chatglm_sagemaker_simple_loRA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "252de0de",
   "metadata": {
    "tags": []
   },
   "source": [
    "### SageMaker fine tune ChatGLM\n",
    "\n",
    "#### 准备\n",
    "1. 升级boto3, sagemaker python sdk  \n",
    "2. 准备requirements.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b8f2c403",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://pypi.org/simple, https://pip.repos.neuron.amazonaws.com\n",
      "Requirement already satisfied: boto3 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (1.26.71)\n",
      "Collecting boto3\n",
      "  Downloading boto3-1.26.116-py3-none-any.whl (135 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m135.6/135.6 kB\u001b[0m \u001b[31m7.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: jmespath<2.0.0,>=0.7.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from boto3) (1.0.1)\n",
      "Requirement already satisfied: s3transfer<0.7.0,>=0.6.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from boto3) (0.6.0)\n",
      "Collecting botocore<1.30.0,>=1.29.116\n",
      "  Downloading botocore-1.29.116-py3-none-any.whl (10.6 MB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10.6/10.6 MB\u001b[0m \u001b[31m111.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: python-dateutil<3.0.0,>=2.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from botocore<1.30.0,>=1.29.116->boto3) (2.8.2)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.25.4 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from botocore<1.30.0,>=1.29.116->boto3) (1.26.8)\n",
      "Requirement already satisfied: six>=1.5 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from python-dateutil<3.0.0,>=2.1->botocore<1.30.0,>=1.29.116->boto3) (1.16.0)\n",
      "Installing collected packages: botocore, boto3\n",
      "  Attempting uninstall: botocore\n",
      "    Found existing installation: botocore 1.29.71\n",
      "    Uninstalling botocore-1.29.71:\n",
      "      Successfully uninstalled botocore-1.29.71\n",
      "  Attempting uninstall: boto3\n",
      "    Found existing installation: boto3 1.26.71\n",
      "    Uninstalling boto3-1.26.71:\n",
      "      Successfully uninstalled boto3-1.26.71\n",
      "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "awscli 1.27.71 requires botocore==1.29.71, but you have botocore 1.29.116 which is incompatible.\u001b[0m\u001b[31m\n",
      "\u001b[0mSuccessfully installed boto3-1.26.116 botocore-1.29.116\n",
      "Looking in indexes: https://pypi.org/simple, https://pip.repos.neuron.amazonaws.com\n",
      "Requirement already satisfied: sagemaker in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (2.132.0)\n",
      "Collecting sagemaker\n",
      "  Downloading sagemaker-2.148.0.tar.gz (743 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m743.3/743.3 kB\u001b[0m \u001b[31m14.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25hRequirement already satisfied: attrs<23,>=20.3.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (22.2.0)\n",
      "Requirement already satisfied: boto3<2.0,>=1.26.28 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (1.26.116)\n",
      "Collecting cloudpickle==2.2.1\n",
      "  Downloading cloudpickle-2.2.1-py3-none-any.whl (25 kB)\n",
      "Requirement already satisfied: google-pasta in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (0.2.0)\n",
      "Requirement already satisfied: numpy<2.0,>=1.9.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (1.23.5)\n",
      "Requirement already satisfied: protobuf<4.0,>=3.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (3.20.2)\n",
      "Requirement already satisfied: protobuf3-to-dict<1.0,>=0.1.5 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (0.1.5)\n",
      "Requirement already satisfied: smdebug_rulesconfig==1.0.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (1.0.1)\n",
      "Requirement already satisfied: importlib-metadata<5.0,>=1.4.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (4.13.0)\n",
      "Requirement already satisfied: packaging>=20.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (21.3)\n",
      "Requirement already satisfied: pandas in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (1.4.4)\n",
      "Requirement already satisfied: pathos in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (0.3.0)\n",
      "Requirement already satisfied: schema in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (0.7.5)\n",
      "Requirement already satisfied: PyYAML==5.4.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (5.4.1)\n",
      "Requirement already satisfied: jsonschema in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (3.2.0)\n",
      "Requirement already satisfied: platformdirs in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (2.6.2)\n",
      "Requirement already satisfied: tblib==1.7.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from sagemaker) (1.7.0)\n",
      "Requirement already satisfied: botocore<1.30.0,>=1.29.116 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from boto3<2.0,>=1.26.28->sagemaker) (1.29.116)\n",
      "Requirement already satisfied: s3transfer<0.7.0,>=0.6.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from boto3<2.0,>=1.26.28->sagemaker) (0.6.0)\n",
      "Requirement already satisfied: jmespath<2.0.0,>=0.7.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from boto3<2.0,>=1.26.28->sagemaker) (1.0.1)\n",
      "Requirement already satisfied: zipp>=0.5 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from importlib-metadata<5.0,>=1.4.0->sagemaker) (3.11.0)\n",
      "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from packaging>=20.0->sagemaker) (3.0.9)\n",
      "Requirement already satisfied: six in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from protobuf3-to-dict<1.0,>=0.1.5->sagemaker) (1.16.0)\n",
      "Requirement already satisfied: pyrsistent>=0.14.0 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from jsonschema->sagemaker) (0.19.3)\n",
      "Requirement already satisfied: setuptools in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from jsonschema->sagemaker) (65.6.3)\n",
      "Requirement already satisfied: python-dateutil>=2.8.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from pandas->sagemaker) (2.8.2)\n",
      "Requirement already satisfied: pytz>=2020.1 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from pandas->sagemaker) (2022.7)\n",
      "Requirement already satisfied: ppft>=1.7.6.6 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from pathos->sagemaker) (1.7.6.6)\n",
      "Requirement already satisfied: dill>=0.3.6 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from pathos->sagemaker) (0.3.6)\n",
      "Requirement already satisfied: multiprocess>=0.70.14 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from pathos->sagemaker) (0.70.14)\n",
      "Requirement already satisfied: pox>=0.3.2 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from pathos->sagemaker) (0.3.2)\n",
      "Requirement already satisfied: contextlib2>=0.5.5 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from schema->sagemaker) (21.6.0)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.25.4 in /home/ec2-user/anaconda3/envs/pytorch_p39/lib/python3.9/site-packages (from botocore<1.30.0,>=1.29.116->boto3<2.0,>=1.26.28->sagemaker) (1.26.8)\n",
      "Building wheels for collected packages: sagemaker\n",
      "  Building wheel for sagemaker (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25h  Created wheel for sagemaker: filename=sagemaker-2.148.0-py2.py3-none-any.whl size=998496 sha256=60972853c9019b0256fc91cbb9a01495e4e968902662ea7a9093c45a60210476\n",
      "  Stored in directory: /home/ec2-user/.cache/pip/wheels/66/36/07/56de705f4ad8ea09e40419f57f8478bc60aae5b1e095dda1f0\n",
      "Successfully built sagemaker\n",
      "Installing collected packages: cloudpickle, sagemaker\n",
      "  Attempting uninstall: cloudpickle\n",
      "    Found existing installation: cloudpickle 2.2.0\n",
      "    Uninstalling cloudpickle-2.2.0:\n",
      "      Successfully uninstalled cloudpickle-2.2.0\n",
      "  Attempting uninstall: sagemaker\n",
      "    Found existing installation: sagemaker 2.132.0\n",
      "    Uninstalling sagemaker-2.132.0:\n",
      "      Successfully uninstalled sagemaker-2.132.0\n",
      "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "distributed 2022.11.0 requires tornado<6.2,>=6.0.3, but you have tornado 6.2 which is incompatible.\u001b[0m\u001b[31m\n",
      "\u001b[0mSuccessfully installed cloudpickle-2.2.1 sagemaker-2.148.0\n"
     ]
    }
   ],
   "source": [
    "!pip install --upgrade boto3\n",
    "!pip install --upgrade sagemaker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a4a30f3a",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "arn:aws:iam::687912291502:role/service-role/AmazonSageMaker-ExecutionRole-20211013T113123\n",
      "sagemaker-us-west-2-687912291502\n"
     ]
    }
   ],
   "source": [
    "import boto3\n",
    "import sagemaker\n",
    "\n",
    "account = boto3.client('sts').get_caller_identity().get('Account')\n",
    "region = boto3.session.Session().region_name\n",
    "\n",
    "sagemaker_session = sagemaker.Session()\n",
    "bucket = sagemaker_session.default_bucket()\n",
    "role = sagemaker.get_execution_role()\n",
    "\n",
    "print(role)\n",
    "print(bucket)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6f59e3f",
   "metadata": {
    "tags": []
   },
   "source": [
    "### chatglm lora方式（单机单卡）\n",
    "1: 使用羊驼语料数据  \n",
    "2：语料处理，tokenization及label标注  \n",
    "3：HF trainer API  \n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "9db2f288-0610-4b18-9409-c8ef5ee91ca0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cloning into 'ChatGLM-Tuning'...\n",
      "remote: Enumerating objects: 158, done.\u001b[K\n",
      "remote: Counting objects: 100% (106/106), done.\u001b[K\n",
      "remote: Compressing objects: 100% (51/51), done.\u001b[K\n",
      "remote: Total 158 (delta 71), reused 55 (delta 55), pack-reused 52\u001b[K\n",
      "Receiving objects: 100% (158/158), 9.33 MiB | 17.46 MiB/s, done.\n",
      "Resolving deltas: 100% (83/83), done.\n"
     ]
    }
   ],
   "source": [
    "!rm -rf ./ChatGLM-Tuning\n",
    "!git clone https://github.com/qingyuan18/ChatGLM-Tuning.git\n",
    "!cp ./s5cmd ./ChatGLM-Tuning/"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b42e2794-2ed4-49f9-b1b0-054f1a0af54f",
   "metadata": {},
   "source": [
    "## prepare docker images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8aceb270-324f-445e-ae84-0da1d1b98adf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overwriting Dockerfile\n"
     ]
    }
   ],
   "source": [
    "%%writefile Dockerfile\n",
    "## You should change below region code to the region you used, here sample is use us-west-2\n",
    "From 763104351884.dkr.ecr.us-west-2.amazonaws.com/huggingface-pytorch-training:1.13.1-transformers4.26.0-gpu-py39-cu117-ubuntu20.04 \n",
    "#From pytorch/pytorch:1.5-cuda10.1-cudnn7-runtime\n",
    "\n",
    "ENV LANG=C.UTF-8\n",
    "ENV PYTHONUNBUFFERED=TRUE\n",
    "ENV PYTHONDONTWRITEBYTECODE=TRUE\n",
    "\n",
    "RUN mkdir -p /opt/\n",
    "COPY ./ChatGLM-Tuning/ /opt/ChatGLM-Tuning/\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3b8ee553",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json.\n",
      "Configure a credential helper to remove this warning. See\n",
      "https://docs.docker.com/engine/reference/commandline/login/#credentials-store\n",
      "\n",
      "Login Succeeded\n"
     ]
    }
   ],
   "source": [
    "## You should change below region code to the region you used, here sample is use us-west-2\n",
    "!aws ecr get-login-password --region us-west-2 | docker login --username AWS --password-stdin 763104351884.dkr.ecr.us-west-2.amazonaws.com"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c00b23c6-6e88-48f2-96dd-99140c147be5",
   "metadata": {},
   "source": [
    "**Build image and push to ECR.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fa970133-14f1-40d4-963f-895154a43f94",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "## define repo name, should contain *sagemaker* in the name\n",
    "repo_name = \"sagemaker-chatglm-lora-demo\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "51c2d43e-ff0f-4718-b772-555c95d6aaaf",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Login Succeeded\n",
      "Sending build context to Docker daemon  13.57GB\n",
      "Step 1/6 : From 763104351884.dkr.ecr.us-west-2.amazonaws.com/huggingface-pytorch-training:1.13.1-transformers4.26.0-gpu-py39-cu117-ubuntu20.04\n",
      " ---> c5a6ef695006\n",
      "Step 2/6 : ENV LANG=C.UTF-8\n",
      " ---> Using cache\n",
      " ---> af49cfa7feae\n",
      "Step 3/6 : ENV PYTHONUNBUFFERED=TRUE\n",
      " ---> Using cache\n",
      " ---> 287106637dc6\n",
      "Step 4/6 : ENV PYTHONDONTWRITEBYTECODE=TRUE\n",
      " ---> Using cache\n",
      " ---> 773b4cf30c90\n",
      "Step 5/6 : RUN mkdir -p /opt/\n",
      " ---> Using cache\n",
      " ---> 953bc66ff020\n",
      "Step 6/6 : COPY ./ChatGLM-Tuning/ /opt/ChatGLM-Tuning/\n",
      " ---> 8c07350c0ce6\n",
      "Successfully built 8c07350c0ce6\n",
      "Successfully tagged sagemaker-chatglm-lora-demo:latest\n",
      "The push refers to repository [687912291502.dkr.ecr.us-west-2.amazonaws.com/sagemaker-chatglm-lora-demo]\n",
      "3f43c43c1d2b: Preparing\n",
      "f8dae5c3df1e: Preparing\n",
      "e3221f18601a: Preparing\n",
      "b6f286626882: Preparing\n",
      "76fe97d80cdb: Preparing\n",
      "f5f76489fff8: Preparing\n",
      "621c3f07daa7: Preparing\n",
      "9b484bb42e11: Preparing\n",
      "54c7c0b58471: Preparing\n",
      "c34adc3ab668: Preparing\n",
      "bbf651e48b84: Preparing\n",
      "f61045791108: Preparing\n",
      "4e2ac0cda74a: Preparing\n",
      "658a33d555eb: Preparing\n",
      "bd16d9a61a98: Preparing\n",
      "f0c0cd2accfa: Preparing\n",
      "1275469c066c: Preparing\n",
      "b802dd3babf4: Preparing\n",
      "a3834ec63558: Preparing\n",
      "63edcef6dedf: Preparing\n",
      "0154e84cc2dd: Preparing\n",
      "7085d1c151f6: Preparing\n",
      "a77a2104cfb6: Preparing\n",
      "6808e7f9da2f: Preparing\n",
      "3bc059a9dec6: Preparing\n",
      "f5f76489fff8: Waiting\n",
      "de783f3fec23: Preparing\n",
      "18ca52d74b2f: Preparing\n",
      "621c3f07daa7: Waiting\n",
      "73df6ccd636c: Preparing\n",
      "9b484bb42e11: Waiting\n",
      "6738b73ff7a8: Preparing\n",
      "2a8292d9bfcc: Preparing\n",
      "54c7c0b58471: Waiting\n",
      "5b75a5ef32a7: Preparing\n",
      "c34adc3ab668: Waiting\n",
      "25a5f55a11f0: Preparing\n",
      "707f484816ae: Preparing\n",
      "0430aa1e47d4: Preparing\n",
      "bbf651e48b84: Waiting\n",
      "65448e793131: Preparing\n",
      "15af6e2d42ba: Preparing\n",
      "b46caef92993: Preparing\n",
      "f61045791108: Waiting\n",
      "53ce33a12646: Preparing\n",
      "aad68760f4ce: Preparing\n",
      "4e2ac0cda74a: Waiting\n",
      "323d67ab1719: Preparing\n",
      "e72743a0fdfe: Preparing\n",
      "3996353f5820: Preparing\n",
      "ea87e0b9c30f: Preparing\n",
      "af18356cdf10: Preparing\n",
      "f6e30dd4497e: Preparing\n",
      "658a33d555eb: Waiting\n",
      "99832d04a153: Preparing\n",
      "a5981ed7a378: Preparing\n",
      "250519a2f830: Preparing\n",
      "6cadbde53f94: Preparing\n",
      "0002c93bdb37: Preparing\n",
      "bd16d9a61a98: Waiting\n",
      "73df6ccd636c: Waiting\n",
      "6738b73ff7a8: Waiting\n",
      "f0c0cd2accfa: Waiting\n",
      "2a8292d9bfcc: Waiting\n",
      "5b75a5ef32a7: Waiting\n",
      "1275469c066c: Waiting\n",
      "25a5f55a11f0: Waiting\n",
      "707f484816ae: Waiting\n",
      "0430aa1e47d4: Waiting\n",
      "65448e793131: Waiting\n",
      "323d67ab1719: Waiting\n",
      "e72743a0fdfe: Waiting\n",
      "15af6e2d42ba: Waiting\n",
      "b46caef92993: Waiting\n",
      "af18356cdf10: Waiting\n",
      "3996353f5820: Waiting\n",
      "99832d04a153: Waiting\n",
      "f6e30dd4497e: Waiting\n",
      "aad68760f4ce: Waiting\n",
      "b802dd3babf4: Waiting\n",
      "0154e84cc2dd: Waiting\n",
      "a3834ec63558: Waiting\n",
      "53ce33a12646: Waiting\n",
      "7085d1c151f6: Waiting\n",
      "a77a2104cfb6: Waiting\n",
      "6808e7f9da2f: Waiting\n",
      "3bc059a9dec6: Waiting\n",
      "0002c93bdb37: Waiting\n",
      "250519a2f830: Waiting\n",
      "6cadbde53f94: Waiting\n",
      "63edcef6dedf: Waiting\n",
      "de783f3fec23: Waiting\n",
      "18ca52d74b2f: Waiting\n",
      "f8dae5c3df1e: Layer already exists\n",
      "76fe97d80cdb: Layer already exists\n",
      "b6f286626882: Layer already exists\n",
      "e3221f18601a: Layer already exists\n",
      "f5f76489fff8: Layer already exists\n",
      "621c3f07daa7: Layer already exists\n",
      "9b484bb42e11: Layer already exists\n",
      "54c7c0b58471: Layer already exists\n",
      "c34adc3ab668: Layer already exists\n",
      "bbf651e48b84: Layer already exists\n",
      "f61045791108: Layer already exists\n",
      "4e2ac0cda74a: Layer already exists\n",
      "658a33d555eb: Layer already exists\n",
      "bd16d9a61a98: Layer already exists\n",
      "f0c0cd2accfa: Layer already exists\n",
      "1275469c066c: Layer already exists\n",
      "b802dd3babf4: Layer already exists\n",
      "a3834ec63558: Layer already exists\n",
      "63edcef6dedf: Layer already exists\n",
      "0154e84cc2dd: Layer already exists\n",
      "7085d1c151f6: Layer already exists\n",
      "a77a2104cfb6: Layer already exists\n",
      "6808e7f9da2f: Layer already exists\n",
      "3bc059a9dec6: Layer already exists\n",
      "de783f3fec23: Layer already exists\n",
      "18ca52d74b2f: Layer already exists\n",
      "73df6ccd636c: Layer already exists\n",
      "6738b73ff7a8: Layer already exists\n",
      "2a8292d9bfcc: Layer already exists\n",
      "5b75a5ef32a7: Layer already exists\n",
      "25a5f55a11f0: Layer already exists\n",
      "707f484816ae: Layer already exists\n",
      "65448e793131: Layer already exists\n",
      "0430aa1e47d4: Layer already exists\n",
      "15af6e2d42ba: Layer already exists\n",
      "b46caef92993: Layer already exists\n",
      "53ce33a12646: Layer already exists\n",
      "aad68760f4ce: Layer already exists\n",
      "323d67ab1719: Layer already exists\n",
      "3996353f5820: Layer already exists\n",
      "e72743a0fdfe: Layer already exists\n",
      "ea87e0b9c30f: Layer already exists\n",
      "af18356cdf10: Layer already exists\n",
      "99832d04a153: Layer already exists\n",
      "f6e30dd4497e: Layer already exists\n",
      "a5981ed7a378: Layer already exists\n",
      "250519a2f830: Layer already exists\n",
      "6cadbde53f94: Layer already exists\n",
      "0002c93bdb37: Layer already exists\n",
      "3f43c43c1d2b: Pushed\n",
      "latest: digest: sha256:09dade21757a43ede9957368afdf968d01298ca46d0ae90f02b594e45d18ce12 size: 10833\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json.\n",
      "Configure a credential helper to remove this warning. See\n",
      "https://docs.docker.com/engine/reference/commandline/login/#credentials-store\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%script env repo_name=$repo_name bash\n",
    "\n",
    "#!/usr/bin/env bash\n",
    "\n",
    "# This script shows how to build the Docker image and push it to ECR to be ready for use\n",
    "# by SageMaker.\n",
    "\n",
    "# The argument to this script is the image name. This will be used as the image on the local\n",
    "# machine and combined with the account and region to form the repository name for ECR.\n",
    "# The name of our algorithm\n",
    "algorithm_name=${repo_name}\n",
    "\n",
    "account=$(aws sts get-caller-identity --query Account --output text)\n",
    "\n",
    "# Get the region defined in the current configuration (default to us-west-2 if none defined)\n",
    "region=$(aws configure get region)\n",
    "region=${region:-us-west-2}\n",
    "\n",
    "fullname=\"${account}.dkr.ecr.${region}.amazonaws.com/${algorithm_name}:latest\"\n",
    "\n",
    "# If the repository doesn't exist in ECR, create it.\n",
    "aws ecr describe-repositories --repository-names \"${algorithm_name}\" > /dev/null 2>&1\n",
    "\n",
    "if [ $? -ne 0 ]\n",
    "then\n",
    "    aws ecr create-repository --repository-name \"${algorithm_name}\" > /dev/null\n",
    "fi\n",
    "\n",
    "# Get the login command from ECR and execute it directly\n",
    "aws ecr get-login-password --region ${region}|docker login --username AWS --password-stdin ${fullname}\n",
    "\n",
    "# Build the docker image locally with the image name and then push it to ECR\n",
    "# with the full name.\n",
    "\n",
    "docker build -t ${algorithm_name} .\n",
    "docker tag ${algorithm_name} ${fullname}\n",
    "\n",
    "docker push ${fullname}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5dcb4cf6-4413-42b8-9ec1-9dcb9dcde793",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Train!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "55cedcb2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overwriting ./ChatGLM-Tuning/train.sh\n"
     ]
    }
   ],
   "source": [
    "%%writefile ./ChatGLM-Tuning/train.sh\n",
    "#!/bin/bash\n",
    "\n",
    "chmod +x ./s5cmd\n",
    "pip install -r ./requirements.txt\n",
    "\n",
    "###转化alpaca数据集为jsonl\n",
    "python ./cover_alpaca2jsonl.py \\\n",
    "    --data_path ./data/alpaca_data.json \\\n",
    "    --save_path ./data/alpaca_data.jsonl \n",
    "\n",
    "#tokenization\n",
    "python ./tokenize_dataset_rows.py \\\n",
    "    --jsonl_path ./data/alpaca_data.jsonl \\\n",
    "    --save_path ./data/alpaca \\\n",
    "    --max_seq_length 200\n",
    "\n",
    "python ./finetune.py \\\n",
    "    --dataset_path ./data/alpaca \\\n",
    "    --lora_rank 8 \\\n",
    "    --per_device_train_batch_size 6 \\\n",
    "    --gradient_accumulation_steps 1 \\\n",
    "    --max_steps 1000 \\\n",
    "    --save_steps 1000 \\\n",
    "    --save_total_limit 2 \\\n",
    "    --learning_rate 1e-4 \\\n",
    "    --fp16 \\\n",
    "    --remove_unused_columns false \\\n",
    "    --logging_steps 50 \\\n",
    "    --output_dir /tmp/output\n",
    "\n",
    "./s5cmd sync /tmp/output/ s3://$MODEL_S3_BUCKET/models/chatglm-lora/output/$(date +%Y-%m-%d-%H-%M-%S)/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "91dc3c13-ba5c-41b9-9682-11dd61cc8382",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'687912291502.dkr.ecr.us-west-2.amazonaws.com/sagemaker-chatglm-lora-demo:latest'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## The image uri which is build and pushed above\n",
    "image_uri = \"{}.dkr.ecr.{}.amazonaws.com/{}:latest\".format(account, region, repo_name)\n",
    "image_uri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8ab36100",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using provided s3_resource\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:sagemaker:Creating training-job with name: chatglm-lora-demo-2023-06-20-09-24-44-574\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2023-06-20 09:24:50 Starting - Starting the training job...\n",
      "2023-06-20 09:25:15 Starting - Preparing the instances for training......\n",
      "2023-06-20 09:26:17 Downloading - Downloading input data...\n",
      "2023-06-20 09:26:37 Training - Downloading the training image.....................\n",
      "2023-06-20 09:30:13 Training - Training image download completed. Training in progress.....\u001b[34mbash: cannot set terminal process group (-1): Inappropriate ioctl for device\u001b[0m\n",
      "\u001b[34mbash: no job control in this shell\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:52,313 sagemaker-training-toolkit INFO     Imported framework sagemaker_pytorch_container.training\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:52,347 sagemaker-training-toolkit INFO     No Neurons detected (normal if no neurons installed)\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:52,357 sagemaker_pytorch_container.training INFO     Block until all host DNS lookups succeed.\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:52,359 sagemaker_pytorch_container.training INFO     Invoking user training script.\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:53,246 sagemaker-training-toolkit INFO     No Neurons detected (normal if no neurons installed)\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:53,294 sagemaker-training-toolkit INFO     No Neurons detected (normal if no neurons installed)\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:53,339 sagemaker-training-toolkit INFO     No Neurons detected (normal if no neurons installed)\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:53,350 sagemaker-training-toolkit INFO     Invoking user script\u001b[0m\n",
      "\u001b[34mTraining Env:\u001b[0m\n",
      "\u001b[34m{\n",
      "    \"additional_framework_parameters\": {},\n",
      "    \"channel_input_dirs\": {},\n",
      "    \"current_host\": \"algo-1\",\n",
      "    \"current_instance_group\": \"homogeneousCluster\",\n",
      "    \"current_instance_group_hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"current_instance_type\": \"ml.g5.12xlarge\",\n",
      "    \"distribution_hosts\": [],\n",
      "    \"distribution_instance_groups\": [],\n",
      "    \"framework_module\": \"sagemaker_pytorch_container.training:main\",\n",
      "    \"hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"hyperparameters\": {},\n",
      "    \"input_config_dir\": \"/opt/ml/input/config\",\n",
      "    \"input_data_config\": {},\n",
      "    \"input_dir\": \"/opt/ml/input\",\n",
      "    \"instance_groups\": [\n",
      "        \"homogeneousCluster\"\n",
      "    ],\n",
      "    \"instance_groups_dict\": {\n",
      "        \"homogeneousCluster\": {\n",
      "            \"instance_group_name\": \"homogeneousCluster\",\n",
      "            \"instance_type\": \"ml.g5.12xlarge\",\n",
      "            \"hosts\": [\n",
      "                \"algo-1\"\n",
      "            ]\n",
      "        }\n",
      "    },\n",
      "    \"is_hetero\": false,\n",
      "    \"is_master\": true,\n",
      "    \"is_modelparallel_enabled\": null,\n",
      "    \"is_smddpmprun_installed\": true,\n",
      "    \"job_name\": \"chatglm-lora-demo-2023-06-20-09-24-44-574\",\n",
      "    \"log_level\": 20,\n",
      "    \"master_hostname\": \"algo-1\",\n",
      "    \"model_dir\": \"/opt/ml/model\",\n",
      "    \"module_dir\": \"s3://sagemaker-us-west-2-687912291502/chatglm-lora-demo-2023-06-20-09-24-44-574/source/sourcedir.tar.gz\",\n",
      "    \"module_name\": \"train.sh\",\n",
      "    \"network_interface_name\": \"eth0\",\n",
      "    \"num_cpus\": 48,\n",
      "    \"num_gpus\": 4,\n",
      "    \"num_neurons\": 0,\n",
      "    \"output_data_dir\": \"/opt/ml/output/data\",\n",
      "    \"output_dir\": \"/opt/ml/output\",\n",
      "    \"output_intermediate_dir\": \"/opt/ml/output/intermediate\",\n",
      "    \"resource_config\": {\n",
      "        \"current_host\": \"algo-1\",\n",
      "        \"current_instance_type\": \"ml.g5.12xlarge\",\n",
      "        \"current_group_name\": \"homogeneousCluster\",\n",
      "        \"hosts\": [\n",
      "            \"algo-1\"\n",
      "        ],\n",
      "        \"instance_groups\": [\n",
      "            {\n",
      "                \"instance_group_name\": \"homogeneousCluster\",\n",
      "                \"instance_type\": \"ml.g5.12xlarge\",\n",
      "                \"hosts\": [\n",
      "                    \"algo-1\"\n",
      "                ]\n",
      "            }\n",
      "        ],\n",
      "        \"network_interface_name\": \"eth0\"\n",
      "    },\n",
      "    \"user_entry_point\": \"train.sh\"\u001b[0m\n",
      "\u001b[34m}\u001b[0m\n",
      "\u001b[34mEnvironment variables:\u001b[0m\n",
      "\u001b[34mSM_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_NETWORK_INTERFACE_NAME=eth0\u001b[0m\n",
      "\u001b[34mSM_HPS={}\u001b[0m\n",
      "\u001b[34mSM_USER_ENTRY_POINT=train.sh\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_PARAMS={}\u001b[0m\n",
      "\u001b[34mSM_RESOURCE_CONFIG={\"current_group_name\":\"homogeneousCluster\",\"current_host\":\"algo-1\",\"current_instance_type\":\"ml.g5.12xlarge\",\"hosts\":[\"algo-1\"],\"instance_groups\":[{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.g5.12xlarge\"}],\"network_interface_name\":\"eth0\"}\u001b[0m\n",
      "\u001b[34mSM_INPUT_DATA_CONFIG={}\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DATA_DIR=/opt/ml/output/data\u001b[0m\n",
      "\u001b[34mSM_CHANNELS=[]\u001b[0m\n",
      "\u001b[34mSM_CURRENT_HOST=algo-1\u001b[0m\n",
      "\u001b[34mSM_CURRENT_INSTANCE_TYPE=ml.g5.12xlarge\u001b[0m\n",
      "\u001b[34mSM_CURRENT_INSTANCE_GROUP=homogeneousCluster\u001b[0m\n",
      "\u001b[34mSM_CURRENT_INSTANCE_GROUP_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_INSTANCE_GROUPS=[\"homogeneousCluster\"]\u001b[0m\n",
      "\u001b[34mSM_INSTANCE_GROUPS_DICT={\"homogeneousCluster\":{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.g5.12xlarge\"}}\u001b[0m\n",
      "\u001b[34mSM_DISTRIBUTION_INSTANCE_GROUPS=[]\u001b[0m\n",
      "\u001b[34mSM_IS_HETERO=false\u001b[0m\n",
      "\u001b[34mSM_MODULE_NAME=train.sh\u001b[0m\n",
      "\u001b[34mSM_LOG_LEVEL=20\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_MODULE=sagemaker_pytorch_container.training:main\u001b[0m\n",
      "\u001b[34mSM_INPUT_DIR=/opt/ml/input\u001b[0m\n",
      "\u001b[34mSM_INPUT_CONFIG_DIR=/opt/ml/input/config\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DIR=/opt/ml/output\u001b[0m\n",
      "\u001b[34mSM_NUM_CPUS=48\u001b[0m\n",
      "\u001b[34mSM_NUM_GPUS=4\u001b[0m\n",
      "\u001b[34mSM_NUM_NEURONS=0\u001b[0m\n",
      "\u001b[34mSM_MODEL_DIR=/opt/ml/model\u001b[0m\n",
      "\u001b[34mSM_MODULE_DIR=s3://sagemaker-us-west-2-687912291502/chatglm-lora-demo-2023-06-20-09-24-44-574/source/sourcedir.tar.gz\u001b[0m\n",
      "\u001b[34mSM_TRAINING_ENV={\"additional_framework_parameters\":{},\"channel_input_dirs\":{},\"current_host\":\"algo-1\",\"current_instance_group\":\"homogeneousCluster\",\"current_instance_group_hosts\":[\"algo-1\"],\"current_instance_type\":\"ml.g5.12xlarge\",\"distribution_hosts\":[],\"distribution_instance_groups\":[],\"framework_module\":\"sagemaker_pytorch_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{},\"input_dir\":\"/opt/ml/input\",\"instance_groups\":[\"homogeneousCluster\"],\"instance_groups_dict\":{\"homogeneousCluster\":{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.g5.12xlarge\"}},\"is_hetero\":false,\"is_master\":true,\"is_modelparallel_enabled\":null,\"is_smddpmprun_installed\":true,\"job_name\":\"chatglm-lora-demo-2023-06-20-09-24-44-574\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://sagemaker-us-west-2-687912291502/chatglm-lora-demo-2023-06-20-09-24-44-574/source/sourcedir.tar.gz\",\"module_name\":\"train.sh\",\"network_interface_name\":\"eth0\",\"num_cpus\":48,\"num_gpus\":4,\"num_neurons\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_group_name\":\"homogeneousCluster\",\"current_host\":\"algo-1\",\"current_instance_type\":\"ml.g5.12xlarge\",\"hosts\":[\"algo-1\"],\"instance_groups\":[{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.g5.12xlarge\"}],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"train.sh\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ARGS=[]\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_INTERMEDIATE_DIR=/opt/ml/output/intermediate\u001b[0m\n",
      "\u001b[34mPYTHONPATH=/opt/ml/code:/opt/conda/bin:/opt/conda/lib/python39.zip:/opt/conda/lib/python3.9:/opt/conda/lib/python3.9/lib-dynload:/opt/conda/lib/python3.9/site-packages\u001b[0m\n",
      "\u001b[34mInvoking script with the following command:\u001b[0m\n",
      "\u001b[34m/bin/sh -c \"./train.sh \"\u001b[0m\n",
      "\u001b[34m[2023-06-20 09:30:54.931: W smdistributed/modelparallel/torch/nn/predefined_hooks.py:78] Found unsupported HuggingFace version 4.26.0 for automated tensor parallelism. HuggingFace modules will not be automatically distributed. You can use smp.tp_register_with_module API to register desired modules for tensor parallelism, or directly instantiate an smp.nn.DistributedModule. Supported HuggingFace transformers versions for automated tensor parallelism: ['4.17.0', '4.20.1', '4.21.0']\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:54,935 root         INFO     Using NamedTuple = typing._NamedTuple instead.\u001b[0m\n",
      "\u001b[34m2023-06-20 09:30:54,955 sagemaker-training-toolkit INFO     Exceptions not imported for SageMaker TF as Tensorflow is not installed.\u001b[0m\n",
      "\u001b[34mCollecting git+https://github.com/huggingface/peft.git (from -r ./requirements.txt (line 15))\u001b[0m\n",
      "\u001b[34mCloning https://github.com/huggingface/peft.git to /tmp/pip-req-build-53y2zuxi\u001b[0m\n",
      "\u001b[34mRunning command git clone --filter=blob:none --quiet https://github.com/huggingface/peft.git /tmp/pip-req-build-53y2zuxi\u001b[0m\n",
      "\u001b[34mResolved https://github.com/huggingface/peft.git to commit 08cb3dde577747f6ca6638c884fd66fd16cf2e9d\u001b[0m\n",
      "\u001b[34mInstalling build dependencies: started\u001b[0m\n",
      "\u001b[34mInstalling build dependencies: finished with status 'done'\u001b[0m\n",
      "\u001b[34mGetting requirements to build wheel: started\u001b[0m\n",
      "\u001b[34mGetting requirements to build wheel: finished with status 'done'\u001b[0m\n",
      "\u001b[34mPreparing metadata (pyproject.toml): started\u001b[0m\n",
      "\u001b[34mPreparing metadata (pyproject.toml): finished with status 'done'\u001b[0m\n",
      "\u001b[34mCollecting bitsandbytes==0.37.1\u001b[0m\n",
      "\u001b[34mDownloading bitsandbytes-0.37.1-py3-none-any.whl (76.3 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 76.3/76.3 MB 29.2 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting accelerate==0.17.1\u001b[0m\n",
      "\u001b[34mDownloading accelerate-0.17.1-py3-none-any.whl (212 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 212.8/212.8 kB 47.3 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting protobuf<3.20.1,>=3.19.5\u001b[0m\n",
      "\u001b[34mDownloading protobuf-3.20.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.whl (1.0 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.0/1.0 MB 90.9 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting transformers==4.27.1\u001b[0m\n",
      "\u001b[34mDownloading transformers-4.27.1-py3-none-any.whl (6.7 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.7/6.7 MB 117.1 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting icetk\u001b[0m\n",
      "\u001b[34mDownloading icetk-0.0.7-py3-none-any.whl (16 kB)\u001b[0m\n",
      "\u001b[34mCollecting cpm_kernels==1.0.11\u001b[0m\n",
      "\u001b[34mDownloading cpm_kernels-1.0.11-py3-none-any.whl (416 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 416.6/416.6 kB 73.0 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: torch>=1.13.1 in /opt/conda/lib/python3.9/site-packages (from -r ./requirements.txt (line 10)) (1.13.1+cu117)\u001b[0m\n",
      "\u001b[34mCollecting tensorboard\u001b[0m\n",
      "\u001b[34mDownloading tensorboard-2.13.0-py3-none-any.whl (5.6 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.6/5.6 MB 113.0 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting datasets==2.10.1\u001b[0m\n",
      "\u001b[34mDownloading datasets-2.10.1-py3-none-any.whl (469 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 469.0/469.0 kB 71.5 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: pyyaml in /opt/conda/lib/python3.9/site-packages (from accelerate==0.17.1->-r ./requirements.txt (line 3)) (5.4.1)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: packaging>=20.0 in /opt/conda/lib/python3.9/site-packages (from accelerate==0.17.1->-r ./requirements.txt (line 3)) (23.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: psutil in /opt/conda/lib/python3.9/site-packages (from accelerate==0.17.1->-r ./requirements.txt (line 3)) (5.9.4)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: numpy>=1.17 in /opt/conda/lib/python3.9/site-packages (from accelerate==0.17.1->-r ./requirements.txt (line 3)) (1.23.5)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: tokenizers!=0.11.3,<0.14,>=0.11.1 in /opt/conda/lib/python3.9/site-packages (from transformers==4.27.1->-r ./requirements.txt (line 7)) (0.13.2)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: requests in /opt/conda/lib/python3.9/site-packages (from transformers==4.27.1->-r ./requirements.txt (line 7)) (2.28.2)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: regex!=2019.12.17 in /opt/conda/lib/python3.9/site-packages (from transformers==4.27.1->-r ./requirements.txt (line 7)) (2022.10.31)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: tqdm>=4.27 in /opt/conda/lib/python3.9/site-packages (from transformers==4.27.1->-r ./requirements.txt (line 7)) (4.64.1)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: filelock in /opt/conda/lib/python3.9/site-packages (from transformers==4.27.1->-r ./requirements.txt (line 7)) (3.9.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: huggingface-hub<1.0,>=0.11.0 in /opt/conda/lib/python3.9/site-packages (from transformers==4.27.1->-r ./requirements.txt (line 7)) (0.12.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: aiohttp in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (3.8.4)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: responses<0.19 in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (0.18.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: pandas in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (1.5.3)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: pyarrow>=6.0.0 in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (11.0.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: dill<0.3.7,>=0.3.0 in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (0.3.6)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: xxhash in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (3.2.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: multiprocess in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (0.70.14)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: fsspec[http]>=2021.11.1 in /opt/conda/lib/python3.9/site-packages (from datasets==2.10.1->-r ./requirements.txt (line 14)) (2023.1.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: sentencepiece in /opt/conda/lib/python3.9/site-packages (from icetk->-r ./requirements.txt (line 8)) (0.1.97)\u001b[0m\n",
      "\u001b[34mCollecting icetk\u001b[0m\n",
      "\u001b[34mDownloading icetk-0.0.6-py3-none-any.whl (15 kB)\u001b[0m\n",
      "\u001b[34mDownloading icetk-0.0.5-py3-none-any.whl (15 kB)\u001b[0m\n",
      "\u001b[34mDownloading icetk-0.0.4-py3-none-any.whl (15 kB)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: torchvision in /opt/conda/lib/python3.9/site-packages (from icetk->-r ./requirements.txt (line 8)) (0.14.1+cu117)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: typing-extensions in /opt/conda/lib/python3.9/site-packages (from torch>=1.13.1->-r ./requirements.txt (line 10)) (4.4.0)\u001b[0m\n",
      "\u001b[34mCollecting google-auth<3,>=1.6.3\u001b[0m\n",
      "\u001b[34mDownloading google_auth-2.20.0-py2.py3-none-any.whl (181 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 181.5/181.5 kB 44.4 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: setuptools>=41.0.0 in /opt/conda/lib/python3.9/site-packages (from tensorboard->-r ./requirements.txt (line 11)) (65.6.3)\u001b[0m\n",
      "\u001b[34mCollecting grpcio>=1.48.2\u001b[0m\n",
      "\u001b[34mDownloading grpcio-1.54.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (5.1 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.1/5.1 MB 124.8 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: werkzeug>=1.0.1 in /opt/conda/lib/python3.9/site-packages (from tensorboard->-r ./requirements.txt (line 11)) (2.2.3)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: wheel>=0.26 in /opt/conda/lib/python3.9/site-packages (from tensorboard->-r ./requirements.txt (line 11)) (0.38.4)\u001b[0m\n",
      "\u001b[34mCollecting tensorboard-data-server<0.8.0,>=0.7.0\u001b[0m\n",
      "\u001b[34mDownloading tensorboard_data_server-0.7.1-py3-none-manylinux2014_x86_64.whl (6.6 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.6/6.6 MB 123.5 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting google-auth-oauthlib<1.1,>=0.5\u001b[0m\n",
      "\u001b[34mDownloading google_auth_oauthlib-1.0.0-py2.py3-none-any.whl (18 kB)\u001b[0m\n",
      "\u001b[34mCollecting absl-py>=0.4\u001b[0m\n",
      "\u001b[34mDownloading absl_py-1.4.0-py3-none-any.whl (126 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 126.5/126.5 kB 39.3 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: markdown>=2.6.8 in /opt/conda/lib/python3.9/site-packages (from tensorboard->-r ./requirements.txt (line 11)) (3.4.1)\u001b[0m\n",
      "\u001b[34mCollecting safetensors\u001b[0m\n",
      "\u001b[34mDownloading safetensors-0.3.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.3/1.3 MB 94.0 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: charset-normalizer<4.0,>=2.0 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (2.1.1)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: frozenlist>=1.1.1 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (1.3.3)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (1.8.2)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (4.0.2)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: aiosignal>=1.1.2 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (1.3.1)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: attrs>=17.3.0 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (22.2.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/lib/python3.9/site-packages (from aiohttp->datasets==2.10.1->-r ./requirements.txt (line 14)) (6.0.4)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: rsa<5,>=3.1.4 in /opt/conda/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard->-r ./requirements.txt (line 11)) (4.7.2)\u001b[0m\n",
      "\u001b[34mCollecting pyasn1-modules>=0.2.1\u001b[0m\n",
      "\u001b[34mDownloading pyasn1_modules-0.3.0-py2.py3-none-any.whl (181 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 181.3/181.3 kB 47.8 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mCollecting cachetools<6.0,>=2.0.0\u001b[0m\n",
      "\u001b[34mDownloading cachetools-5.3.1-py3-none-any.whl (9.3 kB)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: urllib3<2.0 in /opt/conda/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard->-r ./requirements.txt (line 11)) (1.26.14)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: six>=1.9.0 in /opt/conda/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard->-r ./requirements.txt (line 11)) (1.16.0)\u001b[0m\n",
      "\u001b[34mCollecting requests-oauthlib>=0.7.0\u001b[0m\n",
      "\u001b[34mDownloading requests_oauthlib-1.3.1-py2.py3-none-any.whl (23 kB)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: importlib-metadata>=4.4 in /opt/conda/lib/python3.9/site-packages (from markdown>=2.6.8->tensorboard->-r ./requirements.txt (line 11)) (4.13.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.9/site-packages (from requests->transformers==4.27.1->-r ./requirements.txt (line 7)) (2022.12.7)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.9/site-packages (from requests->transformers==4.27.1->-r ./requirements.txt (line 7)) (3.4)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: MarkupSafe>=2.1.1 in /opt/conda/lib/python3.9/site-packages (from werkzeug>=1.0.1->tensorboard->-r ./requirements.txt (line 11)) (2.1.2)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: python-dateutil>=2.8.1 in /opt/conda/lib/python3.9/site-packages (from pandas->datasets==2.10.1->-r ./requirements.txt (line 14)) (2.8.2)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: pytz>=2020.1 in /opt/conda/lib/python3.9/site-packages (from pandas->datasets==2.10.1->-r ./requirements.txt (line 14)) (2022.7.1)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: pillow!=8.3.*,>=5.3.0 in /opt/conda/lib/python3.9/site-packages (from torchvision->icetk->-r ./requirements.txt (line 8)) (9.4.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: zipp>=0.5 in /opt/conda/lib/python3.9/site-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard->-r ./requirements.txt (line 11)) (3.13.0)\u001b[0m\n",
      "\u001b[34mRequirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /opt/conda/lib/python3.9/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard->-r ./requirements.txt (line 11)) (0.4.8)\u001b[0m\n",
      "\u001b[34mCollecting oauthlib>=3.0.0\u001b[0m\n",
      "\u001b[34mDownloading oauthlib-3.2.2-py3-none-any.whl (151 kB)\u001b[0m\n",
      "\u001b[34m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 151.7/151.7 kB 42.0 MB/s eta 0:00:00\u001b[0m\n",
      "\u001b[34mBuilding wheels for collected packages: peft\u001b[0m\n",
      "\u001b[34mBuilding wheel for peft (pyproject.toml): started\u001b[0m\n",
      "\u001b[34mBuilding wheel for peft (pyproject.toml): finished with status 'done'\u001b[0m\n",
      "\u001b[34mCreated wheel for peft: filename=peft-0.4.0.dev0-py3-none-any.whl size=61644 sha256=3b09b7a0ccb7bc310b8d64a84f4ce4acf734995d9998224ac675d9b5aadb8847\u001b[0m\n",
      "\u001b[34mStored in directory: /tmp/pip-ephem-wheel-cache-lussg_td/wheels/2d/60/1b/0edd9dc0f0c489738b1166bc1b0b560ee368f7721f89d06e3a\u001b[0m\n",
      "\u001b[34mSuccessfully built peft\u001b[0m\n",
      "\u001b[34mInstalling collected packages: safetensors, cpm_kernels, bitsandbytes, tensorboard-data-server, pyasn1-modules, protobuf, oauthlib, grpcio, cachetools, absl-py, requests-oauthlib, google-auth, accelerate, transformers, icetk, google-auth-oauthlib, tensorboard, peft, datasets\u001b[0m\n",
      "\u001b[34mAttempting uninstall: protobuf\u001b[0m\n",
      "\u001b[34mFound existing installation: protobuf 3.20.2\u001b[0m\n",
      "\u001b[34mUninstalling protobuf-3.20.2:\u001b[0m\n",
      "\u001b[34mSuccessfully uninstalled protobuf-3.20.2\u001b[0m\n",
      "\u001b[34mAttempting uninstall: accelerate\u001b[0m\n",
      "\u001b[34mFound existing installation: accelerate 0.16.0\u001b[0m\n",
      "\u001b[34mUninstalling accelerate-0.16.0:\u001b[0m\n",
      "\u001b[34mSuccessfully uninstalled accelerate-0.16.0\u001b[0m\n",
      "\u001b[34mAttempting uninstall: transformers\u001b[0m\n",
      "\u001b[34mFound existing installation: transformers 4.26.0\u001b[0m\n",
      "\u001b[34mUninstalling transformers-4.26.0:\u001b[0m\n",
      "\u001b[34mSuccessfully uninstalled transformers-4.26.0\u001b[0m\n",
      "\u001b[34mAttempting uninstall: datasets\u001b[0m\n",
      "\u001b[34mFound existing installation: datasets 2.9.0\u001b[0m\n",
      "\u001b[34mUninstalling datasets-2.9.0:\u001b[0m\n",
      "\u001b[34mSuccessfully uninstalled datasets-2.9.0\u001b[0m\n",
      "\u001b[34mSuccessfully installed absl-py-1.4.0 accelerate-0.17.1 bitsandbytes-0.37.1 cachetools-5.3.1 cpm_kernels-1.0.11 datasets-2.10.1 google-auth-2.20.0 google-auth-oauthlib-1.0.0 grpcio-1.54.2 icetk-0.0.4 oauthlib-3.2.2 peft-0.4.0.dev0 protobuf-3.20.0 pyasn1-modules-0.3.0 requests-oauthlib-1.3.1 safetensors-0.3.1 tensorboard-2.13.0 tensorboard-data-server-0.7.1 transformers-4.27.1\u001b[0m\n",
      "\u001b[34mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\n",
      "\u001b[34m[notice] A new release of pip is available: 23.0 -> 23.1.2\u001b[0m\n",
      "\u001b[34m[notice] To update, run: pip install --upgrade pip\u001b[0m\n",
      "\u001b[34mformatting..:   0%|          | 0/52002 [00:00<?, ?it/s]\u001b[0m\n",
      "\u001b[34mformatting..:  35%|███▌      | 18406/52002 [00:00<00:00, 184050.96it/s]\u001b[0m\n",
      "\u001b[34mformatting..:  71%|███████▏  | 37155/52002 [00:00<00:00, 186067.20it/s]\u001b[0m\n",
      "\u001b[34mformatting..: 100%|██████████| 52002/52002 [00:00<00:00, 185926.31it/s]\u001b[0m\n",
      "\u001b[34mDownloading and preparing dataset generator/default to /root/.cache/huggingface/datasets/generator/default-20105bf96995fb33/0.0.0...\u001b[0m\n",
      "\u001b[34mGenerating train split: 0 examples [00:00, ? examples/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)okenizer_config.json:   0%|          | 0.00/441 [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)okenizer_config.json: 100%|██████████| 441/441 [00:00<00:00, 116kB/s]\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mDownloading (…)enization_chatglm.py:   0%|          | 0.00/17.0k [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)enization_chatglm.py: 100%|██████████| 17.0k/17.0k [00:00<00:00, 4.12MB/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)\"ice_text.model\";:   0%|          | 0.00/2.71M [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)\"ice_text.model\";: 100%|██████████| 2.71M/2.71M [00:00<00:00, 64.0MB/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)lve/main/config.json:   0%|          | 0.00/773 [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)lve/main/config.json: 100%|██████████| 773/773 [00:00<00:00, 253kB/s]\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mDownloading (…)iguration_chatglm.py:   0%|          | 0.00/4.28k [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)iguration_chatglm.py: 100%|██████████| 4.28k/4.28k [00:00<00:00, 3.23MB/s]\u001b[0m\n",
      "\u001b[34m0%|          | 0/52002 [00:00<?, ?it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1 examples [00:03,  3.27s/ examples]\u001b[0m\n",
      "\u001b[34m0%|          | 135/52002 [00:00<00:38, 1346.42it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 137 examples [00:03, 57.19 examples/s]\u001b[0m\n",
      "\u001b[34m1%|          | 282/52002 [00:00<00:36, 1413.69it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 284 examples [00:03, 136.81 examples/s]\u001b[0m\n",
      "\u001b[34m1%|          | 428/52002 [00:00<00:36, 1431.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 431 examples [00:03, 236.02 examples/s]\u001b[0m\n",
      "\u001b[34mGenerating train split: 566 examples [00:03, 342.79 examples/s]\u001b[0m\n",
      "\u001b[34m1%|          | 572/52002 [00:00<00:37, 1389.68it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 706 examples [00:03, 469.32 examples/s]\u001b[0m\n",
      "\u001b[34m1%|▏         | 712/52002 [00:00<00:37, 1381.48it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 852 examples [00:03, 613.57 examples/s]\u001b[0m\n",
      "\u001b[34m2%|▏         | 857/52002 [00:00<00:36, 1404.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 989 examples [00:03, 743.10 examples/s]\u001b[0m\n",
      "\u001b[34m2%|▏         | 998/52002 [00:00<00:36, 1398.00it/s]#033[A\u001b[0m\n",
      "\u001b[34m2%|▏         | 1138/52002 [00:00<00:37, 1358.85it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1193 examples [00:04, 908.99 examples/s]\u001b[0m\n",
      "\u001b[34m2%|▏         | 1286/52002 [00:00<00:36, 1395.01it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1334 examples [00:04, 1008.80 examples/s]\u001b[0m\n",
      "\u001b[34m3%|▎         | 1428/52002 [00:01<00:36, 1401.71it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1485 examples [00:04, 1120.87 examples/s]\u001b[0m\n",
      "\u001b[34m3%|▎         | 1572/52002 [00:01<00:35, 1411.40it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1679 examples [00:04, 1173.82 examples/s]\u001b[0m\n",
      "\u001b[34m3%|▎         | 1714/52002 [00:01<00:37, 1353.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1816 examples [00:04, 1219.70 examples/s]\u001b[0m\n",
      "\u001b[34m4%|▎         | 1853/52002 [00:01<00:36, 1362.01it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 1953 examples [00:04, 1255.33 examples/s]\u001b[0m\n",
      "\u001b[34m4%|▍         | 1997/52002 [00:01<00:36, 1382.01it/s]#033[A\u001b[0m\n",
      "\u001b[34m4%|▍         | 2136/52002 [00:01<00:36, 1353.91it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 2157 examples [00:04, 1287.77 examples/s]\u001b[0m\n",
      "\u001b[34m4%|▍         | 2281/52002 [00:01<00:36, 1380.75it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 2308 examples [00:04, 1341.04 examples/s]\u001b[0m\n",
      "\u001b[34m5%|▍         | 2426/52002 [00:01<00:35, 1397.78it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 2524 examples [00:05, 1371.89 examples/s]\u001b[0m\n",
      "\u001b[34m5%|▍         | 2569/52002 [00:01<00:35, 1405.34it/s]#033[A\u001b[0m\n",
      "\u001b[34m5%|▌         | 2712/52002 [00:01<00:35, 1407.77it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 2740 examples [00:05, 1393.21 examples/s]\u001b[0m\n",
      "\u001b[34m5%|▌         | 2858/52002 [00:02<00:34, 1419.40it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 2948 examples [00:05, 1387.13 examples/s]\u001b[0m\n",
      "\u001b[34m6%|▌         | 3001/52002 [00:02<00:35, 1370.52it/s]#033[A\u001b[0m\n",
      "\u001b[34m6%|▌         | 3143/52002 [00:02<00:35, 1383.29it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 3149 examples [00:05, 1367.74 examples/s]\u001b[0m\n",
      "\u001b[34m6%|▋         | 3282/52002 [00:02<00:35, 1379.38it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 3350 examples [00:05, 1351.12 examples/s]\u001b[0m\n",
      "\u001b[34m7%|▋         | 3421/52002 [00:02<00:35, 1353.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 3560 examples [00:05, 1358.35 examples/s]\u001b[0m\n",
      "\u001b[34m7%|▋         | 3560/52002 [00:02<00:35, 1357.34it/s]#033[A\u001b[0m\n",
      "\u001b[34m7%|▋         | 3696/52002 [00:02<00:36, 1326.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 3759 examples [00:05, 1345.40 examples/s]\u001b[0m\n",
      "\u001b[34m7%|▋         | 3839/52002 [00:02<00:35, 1352.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 3902 examples [00:06, 1363.84 examples/s]\u001b[0m\n",
      "\u001b[34m8%|▊         | 3978/52002 [00:02<00:35, 1360.88it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 4090 examples [00:06, 1323.05 examples/s]\u001b[0m\n",
      "\u001b[34m8%|▊         | 4115/52002 [00:02<00:36, 1307.39it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 4232 examples [00:06, 1343.11 examples/s]\u001b[0m\n",
      "\u001b[34m8%|▊         | 4264/52002 [00:03<00:35, 1359.81it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 4383 examples [00:06, 1380.32 examples/s]\u001b[0m\n",
      "\u001b[34m8%|▊         | 4403/52002 [00:03<00:34, 1368.53it/s]#033[A\u001b[0m\n",
      "\u001b[34m9%|▊         | 4545/52002 [00:03<00:34, 1381.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 4588 examples [00:06, 1371.96 examples/s]\u001b[0m\n",
      "\u001b[34m9%|▉         | 4684/52002 [00:03<00:35, 1345.96it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 4801 examples [00:06, 1382.81 examples/s]\u001b[0m\n",
      "\u001b[34m9%|▉         | 4830/52002 [00:03<00:34, 1378.53it/s]#033[A\u001b[0m\n",
      "\u001b[34m10%|▉         | 4977/52002 [00:03<00:33, 1405.08it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 5000 examples [00:06, 1349.26 examples/s]\u001b[0m\n",
      "\u001b[34m10%|▉         | 5118/52002 [00:03<00:35, 1336.71it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 5206 examples [00:07, 1351.89 examples/s]\u001b[0m\n",
      "\u001b[34m10%|█         | 5255/52002 [00:03<00:34, 1344.67it/s]#033[A\u001b[0m\n",
      "\u001b[34m10%|█         | 5391/52002 [00:03<00:34, 1344.53it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 5410 examples [00:07, 1352.03 examples/s]\u001b[0m\n",
      "\u001b[34m11%|█         | 5543/52002 [00:04<00:33, 1393.89it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 5562 examples [00:07, 1386.05 examples/s]\u001b[0m\n",
      "\u001b[34m11%|█         | 5690/52002 [00:04<00:32, 1415.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 5708 examples [00:07, 1403.47 examples/s]\u001b[0m\n",
      "\u001b[34m11%|█         | 5832/52002 [00:04<00:32, 1401.23it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 5906 examples [00:07, 1371.67 examples/s]\u001b[0m\n",
      "\u001b[34m11%|█▏        | 5973/52002 [00:04<00:34, 1349.44it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 6100 examples [00:07, 1341.61 examples/s]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 6109/52002 [00:04<00:34, 1337.84it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 6236 examples [00:07, 1344.65 examples/s]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 6246/52002 [00:04<00:34, 1345.65it/s]#033[A\u001b[0m\n",
      "\u001b[34m12%|█▏        | 6381/52002 [00:04<00:33, 1342.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 6439 examples [00:07, 1345.81 examples/s]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 6517/52002 [00:04<00:33, 1346.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 6643 examples [00:08, 1345.22 examples/s]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 6652/52002 [00:04<00:33, 1339.32it/s]#033[A\u001b[0m\n",
      "\u001b[34m13%|█▎        | 6787/52002 [00:04<00:33, 1339.32it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 6836 examples [00:08, 1324.53 examples/s]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 6921/52002 [00:05<00:34, 1325.42it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 6976 examples [00:08, 1333.75 examples/s]\u001b[0m\n",
      "\u001b[34m14%|█▎        | 7054/52002 [00:05<00:35, 1275.19it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 7159 examples [00:08, 1294.85 examples/s]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 7183/52002 [00:05<00:35, 1279.16it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 7298 examples [00:08, 1315.27 examples/s]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 7325/52002 [00:05<00:33, 1320.15it/s]#033[A\u001b[0m\n",
      "\u001b[34m14%|█▍        | 7458/52002 [00:05<00:34, 1287.18it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 7488 examples [00:08, 1295.40 examples/s]\u001b[0m\n",
      "\u001b[34m15%|█▍        | 7601/52002 [00:05<00:33, 1326.41it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 7631 examples [00:08, 1325.59 examples/s]\u001b[0m\n",
      "\u001b[34m15%|█▍        | 7747/52002 [00:05<00:32, 1360.60it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 7771 examples [00:08, 1341.00 examples/s]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 7884/52002 [00:05<00:33, 1305.17it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 7967 examples [00:09, 1325.68 examples/s]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 8016/52002 [00:05<00:34, 1291.89it/s]#033[A\u001b[0m\n",
      "\u001b[34m16%|█▌        | 8155/52002 [00:06<00:33, 1319.12it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 8160 examples [00:09, 1308.95 examples/s]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 8288/52002 [00:06<00:33, 1312.28it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 8354 examples [00:09, 1298.65 examples/s]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 8420/52002 [00:06<00:33, 1303.47it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 8498 examples [00:09, 1329.95 examples/s]\u001b[0m\n",
      "\u001b[34m16%|█▋        | 8570/52002 [00:06<00:31, 1360.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 8636 examples [00:09, 1339.55 examples/s]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 8709/52002 [00:06<00:31, 1368.47it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 8775 examples [00:09, 1351.24 examples/s]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 8858/52002 [00:06<00:30, 1402.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 8934 examples [00:09, 1414.34 examples/s]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 9000/52002 [00:06<00:31, 1385.41it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 9127 examples [00:09, 1351.69 examples/s]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 9139/52002 [00:06<00:31, 1343.62it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 9267 examples [00:10, 1361.19 examples/s]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 9282/52002 [00:06<00:31, 1365.56it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 9414 examples [00:10, 1386.92 examples/s]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 9431/52002 [00:06<00:30, 1401.88it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 9555 examples [00:10, 1389.26 examples/s]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 9572/52002 [00:07<00:30, 1375.08it/s]#033[A\u001b[0m\n",
      "\u001b[34m19%|█▊        | 9710/52002 [00:07<00:31, 1346.15it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 9753 examples [00:10, 1354.17 examples/s]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 9845/52002 [00:07<00:32, 1317.38it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 9951 examples [00:10, 1337.45 examples/s]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 9986/52002 [00:07<00:31, 1342.66it/s]#033[A\u001b[0m\n",
      "\u001b[34m19%|█▉        | 10121/52002 [00:07<00:31, 1315.38it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 10148 examples [00:10, 1326.10 examples/s]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 10263/52002 [00:07<00:31, 1343.75it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 10287 examples [00:10, 1338.95 examples/s]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 10398/52002 [00:07<00:31, 1326.59it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 10490 examples [00:10, 1340.60 examples/s]\u001b[0m\n",
      "\u001b[34m20%|██        | 10542/52002 [00:07<00:30, 1357.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 10626 examples [00:11, 1343.55 examples/s]\u001b[0m\n",
      "\u001b[34m21%|██        | 10678/52002 [00:07<00:30, 1352.85it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 10767 examples [00:11, 1358.16 examples/s]\u001b[0m\n",
      "\u001b[34m21%|██        | 10821/52002 [00:07<00:29, 1373.61it/s]#033[A\u001b[0m\n",
      "\u001b[34m21%|██        | 10959/52002 [00:08<00:30, 1351.04it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 10970 examples [00:11, 1351.47 examples/s]\u001b[0m\n",
      "\u001b[34m21%|██▏       | 11095/52002 [00:08<00:30, 1327.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 11161 examples [00:11, 1316.23 examples/s]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 11228/52002 [00:08<00:31, 1301.09it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 11353 examples [00:11, 1302.92 examples/s]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 11359/52002 [00:08<00:31, 1302.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 11505 examples [00:11, 1350.69 examples/s]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 11508/52002 [00:08<00:29, 1353.78it/s]#033[A\u001b[0m\n",
      "\u001b[34m22%|██▏       | 11644/52002 [00:08<00:30, 1313.47it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 11703 examples [00:11, 1331.64 examples/s]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 11776/52002 [00:08<00:30, 1307.94it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 11894 examples [00:12, 1306.19 examples/s]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 11910/52002 [00:08<00:30, 1316.02it/s]#033[A\u001b[0m\n",
      "\u001b[34m23%|██▎       | 12042/52002 [00:08<00:30, 1307.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 12086 examples [00:12, 1294.34 examples/s]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 12175/52002 [00:09<00:30, 1313.48it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 12221 examples [00:12, 1303.85 examples/s]\u001b[0m\n",
      "\u001b[34m24%|██▎       | 12313/52002 [00:09<00:29, 1332.54it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 12365 examples [00:12, 1336.24 examples/s]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 12449/52002 [00:09<00:29, 1338.62it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 12508 examples [00:12, 1354.70 examples/s]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 12588/52002 [00:09<00:29, 1351.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 12711 examples [00:12, 1350.46 examples/s]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 12724/52002 [00:09<00:29, 1346.95it/s]#033[A\u001b[0m\n",
      "\u001b[34m25%|██▍       | 12859/52002 [00:09<00:29, 1340.73it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 12918 examples [00:12, 1353.45 examples/s]\u001b[0m\n",
      "\u001b[34m25%|██▍       | 12997/52002 [00:09<00:28, 1350.63it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 13108 examples [00:12, 1323.84 examples/s]\u001b[0m\n",
      "\u001b[34m25%|██▌       | 13133/52002 [00:09<00:29, 1335.83it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 13253 examples [00:13, 1351.70 examples/s]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 13275/52002 [00:09<00:28, 1357.77it/s]#033[A\u001b[0m\n",
      "\u001b[34m26%|██▌       | 13411/52002 [00:09<00:29, 1324.87it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 13450 examples [00:13, 1335.90 examples/s]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 13559/52002 [00:10<00:28, 1369.19it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 13591 examples [00:13, 1351.05 examples/s]\u001b[0m\n",
      "\u001b[34m26%|██▋       | 13697/52002 [00:10<00:28, 1336.17it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 13792 examples [00:13, 1343.07 examples/s]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 13835/52002 [00:10<00:28, 1348.49it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 13931 examples [00:13, 1347.48 examples/s]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 13971/52002 [00:10<00:28, 1347.19it/s]#033[A\u001b[0m\n",
      "\u001b[34m27%|██▋       | 14106/52002 [00:10<00:29, 1275.95it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 14109 examples [00:13, 1289.99 examples/s]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 14236/52002 [00:10<00:29, 1279.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 14240 examples [00:13, 1293.06 examples/s]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 14394/52002 [00:10<00:27, 1364.40it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 14397 examples [00:13, 1363.39 examples/s]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 14534/52002 [00:10<00:27, 1371.88it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 14536 examples [00:14, 1369.47 examples/s]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 14672/52002 [00:10<00:27, 1349.80it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 14741 examples [00:14, 1364.86 examples/s]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 14808/52002 [00:10<00:27, 1352.43it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 14944 examples [00:14, 1357.72 examples/s]\u001b[0m\n",
      "\u001b[34m29%|██▊       | 14948/52002 [00:11<00:27, 1364.87it/s]#033[A\u001b[0m\n",
      "\u001b[34m29%|██▉       | 15085/52002 [00:11<00:27, 1343.14it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 15137 examples [00:14, 1322.94 examples/s]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 15220/52002 [00:11<00:27, 1344.67it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 15281 examples [00:14, 1349.16 examples/s]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 15355/52002 [00:11<00:27, 1326.01it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 15482 examples [00:14, 1341.82 examples/s]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 15492/52002 [00:11<00:27, 1337.89it/s]#033[A\u001b[0m\n",
      "\u001b[34m30%|███       | 15626/52002 [00:11<00:27, 1335.34it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 15682 examples [00:14, 1332.82 examples/s]\u001b[0m\n",
      "\u001b[34m30%|███       | 15760/52002 [00:11<00:27, 1331.28it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 15821 examples [00:14, 1342.89 examples/s]\u001b[0m\n",
      "\u001b[34m31%|███       | 15896/52002 [00:11<00:27, 1337.18it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 15957 examples [00:15, 1344.23 examples/s]\u001b[0m\n",
      "\u001b[34m31%|███       | 16030/52002 [00:11<00:26, 1335.09it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 16096 examples [00:15, 1354.00 examples/s]\u001b[0m\n",
      "\u001b[34m31%|███       | 16175/52002 [00:11<00:26, 1368.42it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 16298 examples [00:15, 1348.34 examples/s]\u001b[0m\n",
      "\u001b[34m31%|███▏      | 16312/52002 [00:12<00:26, 1362.00it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 16437 examples [00:15, 1356.43 examples/s]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 16449/52002 [00:12<00:26, 1344.56it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 16579 examples [00:15, 1370.83 examples/s]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 16599/52002 [00:12<00:25, 1388.60it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 16725 examples [00:15, 1392.20 examples/s]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 16744/52002 [00:12<00:25, 1406.15it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 16867 examples [00:15, 1396.12 examples/s]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 16885/52002 [00:12<00:25, 1389.32it/s]#033[A\u001b[0m\n",
      "\u001b[34m33%|███▎      | 17025/52002 [00:12<00:25, 1361.61it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 17065 examples [00:15, 1350.63 examples/s]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 17162/52002 [00:12<00:26, 1336.95it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 17202 examples [00:15, 1354.12 examples/s]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 17307/52002 [00:12<00:25, 1368.78it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 17349 examples [00:16, 1381.93 examples/s]\u001b[0m\n",
      "\u001b[34m34%|███▎      | 17445/52002 [00:12<00:25, 1364.11it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 17560 examples [00:16, 1390.12 examples/s]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 17596/52002 [00:13<00:24, 1404.04it/s]#033[A\u001b[0m\n",
      "\u001b[34m34%|███▍      | 17737/52002 [00:13<00:24, 1392.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 17765 examples [00:16, 1377.63 examples/s]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 17878/52002 [00:13<00:24, 1397.62it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 17966 examples [00:16, 1363.89 examples/s]\u001b[0m\n",
      "\u001b[34m35%|███▍      | 18018/52002 [00:13<00:25, 1324.80it/s]#033[A\u001b[0m\n",
      "\u001b[34m35%|███▍      | 18159/52002 [00:13<00:25, 1348.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 18165 examples [00:16, 1349.84 examples/s]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 18295/52002 [00:13<00:25, 1308.23it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 18350 examples [00:16, 1310.78 examples/s]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 18427/52002 [00:13<00:25, 1305.62it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 18540 examples [00:17, 1292.12 examples/s]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 18558/52002 [00:13<00:25, 1290.00it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 18687 examples [00:17, 1328.57 examples/s]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 18700/52002 [00:13<00:25, 1326.71it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 18832 examples [00:17, 1356.57 examples/s]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 18844/52002 [00:13<00:24, 1359.09it/s]#033[A\u001b[0m\n",
      "\u001b[34m37%|███▋      | 18981/52002 [00:14<00:24, 1341.48it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19023 examples [00:17, 1325.95 examples/s]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 19116/52002 [00:14<00:24, 1334.49it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19160 examples [00:17, 1335.96 examples/s]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 19258/52002 [00:14<00:24, 1359.56it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19311 examples [00:17, 1379.06 examples/s]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 19396/52002 [00:14<00:23, 1364.91it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19504 examples [00:17, 1340.80 examples/s]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 19533/52002 [00:14<00:24, 1316.33it/s]#033[A\u001b[0m\n",
      "\u001b[34m38%|███▊      | 19671/52002 [00:14<00:24, 1331.36it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19704 examples [00:17, 1334.98 examples/s]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 19815/52002 [00:14<00:23, 1360.52it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19847 examples [00:17, 1355.82 examples/s]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 19952/52002 [00:14<00:23, 1350.44it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 19984 examples [00:18, 1356.91 examples/s]\u001b[0m\n",
      "\u001b[34m39%|███▊      | 20088/52002 [00:14<00:23, 1351.39it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 20186 examples [00:18, 1347.98 examples/s]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 20224/52002 [00:14<00:23, 1349.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 20329 examples [00:18, 1363.83 examples/s]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 20363/52002 [00:15<00:23, 1359.84it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 20468 examples [00:18, 1367.98 examples/s]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 20500/52002 [00:15<00:23, 1350.78it/s]#033[A\u001b[0m\n",
      "\u001b[34m40%|███▉      | 20636/52002 [00:15<00:23, 1352.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 20674 examples [00:18, 1366.82 examples/s]\u001b[0m\n",
      "\u001b[34m40%|███▉      | 20798/52002 [00:15<00:21, 1428.86it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 20831 examples [00:18, 1416.18 examples/s]\u001b[0m\n",
      "\u001b[34m40%|████      | 20941/52002 [00:15<00:21, 1422.16it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 21033 examples [00:18, 1386.81 examples/s]\u001b[0m\n",
      "\u001b[34m41%|████      | 21084/52002 [00:15<00:21, 1414.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 21182 examples [00:18, 1410.59 examples/s]\u001b[0m\n",
      "\u001b[34m41%|████      | 21228/52002 [00:15<00:21, 1420.18it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 21328 examples [00:19, 1421.46 examples/s]\u001b[0m\n",
      "\u001b[34m41%|████      | 21371/52002 [00:15<00:22, 1391.17it/s]#033[A\u001b[0m\n",
      "\u001b[34m41%|████▏     | 21517/52002 [00:15<00:21, 1410.99it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 21534 examples [00:19, 1399.66 examples/s]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 21659/52002 [00:15<00:21, 1399.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 21742 examples [00:19, 1392.70 examples/s]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 21800/52002 [00:16<00:21, 1393.68it/s]#033[A\u001b[0m\n",
      "\u001b[34m42%|████▏     | 21940/52002 [00:16<00:21, 1378.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 21947 examples [00:19, 1381.75 examples/s]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 22078/52002 [00:16<00:22, 1348.71it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 22141 examples [00:19, 1349.85 examples/s]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 22214/52002 [00:16<00:22, 1344.90it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 22282 examples [00:19, 1360.96 examples/s]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 22349/52002 [00:16<00:22, 1345.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 22486 examples [00:19, 1355.43 examples/s]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 22488/52002 [00:16<00:21, 1354.79it/s]#033[A\u001b[0m\n",
      "\u001b[34m44%|████▎     | 22625/52002 [00:16<00:21, 1358.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 22687 examples [00:20, 1348.65 examples/s]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 22761/52002 [00:16<00:21, 1348.45it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 22886 examples [00:20, 1338.68 examples/s]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 22896/52002 [00:16<00:21, 1328.97it/s]#033[A\u001b[0m\n",
      "\u001b[34m44%|████▍     | 23029/52002 [00:17<00:21, 1321.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 23093 examples [00:20, 1348.76 examples/s]\u001b[0m\n",
      "\u001b[34m45%|████▍     | 23167/52002 [00:17<00:21, 1337.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 23298 examples [00:20, 1350.81 examples/s]\u001b[0m\n",
      "\u001b[34m45%|████▍     | 23310/52002 [00:17<00:21, 1362.96it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 23450 examples [00:20, 1385.69 examples/s]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 23461/52002 [00:17<00:20, 1406.43it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 23591 examples [00:20, 1388.56 examples/s]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 23603/52002 [00:17<00:20, 1408.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 23739 examples [00:20, 1410.78 examples/s]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 23751/52002 [00:17<00:19, 1425.34it/s]#033[A\u001b[0m\n",
      "\u001b[34m46%|████▌     | 23894/52002 [00:17<00:20, 1399.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 23948 examples [00:20, 1397.91 examples/s]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 24035/52002 [00:17<00:20, 1352.61it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 24135 examples [00:21, 1342.17 examples/s]\u001b[0m\n",
      "\u001b[34m46%|████▋     | 24171/52002 [00:17<00:20, 1328.51it/s]#033[A\u001b[0m\n",
      "\u001b[34m47%|████▋     | 24317/52002 [00:17<00:20, 1365.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 24348 examples [00:21, 1364.60 examples/s]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 24459/52002 [00:18<00:19, 1380.94it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 24486 examples [00:21, 1365.42 examples/s]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 24599/52002 [00:18<00:19, 1384.52it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 24628 examples [00:21, 1377.74 examples/s]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 24738/52002 [00:18<00:19, 1384.59it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 24767 examples [00:21, 1380.44 examples/s]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 24877/52002 [00:18<00:19, 1367.04it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 24972 examples [00:21, 1373.69 examples/s]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 25014/52002 [00:18<00:20, 1348.52it/s]#033[A\u001b[0m\n",
      "\u001b[34m48%|████▊     | 25155/52002 [00:18<00:19, 1353.53it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 25169 examples [00:21, 1349.18 examples/s]\u001b[0m\n",
      "\u001b[34m49%|████▊     | 25291/52002 [00:18<00:19, 1350.33it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 25367 examples [00:21, 1337.04 examples/s]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 25428/52002 [00:18<00:19, 1355.73it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 25516 examples [00:22, 1370.87 examples/s]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 25575/52002 [00:18<00:19, 1386.75it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 25660 examples [00:22, 1385.71 examples/s]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 25714/52002 [00:18<00:19, 1380.35it/s]#033[A\u001b[0m\n",
      "\u001b[34m50%|████▉     | 25853/52002 [00:19<00:19, 1371.03it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 25860 examples [00:22, 1365.75 examples/s]\u001b[0m\n",
      "\u001b[34m50%|████▉     | 25991/52002 [00:19<00:19, 1347.32it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 26053 examples [00:22, 1335.43 examples/s]\u001b[0m\n",
      "\u001b[34m50%|█████     | 26127/52002 [00:19<00:19, 1347.84it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 26196 examples [00:22, 1354.82 examples/s]\u001b[0m\n",
      "\u001b[34m51%|█████     | 26270/52002 [00:19<00:18, 1371.61it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 26345 examples [00:22, 1386.43 examples/s]\u001b[0m\n",
      "\u001b[34m51%|█████     | 26408/52002 [00:19<00:18, 1371.26it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 26543 examples [00:22, 1360.73 examples/s]\u001b[0m\n",
      "\u001b[34m51%|█████     | 26546/52002 [00:19<00:18, 1361.84it/s]#033[A\u001b[0m\n",
      "\u001b[34m51%|█████▏    | 26683/52002 [00:19<00:18, 1354.62it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 26741 examples [00:23, 1341.53 examples/s]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 26819/52002 [00:19<00:19, 1318.90it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 26943 examples [00:23, 1339.36 examples/s]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 26960/52002 [00:19<00:18, 1343.67it/s]#033[A\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 27095/52002 [00:19<00:18, 1326.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 27142 examples [00:23, 1330.99 examples/s]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 27238/52002 [00:20<00:18, 1352.50it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 27283 examples [00:23, 1346.04 examples/s]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 27374/52002 [00:20<00:18, 1349.33it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 27426 examples [00:23, 1365.17 examples/s]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 27523/52002 [00:20<00:17, 1389.57it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 27564 examples [00:23, 1366.76 examples/s]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 27663/52002 [00:20<00:17, 1356.13it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 27765 examples [00:23, 1349.96 examples/s]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 27799/52002 [00:20<00:18, 1343.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 27910 examples [00:23, 1367.74 examples/s]\u001b[0m\n",
      "\u001b[34m54%|█████▎    | 27939/52002 [00:20<00:17, 1356.00it/s]#033[A\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 28075/52002 [00:20<00:18, 1317.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28098 examples [00:24, 1323.17 examples/s]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 28208/52002 [00:20<00:18, 1307.03it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28291 examples [00:24, 1305.60 examples/s]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 28339/52002 [00:20<00:18, 1302.59it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28424 examples [00:24, 1307.85 examples/s]\u001b[0m\n",
      "\u001b[34m55%|█████▍    | 28471/52002 [00:21<00:18, 1304.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28556 examples [00:24, 1306.17 examples/s]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 28602/52002 [00:21<00:18, 1296.91it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28690 examples [00:24, 1313.54 examples/s]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 28744/52002 [00:21<00:17, 1330.06it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28839 examples [00:24, 1359.52 examples/s]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 28886/52002 [00:21<00:17, 1355.57it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 28979 examples [00:24, 1369.96 examples/s]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 29023/52002 [00:21<00:16, 1358.34it/s]#033[A\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 29159/52002 [00:21<00:16, 1347.61it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 29171 examples [00:24, 1330.87 examples/s]\u001b[0m\n",
      "\u001b[34m56%|█████▋    | 29299/52002 [00:21<00:16, 1360.19it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 29317 examples [00:24, 1362.31 examples/s]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 29436/52002 [00:21<00:16, 1350.75it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 29518 examples [00:25, 1351.59 examples/s]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 29575/52002 [00:21<00:16, 1360.59it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 29664 examples [00:25, 1375.86 examples/s]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 29717/52002 [00:21<00:16, 1376.84it/s]#033[A\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 29855/52002 [00:22<00:16, 1338.93it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 29856 examples [00:25, 1339.14 examples/s]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 29998/52002 [00:22<00:16, 1363.30it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30000 examples [00:25, 1330.17 examples/s]\u001b[0m\n",
      "\u001b[34mGenerating train split: 30135 examples [00:25, 1334.21 examples/s]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 30135/52002 [00:22<00:16, 1321.84it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30272 examples [00:25, 1341.39 examples/s]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 30272/52002 [00:22<00:16, 1333.14it/s]#033[A\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 30406/52002 [00:22<00:18, 1147.85it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30440 examples [00:25, 1161.53 examples/s]\u001b[0m\n",
      "\u001b[34m59%|█████▊    | 30529/52002 [00:22<00:18, 1168.24it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30562 examples [00:25, 1172.94 examples/s]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 30652/52002 [00:22<00:18, 1183.06it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30694 examples [00:26, 1209.06 examples/s]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 30794/52002 [00:22<00:16, 1249.10it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30836 examples [00:26, 1262.62 examples/s]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 30931/52002 [00:22<00:16, 1280.74it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 30976 examples [00:26, 1290.88 examples/s]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 31061/52002 [00:23<00:16, 1259.16it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 31166 examples [00:26, 1277.51 examples/s]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 31191/52002 [00:23<00:16, 1268.65it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 31300 examples [00:26, 1289.94 examples/s]\u001b[0m\n",
      "\u001b[34m60%|██████    | 31329/52002 [00:23<00:15, 1297.81it/s]#033[A\u001b[0m\n",
      "\u001b[34m61%|██████    | 31468/52002 [00:23<00:15, 1323.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 31504 examples [00:26, 1312.80 examples/s]\u001b[0m\n",
      "\u001b[34m61%|██████    | 31601/52002 [00:23<00:15, 1319.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 31637 examples [00:26, 1315.35 examples/s]\u001b[0m\n",
      "\u001b[34m61%|██████    | 31754/52002 [00:23<00:14, 1377.66it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 31790 examples [00:26, 1371.17 examples/s]\u001b[0m\n",
      "\u001b[34m61%|██████▏   | 31905/52002 [00:23<00:14, 1415.20it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 31942 examples [00:26, 1410.40 examples/s]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 32047/52002 [00:23<00:14, 1391.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 32141 examples [00:27, 1376.59 examples/s]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 32187/52002 [00:23<00:14, 1393.30it/s]#033[A\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 32327/52002 [00:23<00:14, 1351.98it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 32338 examples [00:27, 1349.76 examples/s]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 32463/52002 [00:24<00:14, 1342.63it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 32549 examples [00:27, 1366.92 examples/s]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 32612/52002 [00:24<00:14, 1384.42it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 32688 examples [00:27, 1365.24 examples/s]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 32751/52002 [00:24<00:14, 1331.06it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 32890 examples [00:27, 1352.11 examples/s]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 32894/52002 [00:24<00:14, 1357.53it/s]#033[A\u001b[0m\n",
      "\u001b[34m64%|██████▎   | 33031/52002 [00:24<00:13, 1355.87it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 33096 examples [00:27, 1354.27 examples/s]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 33167/52002 [00:24<00:14, 1338.42it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 33296 examples [00:27, 1344.81 examples/s]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 33306/52002 [00:24<00:13, 1351.16it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 33438 examples [00:28, 1359.13 examples/s]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 33446/52002 [00:24<00:13, 1365.25it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 33577 examples [00:28, 1364.78 examples/s]\u001b[0m\n",
      "\u001b[34m65%|██████▍   | 33585/52002 [00:24<00:13, 1372.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 33715 examples [00:28, 1365.58 examples/s]\u001b[0m\n",
      "\u001b[34m65%|██████▍   | 33723/52002 [00:24<00:13, 1368.01it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 33853 examples [00:28, 1364.21 examples/s]\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 33861/52002 [00:25<00:13, 1370.52it/s]#033[A\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 33999/52002 [00:25<00:13, 1359.07it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34040 examples [00:28, 1317.23 examples/s]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 34135/52002 [00:25<00:13, 1322.11it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34179 examples [00:28, 1332.49 examples/s]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 34269/52002 [00:25<00:13, 1327.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34317 examples [00:28, 1339.53 examples/s]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 34404/52002 [00:25<00:13, 1333.45it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34518 examples [00:28, 1334.27 examples/s]\u001b[0m\n",
      "\u001b[34m66%|██████▋   | 34538/52002 [00:25<00:13, 1324.24it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34655 examples [00:28, 1339.74 examples/s]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 34681/52002 [00:25<00:12, 1351.64it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34791 examples [00:29, 1340.55 examples/s]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 34817/52002 [00:25<00:12, 1330.88it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 34927 examples [00:29, 1345.44 examples/s]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 34953/52002 [00:25<00:12, 1336.67it/s]#033[A\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 35087/52002 [00:26<00:12, 1309.46it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 35121 examples [00:29, 1321.15 examples/s]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 35229/52002 [00:26<00:12, 1340.39it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 35256 examples [00:29, 1325.06 examples/s]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 35364/52002 [00:26<00:12, 1334.58it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 35392 examples [00:29, 1331.79 examples/s]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 35507/52002 [00:26<00:12, 1360.56it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 35535 examples [00:29, 1356.86 examples/s]\u001b[0m\n",
      "\u001b[34m69%|██████▊   | 35648/52002 [00:26<00:11, 1372.12it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 35676 examples [00:29, 1370.31 examples/s]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 35786/52002 [00:26<00:11, 1368.33it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 35874 examples [00:29, 1346.25 examples/s]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 35923/52002 [00:26<00:12, 1313.51it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36041 examples [00:30, 1260.64 examples/s]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 36055/52002 [00:26<00:12, 1234.54it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36180 examples [00:30, 1288.19 examples/s]\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 36192/52002 [00:26<00:12, 1271.77it/s]#033[A\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 36323/52002 [00:26<00:12, 1279.81it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36371 examples [00:30, 1280.50 examples/s]\u001b[0m\n",
      "\u001b[34m70%|███████   | 36457/52002 [00:27<00:12, 1292.59it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36503 examples [00:30, 1285.98 examples/s]\u001b[0m\n",
      "\u001b[34m70%|███████   | 36602/52002 [00:27<00:11, 1338.33it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36653 examples [00:30, 1338.59 examples/s]\u001b[0m\n",
      "\u001b[34m71%|███████   | 36737/52002 [00:27<00:11, 1337.40it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36797 examples [00:30, 1360.07 examples/s]\u001b[0m\n",
      "\u001b[34m71%|███████   | 36891/52002 [00:27<00:10, 1393.56it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 36952 examples [00:30, 1410.20 examples/s]\u001b[0m\n",
      "\u001b[34m71%|███████   | 37031/52002 [00:27<00:10, 1363.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 37148 examples [00:30, 1368.01 examples/s]\u001b[0m\n",
      "\u001b[34m71%|███████▏  | 37176/52002 [00:27<00:10, 1385.77it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 37293 examples [00:30, 1386.87 examples/s]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 37317/52002 [00:27<00:10, 1390.18it/s]#033[A\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 37457/52002 [00:27<00:10, 1381.38it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 37499 examples [00:31, 1376.51 examples/s]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 37596/52002 [00:27<00:10, 1337.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 37702 examples [00:31, 1364.28 examples/s]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 37743/52002 [00:27<00:10, 1371.02it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 37843 examples [00:31, 1372.91 examples/s]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 37883/52002 [00:28<00:10, 1378.09it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 37982 examples [00:31, 1373.74 examples/s]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 38022/52002 [00:28<00:10, 1351.10it/s]#033[A\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 38171/52002 [00:28<00:09, 1391.23it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 38195 examples [00:31, 1384.24 examples/s]\u001b[0m\n",
      "\u001b[34m74%|███████▎  | 38313/52002 [00:28<00:09, 1397.78it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 38394 examples [00:31, 1361.77 examples/s]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 38454/52002 [00:28<00:09, 1376.02it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 38535 examples [00:31, 1371.92 examples/s]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 38592/52002 [00:28<00:09, 1363.91it/s]#033[A\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 38730/52002 [00:28<00:09, 1367.29it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 38738 examples [00:31, 1363.03 examples/s]\u001b[0m\n",
      "\u001b[34m75%|███████▍  | 38867/52002 [00:28<00:09, 1337.13it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 38938 examples [00:32, 1351.64 examples/s]\u001b[0m\n",
      "\u001b[34m75%|███████▍  | 39001/52002 [00:28<00:09, 1306.31it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 39138 examples [00:32, 1322.30 examples/s]\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 39139/52002 [00:29<00:09, 1324.86it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 39274 examples [00:32, 1329.53 examples/s]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 39276/52002 [00:29<00:09, 1337.18it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 39424 examples [00:32, 1369.86 examples/s]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 39425/52002 [00:29<00:09, 1379.21it/s]#033[A\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 39564/52002 [00:29<00:09, 1367.35it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 39621 examples [00:32, 1346.88 examples/s]\u001b[0m\n",
      "\u001b[34m76%|███████▋  | 39701/52002 [00:29<00:09, 1345.30it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 39759 examples [00:32, 1353.35 examples/s]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 39837/52002 [00:29<00:09, 1349.12it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 39962 examples [00:32, 1348.94 examples/s]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 39974/52002 [00:29<00:08, 1354.83it/s]#033[A\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 40110/52002 [00:29<00:08, 1336.20it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 40162 examples [00:33, 1337.91 examples/s]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 40248/52002 [00:29<00:08, 1346.57it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 40307 examples [00:33, 1364.59 examples/s]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 40383/52002 [00:29<00:08, 1331.95it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 40489 examples [00:33, 1311.64 examples/s]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 40517/52002 [00:30<00:08, 1320.68it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 40641 examples [00:33, 1360.35 examples/s]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 40669/52002 [00:30<00:08, 1377.13it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 40783 examples [00:33, 1368.03 examples/s]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 40807/52002 [00:30<00:08, 1371.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 40929 examples [00:33, 1391.28 examples/s]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 40957/52002 [00:30<00:07, 1407.45it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 41074 examples [00:33, 1375.81 examples/s]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 41098/52002 [00:30<00:07, 1367.81it/s]#033[A\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 41236/52002 [00:30<00:07, 1368.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 41277 examples [00:33, 1363.78 examples/s]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 41374/52002 [00:30<00:07, 1354.15it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 41474 examples [00:34, 1343.60 examples/s]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 41510/52002 [00:30<00:07, 1351.55it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 41620 examples [00:34, 1366.68 examples/s]\u001b[0m\n",
      "\u001b[34m80%|████████  | 41646/52002 [00:30<00:07, 1350.53it/s]#033[A\u001b[0m\n",
      "\u001b[34m80%|████████  | 41782/52002 [00:30<00:07, 1331.50it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 41810 examples [00:34, 1330.93 examples/s]\u001b[0m\n",
      "\u001b[34m81%|████████  | 41916/52002 [00:31<00:07, 1331.89it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 41945 examples [00:34, 1332.41 examples/s]\u001b[0m\n",
      "\u001b[34m81%|████████  | 42050/52002 [00:31<00:07, 1283.49it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 42139 examples [00:34, 1305.80 examples/s]\u001b[0m\n",
      "\u001b[34m81%|████████  | 42186/52002 [00:31<00:07, 1299.54it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 42272 examples [00:34, 1310.56 examples/s]\u001b[0m\n",
      "\u001b[34m81%|████████▏ | 42320/52002 [00:31<00:07, 1311.16it/s]#033[A\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 42459/52002 [00:31<00:07, 1334.09it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 42477 examples [00:34, 1329.10 examples/s]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 42604/52002 [00:31<00:06, 1368.07it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 42622 examples [00:34, 1357.93 examples/s]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 42742/52002 [00:31<00:06, 1352.05it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 42808 examples [00:35, 1314.69 examples/s]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 42878/52002 [00:31<00:06, 1306.41it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 42941 examples [00:35, 1315.53 examples/s]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 43010/52002 [00:31<00:06, 1291.17it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 43144 examples [00:35, 1319.54 examples/s]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 43153/52002 [00:32<00:06, 1324.31it/s]#033[A\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 43286/52002 [00:32<00:06, 1293.08it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 43327 examples [00:35, 1284.41 examples/s]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 43416/52002 [00:32<00:06, 1268.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 43523 examples [00:35, 1288.29 examples/s]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 43553/52002 [00:32<00:06, 1294.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 43668 examples [00:35, 1322.64 examples/s]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 43693/52002 [00:32<00:06, 1323.96it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 43808 examples [00:35, 1340.05 examples/s]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 43836/52002 [00:32<00:06, 1354.58it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 43947 examples [00:35, 1351.88 examples/s]\u001b[0m\n",
      "\u001b[34m85%|████████▍ | 43980/52002 [00:32<00:05, 1374.53it/s]#033[A\u001b[0m\n",
      "\u001b[34m85%|████████▍ | 44118/52002 [00:32<00:05, 1329.88it/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 44139 examples [00:36, 1322.24 examples/s]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 44252/52002 [00:32<00:05, 1327.52it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 44275 examples [00:36, 1331.05 examples/s]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 44391/52002 [00:32<00:05, 1342.28it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 44411 examples [00:36, 1337.94 examples/s]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 44532/52002 [00:33<00:05, 1360.85it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 44553 examples [00:36, 1359.34 examples/s]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 44675/52002 [00:33<00:05, 1378.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 44698 examples [00:36, 1382.38 examples/s]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 44816/52002 [00:33<00:05, 1386.51it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 44841 examples [00:36, 1393.41 examples/s]\u001b[0m\n",
      "\u001b[34m86%|████████▋ | 44958/52002 [00:33<00:05, 1392.88it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 45038 examples [00:36, 1357.23 examples/s]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 45098/52002 [00:33<00:05, 1360.76it/s]#033[A\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 45235/52002 [00:33<00:05, 1347.52it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 45243 examples [00:36, 1354.66 examples/s]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 45377/52002 [00:33<00:04, 1365.49it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 45386 examples [00:36, 1369.67 examples/s]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 45514/52002 [00:33<00:04, 1358.39it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 45598 examples [00:37, 1382.52 examples/s]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 45668/52002 [00:33<00:04, 1410.76it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 45743 examples [00:37, 1397.28 examples/s]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 45810/52002 [00:33<00:04, 1410.58it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 45943 examples [00:37, 1370.50 examples/s]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 45952/52002 [00:34<00:04, 1368.24it/s]#033[A\u001b[0m\n",
      "\u001b[34m89%|████████▊ | 46090/52002 [00:34<00:04, 1287.17it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 46121 examples [00:37, 1300.96 examples/s]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 46226/52002 [00:34<00:04, 1306.47it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 46260 examples [00:37, 1321.86 examples/s]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 46362/52002 [00:34<00:04, 1318.96it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 46399 examples [00:37, 1335.39 examples/s]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 46507/52002 [00:34<00:04, 1355.85it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 46542 examples [00:37, 1359.44 examples/s]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 46644/52002 [00:34<00:03, 1339.72it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 46740 examples [00:37, 1341.32 examples/s]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 46785/52002 [00:34<00:03, 1356.91it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 46883 examples [00:38, 1362.61 examples/s]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 46922/52002 [00:34<00:03, 1360.60it/s]#033[A\u001b[0m\n",
      "\u001b[34m90%|█████████ | 47059/52002 [00:34<00:03, 1344.17it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 47085 examples [00:38, 1353.46 examples/s]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 47194/52002 [00:35<00:03, 1339.85it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 47283 examples [00:38, 1339.01 examples/s]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 47329/52002 [00:35<00:03, 1342.71it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 47422 examples [00:38, 1348.98 examples/s]\u001b[0m\n",
      "\u001b[34m91%|█████████▏| 47464/52002 [00:35<00:03, 1337.20it/s]#033[A\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 47598/52002 [00:35<00:03, 1331.35it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 47622 examples [00:38, 1339.58 examples/s]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 47733/52002 [00:35<00:03, 1336.69it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 47758 examples [00:38, 1341.04 examples/s]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 47867/52002 [00:35<00:03, 1326.97it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 47949 examples [00:38, 1313.59 examples/s]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 48000/52002 [00:35<00:03, 1297.13it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 48083 examples [00:38, 1317.55 examples/s]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 48142/52002 [00:35<00:02, 1331.03it/s]#033[A\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 48279/52002 [00:35<00:02, 1340.78it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 48288 examples [00:39, 1331.53 examples/s]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 48414/52002 [00:35<00:02, 1325.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 48487 examples [00:39, 1325.29 examples/s]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 48551/52002 [00:36<00:02, 1337.66it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 48626 examples [00:39, 1337.76 examples/s]\u001b[0m\n",
      "\u001b[34m94%|█████████▎| 48688/52002 [00:36<00:02, 1345.82it/s]#033[A\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 48826/52002 [00:36<00:02, 1355.22it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 48836 examples [00:39, 1356.99 examples/s]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 48968/52002 [00:36<00:02, 1373.21it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 48977 examples [00:39, 1367.86 examples/s]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 49106/52002 [00:36<00:02, 1302.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 49156 examples [00:39, 1305.86 examples/s]\u001b[0m\n",
      "\u001b[34m95%|█████████▍| 49250/52002 [00:36<00:02, 1340.46it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 49302 examples [00:39, 1342.06 examples/s]\u001b[0m\n",
      "\u001b[34m95%|█████████▍| 49388/52002 [00:36<00:01, 1350.56it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 49442 examples [00:39, 1355.73 examples/s]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 49524/52002 [00:36<00:01, 1324.70it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 49626 examples [00:40, 1303.10 examples/s]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 49657/52002 [00:36<00:01, 1313.78it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 49777 examples [00:40, 1351.30 examples/s]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 49803/52002 [00:36<00:01, 1355.71it/s]#033[A\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 49939/52002 [00:37<00:01, 1341.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 49975 examples [00:40, 1337.39 examples/s]\u001b[0m\n",
      "\u001b[34m96%|█████████▋| 50074/52002 [00:37<00:01, 1283.48it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 50161 examples [00:40, 1301.68 examples/s]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 50214/52002 [00:37<00:01, 1313.61it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 50304 examples [00:40, 1329.52 examples/s]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 50349/52002 [00:37<00:01, 1321.66it/s]#033[A\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 50482/52002 [00:37<00:01, 1306.27it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 50492 examples [00:40, 1297.97 examples/s]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 50618/52002 [00:37<00:01, 1319.35it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 50631 examples [00:40, 1316.78 examples/s]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 50751/52002 [00:37<00:00, 1317.16it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 50765 examples [00:40, 1320.25 examples/s]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 50901/52002 [00:37<00:00, 1369.73it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 50917 examples [00:41, 1372.32 examples/s]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 51039/52002 [00:37<00:00, 1329.96it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 51110 examples [00:41, 1338.05 examples/s]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 51175/52002 [00:37<00:00, 1336.47it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 51248 examples [00:41, 1348.46 examples/s]\u001b[0m\n",
      "\u001b[34m99%|█████████▊| 51316/52002 [00:38<00:00, 1356.90it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 51448 examples [00:41, 1340.29 examples/s]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 51452/52002 [00:38<00:00, 1342.79it/s]#033[A\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 51588/52002 [00:38<00:00, 1346.49it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 51655 examples [00:41, 1350.60 examples/s]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 51730/52002 [00:38<00:00, 1365.53it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 51798 examples [00:41, 1367.45 examples/s]\u001b[0m\n",
      "\u001b[34m100%|█████████▉| 51867/52002 [00:38<00:00, 1365.59it/s]#033[A\u001b[0m\n",
      "\u001b[34mGenerating train split: 51995 examples [00:41, 1341.45 examples/s]\u001b[0m\n",
      "\u001b[34m100%|██████████| 52002/52002 [00:38<00:00, 1346.65it/s]\u001b[0m\n",
      "\u001b[34mDataset generator downloaded and prepared to /root/.cache/huggingface/datasets/generator/default-20105bf96995fb33/0.0.0. Subsequent calls will reuse this data.\u001b[0m\n",
      "\u001b[34mSaving the dataset (0/1 shards):   0%|          | 0/52002 [00:00<?, ? examples/s]\u001b[0m\n",
      "\u001b[34mSaving the dataset (1/1 shards): 100%|██████████| 52002/52002 [00:00<00:00, 3062685.30 examples/s]\u001b[0m\n",
      "\u001b[34m===================================BUG REPORT===================================\u001b[0m\n",
      "\u001b[34mWelcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues\u001b[0m\n",
      "\u001b[34m================================================================================\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/nvidia/lib'), PosixPath('/usr/local/nvidia/lib64')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: /opt/conda/lib/python3.9/site-packages/smdistributed/dataparallel/lib:/opt/amazon/openmpi/lib/:/opt/amazon/efa/lib/:/opt/conda/lib:/usr/local/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/lib did not contain libcudart.so as expected! Searching further paths...\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/v2/credentials/proxy-8a54c3937801ecc549e303dab581744b8115ea24cbdebcdfa26c831a9a668f6d-customer')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/opt/ml/output/intermediate')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('$(dirname $(which conda))/..')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/opt/conda/lib/python39.zip')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('\"homogeneousCluster\",\"current_instance_group_hosts\"'), PosixPath('true,\"is_modelparallel_enabled\"'), PosixPath('0,\"output_data_dir\"'), PosixPath('\"ml.g5.12xlarge\",\"hosts\"'), PosixPath('{},\"input_config_dir\"'), PosixPath('//sagemaker-us-west-2-687912291502/chatglm-lora-demo-2023-06-20-09-24-44-574/source/sourcedir.tar.gz\",\"module_name\"'), PosixPath('\"chatglm-lora-demo-2023-06-20-09-24-44-574\",\"log_level\"'), PosixPath('\"eth0\",\"num_cpus\"'), PosixPath('\"algo-1\",\"current_instance_type\"'), PosixPath('\"sagemaker_pytorch_container.training'), PosixPath('\"ml.g5.12xlarge\",\"distribution_hosts\"'), PosixPath('20,\"master_hostname\"'), PosixPath('{},\"current_host\"'), PosixPath('\"eth0\"},\"user_entry_point\"'), PosixPath('{\"current_group_name\"'), PosixPath('{\"homogeneousCluster\"'), PosixPath('\"/opt/ml/input\",\"instance_groups\"'), PosixPath('\"s3'), PosixPath('\"ml.g5.12xlarge\"}],\"network_interface_name\"'), PosixPath('\"/opt/ml/output\",\"output_intermediate_dir\"'), PosixPath('main\",\"hosts\"'), PosixPath('[\"algo-1\"],\"hyperparameters\"'), PosixPath('null,\"is_smddpmprun_installed\"'), PosixPath('\"train.sh\"}'), PosixPath('\"/opt/ml/input/config\",\"input_data_config\"'), PosixPath('{},\"input_dir\"'), PosixPath('\"train.sh\",\"network_interface_name\"'), PosixPath('[\"homogeneousCluster\"],\"instance_groups_dict\"'), PosixPath('\"algo-1\",\"model_dir\"'), PosixPath('false,\"is_master\"'), PosixPath('[\"algo-1\"],\"current_instance_type\"'), PosixPath('[],\"framework_module\"'), PosixPath('{\"additional_framework_parameters\"'), PosixPath('\"ml.g5.12xlarge\"}},\"is_hetero\"'), PosixPath('true,\"job_name\"'), PosixPath('48,\"num_gpus\"'), PosixPath('\"/opt/ml/output/intermediate\",\"resource_config\"'), PosixPath('\"/opt/ml/output/data\",\"output_dir\"'), PosixPath('\"algo-1\",\"current_instance_group\"'), PosixPath('{},\"channel_input_dirs\"'), PosixPath('\"homogeneousCluster\",\"current_host\"'), PosixPath('[\"algo-1\"],\"instance_groups\"'), PosixPath('[{\"hosts\"'), PosixPath('\"homogeneousCluster\",\"instance_type\"'), PosixPath('\"/opt/ml/model\",\"module_dir\"'), PosixPath('[],\"distribution_instance_groups\"'), PosixPath('{\"hosts\"'), PosixPath('[\"algo-1\"],\"instance_group_name\"'), PosixPath('4,\"num_neurons\"')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('//sagemaker-us-west-2-687912291502/chatglm-lora-demo-2023-06-20-09-24-44-574/source/sourcedir.tar.gz'), PosixPath('s3')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('us-west-2'), PosixPath('687912291502'), PosixPath('arn'), PosixPath('training-job/chatglm-lora-demo-2023-06-20-09-24-44-574'), PosixPath('sagemaker'), PosixPath('aws')}\n",
      "  warn(msg)\u001b[0m\n",
      "\u001b[34mCUDA_SETUP: WARNING! libcudart.so not found in any environmental path. Searching /usr/local/cuda/lib64...\u001b[0m\n",
      "\u001b[34mCUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so\u001b[0m\n",
      "\u001b[34mCUDA SETUP: Highest compute capability among GPUs detected: 8.6\u001b[0m\n",
      "\u001b[34mCUDA SETUP: Detected CUDA version 117\u001b[0m\n",
      "\u001b[34mCUDA SETUP: Loading binary /opt/conda/lib/python3.9/site-packages/bitsandbytes/libbitsandbytes_cuda117.so...\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mExplicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\u001b[0m\n",
      "\u001b[34mDownloading (…)/modeling_chatglm.py:   0%|          | 0.00/57.6k [00:00<?, ?B/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)/modeling_chatglm.py: 100%|██████████| 57.6k/57.6k [00:00<00:00, 8.97MB/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)main/quantization.py:   0%|          | 0.00/15.1k [00:00<?, ?B/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)main/quantization.py: 100%|██████████| 15.1k/15.1k [00:00<00:00, 4.11MB/s]\u001b[0m\n",
      "\u001b[34mOverriding torch_dtype=None with `torch_dtype=torch.float16` due to requirements of `bitsandbytes` to enable model loading in mixed int8. Either pass torch_dtype=torch.float16 or don't pass this argument at all to remove this warning.\u001b[0m\n",
      "\u001b[34mOverriding torch_dtype=None with `torch_dtype=torch.float16` due to requirements of `bitsandbytes` to enable model loading in mixed int8. Either pass torch_dtype=torch.float16 or don't pass this argument at all to remove this warning.\u001b[0m\n",
      "\u001b[34mDownloading (…)model.bin.index.json:   0%|          | 0.00/33.4k [00:00<?, ?B/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)model.bin.index.json: 100%|██████████| 33.4k/33.4k [00:00<00:00, 11.1MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:   0%|          | 0/8 [00:00<?, ?it/s]\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:   0%|          | 0.00/1.74G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:   2%|▏         | 31.5M/1.74G [00:00<00:06, 275MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:   4%|▍         | 73.4M/1.74G [00:00<00:04, 343MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:   7%|▋         | 115M/1.74G [00:00<00:04, 369MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:   9%|▉         | 157M/1.74G [00:00<00:04, 386MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  11%|█▏        | 199M/1.74G [00:00<00:03, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  14%|█▍        | 241M/1.74G [00:00<00:03, 397MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  16%|█▋        | 283M/1.74G [00:00<00:03, 402MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  19%|█▊        | 325M/1.74G [00:00<00:03, 401MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  21%|██        | 367M/1.74G [00:00<00:03, 405MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  23%|██▎       | 409M/1.74G [00:01<00:03, 403MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  26%|██▌       | 451M/1.74G [00:01<00:03, 405MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  28%|██▊       | 493M/1.74G [00:01<00:03, 403MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  31%|███       | 535M/1.74G [00:01<00:03, 401MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  33%|███▎      | 577M/1.74G [00:01<00:02, 401MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  36%|███▌      | 619M/1.74G [00:01<00:02, 402MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  38%|███▊      | 661M/1.74G [00:01<00:02, 403MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  40%|████      | 703M/1.74G [00:01<00:02, 404MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  43%|████▎     | 744M/1.74G [00:01<00:02, 407MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  45%|████▌     | 786M/1.74G [00:01<00:02, 409MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  48%|████▊     | 828M/1.74G [00:02<00:02, 409MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  50%|████▉     | 870M/1.74G [00:02<00:02, 409MB/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  52%|█████▏    | 912M/1.74G [00:02<00:02, 408MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  55%|█████▍    | 954M/1.74G [00:02<00:01, 407MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  57%|█████▋    | 996M/1.74G [00:02<00:01, 407MB/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  60%|█████▉    | 1.04G/1.74G [00:02<00:01, 409MB/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  62%|██████▏   | 1.08G/1.74G [00:02<00:01, 409MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  64%|██████▍   | 1.12G/1.74G [00:02<00:01, 408MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  67%|██████▋   | 1.16G/1.74G [00:02<00:01, 400MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  69%|██████▉   | 1.21G/1.74G [00:03<00:01, 401MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  72%|███████▏  | 1.25G/1.74G [00:03<00:01, 398MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  74%|███████▍  | 1.29G/1.74G [00:03<00:01, 398MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  77%|███████▋  | 1.33G/1.74G [00:03<00:01, 398MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  79%|███████▉  | 1.37G/1.74G [00:03<00:00, 398MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  81%|████████▏ | 1.42G/1.74G [00:03<00:00, 403MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  84%|████████▎ | 1.46G/1.74G [00:03<00:00, 402MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  86%|████████▌ | 1.50G/1.74G [00:03<00:00, 401MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  89%|████████▊ | 1.54G/1.74G [00:03<00:00, 406MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  91%|█████████ | 1.58G/1.74G [00:03<00:00, 404MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  93%|█████████▎| 1.63G/1.74G [00:04<00:00, 407MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  96%|█████████▌| 1.67G/1.74G [00:04<00:00, 401MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";:  98%|█████████▊| 1.71G/1.74G [00:04<00:00, 398MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00001-of-00008.bin\";: 100%|██████████| 1.74G/1.74G [00:04<00:00, 400MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  12%|█▎        | 1/8 [00:04<00:31,  4.53s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:   0%|          | 0.00/1.88G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:   2%|▏         | 31.5M/1.88G [00:00<00:07, 257MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:   4%|▍         | 73.4M/1.88G [00:00<00:05, 328MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:   6%|▌         | 115M/1.88G [00:00<00:05, 350MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:   8%|▊         | 157M/1.88G [00:00<00:04, 364MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  11%|█         | 199M/1.88G [00:00<00:04, 370MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  13%|█▎        | 241M/1.88G [00:00<00:04, 374MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  15%|█▌        | 283M/1.88G [00:00<00:04, 375MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  17%|█▋        | 325M/1.88G [00:00<00:04, 375MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  20%|█▉        | 367M/1.88G [00:01<00:04, 378MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  22%|██▏       | 409M/1.88G [00:01<00:03, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  24%|██▍       | 451M/1.88G [00:01<00:03, 383MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  26%|██▌       | 493M/1.88G [00:01<00:03, 383MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  28%|██▊       | 535M/1.88G [00:01<00:03, 384MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  31%|███       | 577M/1.88G [00:01<00:03, 383MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  33%|███▎      | 619M/1.88G [00:01<00:03, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  35%|███▌      | 661M/1.88G [00:01<00:03, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  37%|███▋      | 703M/1.88G [00:01<00:03, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  40%|███▉      | 744M/1.88G [00:01<00:02, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  42%|████▏     | 786M/1.88G [00:02<00:02, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  44%|████▍     | 828M/1.88G [00:02<00:02, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  46%|████▋     | 870M/1.88G [00:02<00:02, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  49%|████▊     | 912M/1.88G [00:02<00:02, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  51%|█████     | 954M/1.88G [00:02<00:02, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  53%|█████▎    | 996M/1.88G [00:02<00:02, 370MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  55%|█████▌    | 1.04G/1.88G [00:02<00:02, 373MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  57%|█████▋    | 1.08G/1.88G [00:02<00:02, 374MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  60%|█████▉    | 1.12G/1.88G [00:02<00:02, 375MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  62%|██████▏   | 1.16G/1.88G [00:03<00:01, 378MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  64%|██████▍   | 1.21G/1.88G [00:03<00:01, 379MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  66%|██████▋   | 1.25G/1.88G [00:03<00:01, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  69%|██████▊   | 1.29G/1.88G [00:03<00:01, 380MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  71%|███████   | 1.33G/1.88G [00:03<00:01, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  73%|███████▎  | 1.37G/1.88G [00:03<00:01, 380MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  75%|███████▌  | 1.42G/1.88G [00:03<00:01, 378MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  78%|███████▊  | 1.46G/1.88G [00:03<00:01, 380MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  80%|███████▉  | 1.50G/1.88G [00:03<00:01, 378MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  82%|████████▏ | 1.54G/1.88G [00:04<00:00, 379MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  84%|████████▍ | 1.58G/1.88G [00:04<00:00, 379MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  86%|████████▋ | 1.63G/1.88G [00:04<00:00, 380MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  89%|████████▊ | 1.67G/1.88G [00:04<00:00, 380MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  91%|█████████ | 1.71G/1.88G [00:04<00:00, 378MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  93%|█████████▎| 1.75G/1.88G [00:04<00:00, 380MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  95%|█████████▌| 1.79G/1.88G [00:04<00:00, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";:  98%|█████████▊| 1.84G/1.88G [00:04<00:00, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";: 100%|█████████▉| 1.88G/1.88G [00:04<00:00, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00002-of-00008.bin\";: 100%|██████████| 1.88G/1.88G [00:04<00:00, 377MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  25%|██▌       | 2/8 [00:09<00:29,  4.91s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:   0%|          | 0.00/1.98G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:   1%|          | 21.0M/1.98G [00:00<00:09, 207MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:   3%|▎         | 62.9M/1.98G [00:00<00:06, 313MB/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:   5%|▌         | 105M/1.98G [00:00<00:05, 349MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:   7%|▋         | 147M/1.98G [00:00<00:05, 353MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  10%|▉         | 189M/1.98G [00:00<00:04, 367MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  12%|█▏        | 231M/1.98G [00:00<00:04, 375MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  14%|█▍        | 273M/1.98G [00:00<00:04, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  16%|█▌        | 315M/1.98G [00:00<00:04, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  18%|█▊        | 357M/1.98G [00:00<00:04, 383MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  20%|██        | 398M/1.98G [00:01<00:04, 386MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  22%|██▏       | 440M/1.98G [00:01<00:03, 387MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  24%|██▍       | 482M/1.98G [00:01<00:03, 388MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  26%|██▋       | 524M/1.98G [00:01<00:03, 382MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  29%|██▊       | 566M/1.98G [00:01<00:03, 387MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  31%|███       | 608M/1.98G [00:01<00:03, 390MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  33%|███▎      | 650M/1.98G [00:01<00:03, 389MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  35%|███▍      | 692M/1.98G [00:01<00:03, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  37%|███▋      | 734M/1.98G [00:01<00:03, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  39%|███▉      | 776M/1.98G [00:02<00:03, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  41%|████▏     | 818M/1.98G [00:02<00:02, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  43%|████▎     | 860M/1.98G [00:02<00:02, 390MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  46%|████▌     | 902M/1.98G [00:02<00:02, 391MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  48%|████▊     | 944M/1.98G [00:02<00:02, 391MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  50%|████▉     | 986M/1.98G [00:02<00:02, 388MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  52%|█████▏    | 1.03G/1.98G [00:02<00:02, 389MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  54%|█████▍    | 1.07G/1.98G [00:02<00:02, 389MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  56%|█████▌    | 1.11G/1.98G [00:02<00:02, 390MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  58%|█████▊    | 1.15G/1.98G [00:03<00:02, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  60%|██████    | 1.20G/1.98G [00:03<00:01, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  62%|██████▏   | 1.24G/1.98G [00:03<00:01, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  65%|██████▍   | 1.28G/1.98G [00:03<00:01, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  67%|██████▋   | 1.32G/1.98G [00:03<00:01, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  69%|██████▉   | 1.36G/1.98G [00:03<00:01, 391MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  71%|███████   | 1.41G/1.98G [00:03<00:01, 391MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  73%|███████▎  | 1.45G/1.98G [00:03<00:01, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  75%|███████▌  | 1.49G/1.98G [00:03<00:01, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  77%|███████▋  | 1.53G/1.98G [00:03<00:01, 381MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  79%|███████▉  | 1.57G/1.98G [00:04<00:01, 386MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  82%|████████▏ | 1.61G/1.98G [00:04<00:00, 387MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  84%|████████▎ | 1.66G/1.98G [00:04<00:00, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  86%|████████▌ | 1.70G/1.98G [00:04<00:00, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  88%|████████▊ | 1.74G/1.98G [00:04<00:00, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  90%|█████████ | 1.78G/1.98G [00:04<00:00, 393MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  92%|█████████▏| 1.82G/1.98G [00:04<00:00, 394MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  94%|█████████▍| 1.87G/1.98G [00:04<00:00, 392MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  96%|█████████▋| 1.91G/1.98G [00:04<00:00, 394MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";:  98%|█████████▊| 1.95G/1.98G [00:05<00:00, 394MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00003-of-00008.bin\";: 100%|██████████| 1.98G/1.98G [00:05<00:00, 386MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  38%|███▊      | 3/8 [00:15<00:25,  5.09s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:   0%|          | 0.00/1.91G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:   2%|▏         | 31.5M/1.91G [00:00<00:06, 292MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:   4%|▍         | 73.4M/1.91G [00:00<00:05, 364MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:   6%|▌         | 115M/1.91G [00:00<00:04, 381MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:   8%|▊         | 157M/1.91G [00:00<00:04, 355MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  10%|█         | 199M/1.91G [00:00<00:04, 375MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  13%|█▎        | 241M/1.91G [00:00<00:04, 387MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  15%|█▌        | 294M/1.91G [00:00<00:04, 399MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  18%|█▊        | 336M/1.91G [00:00<00:03, 402MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  20%|█▉        | 377M/1.91G [00:00<00:03, 403MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  22%|██▏       | 430M/1.91G [00:01<00:03, 409MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  25%|██▍       | 472M/1.91G [00:01<00:03, 403MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  27%|██▋       | 524M/1.91G [00:01<00:03, 406MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  30%|██▉       | 566M/1.91G [00:01<00:03, 399MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  32%|███▏      | 608M/1.91G [00:01<00:03, 404MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  34%|███▍      | 650M/1.91G [00:01<00:03, 407MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  36%|███▌      | 692M/1.91G [00:01<00:02, 408MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  38%|███▊      | 734M/1.91G [00:01<00:02, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  41%|████      | 776M/1.91G [00:01<00:02, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  43%|████▎     | 818M/1.91G [00:02<00:02, 412MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  45%|████▍     | 860M/1.91G [00:02<00:02, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  47%|████▋     | 902M/1.91G [00:02<00:02, 412MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  49%|████▉     | 944M/1.91G [00:02<00:02, 411MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  52%|█████▏    | 986M/1.91G [00:02<00:02, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  54%|█████▎    | 1.03G/1.91G [00:02<00:02, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  56%|█████▌    | 1.07G/1.91G [00:02<00:02, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  58%|█████▊    | 1.11G/1.91G [00:02<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  60%|██████    | 1.15G/1.91G [00:02<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  62%|██████▏   | 1.20G/1.91G [00:02<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  65%|██████▍   | 1.24G/1.91G [00:03<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  67%|██████▋   | 1.28G/1.91G [00:03<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  69%|██████▉   | 1.32G/1.91G [00:03<00:01, 411MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  71%|███████   | 1.36G/1.91G [00:03<00:01, 412MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  73%|███████▎  | 1.41G/1.91G [00:03<00:01, 411MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  76%|███████▌  | 1.45G/1.91G [00:03<00:01, 412MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  78%|███████▊  | 1.49G/1.91G [00:03<00:01, 411MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  80%|████████  | 1.53G/1.91G [00:03<00:00, 412MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  82%|████████▏ | 1.57G/1.91G [00:03<00:00, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  84%|████████▍ | 1.61G/1.91G [00:03<00:00, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  87%|████████▋ | 1.66G/1.91G [00:04<00:00, 405MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  89%|████████▉ | 1.70G/1.91G [00:04<00:00, 407MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  91%|█████████ | 1.74G/1.91G [00:04<00:00, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  94%|█████████▎| 1.79G/1.91G [00:04<00:00, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  96%|█████████▌| 1.84G/1.91G [00:04<00:00, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";:  98%|█████████▊| 1.88G/1.91G [00:04<00:00, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00004-of-00008.bin\";: 100%|██████████| 1.91G/1.91G [00:04<00:00, 406MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  50%|█████     | 4/8 [00:19<00:20,  5.02s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:   0%|          | 0.00/1.88G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:   2%|▏         | 31.5M/1.88G [00:00<00:06, 284MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:   4%|▍         | 73.4M/1.88G [00:00<00:05, 359MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:   6%|▌         | 115M/1.88G [00:00<00:04, 382MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:   9%|▉         | 168M/1.88G [00:00<00:04, 406MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  12%|█▏        | 220M/1.88G [00:00<00:04, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  15%|█▍        | 273M/1.88G [00:00<00:03, 422MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  17%|█▋        | 325M/1.88G [00:00<00:03, 425MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  20%|██        | 377M/1.88G [00:00<00:03, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  23%|██▎       | 430M/1.88G [00:01<00:03, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  26%|██▌       | 482M/1.88G [00:01<00:03, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  28%|██▊       | 535M/1.88G [00:01<00:03, 425MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  31%|███       | 587M/1.88G [00:01<00:03, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  34%|███▍      | 640M/1.88G [00:01<00:02, 430MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  37%|███▋      | 692M/1.88G [00:01<00:02, 432MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  40%|███▉      | 744M/1.88G [00:01<00:02, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  42%|████▏     | 797M/1.88G [00:01<00:02, 432MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  45%|████▌     | 849M/1.88G [00:02<00:02, 433MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  48%|████▊     | 902M/1.88G [00:02<00:02, 434MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  51%|█████     | 954M/1.88G [00:02<00:02, 432MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  54%|█████▎    | 1.01G/1.88G [00:02<00:02, 432MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  56%|█████▋    | 1.06G/1.88G [00:02<00:01, 433MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  59%|█████▉    | 1.11G/1.88G [00:02<00:01, 434MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  62%|██████▏   | 1.16G/1.88G [00:02<00:01, 435MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  65%|██████▍   | 1.22G/1.88G [00:02<00:01, 437MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  67%|██████▋   | 1.27G/1.88G [00:02<00:01, 435MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  70%|███████   | 1.32G/1.88G [00:03<00:01, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  73%|███████▎  | 1.37G/1.88G [00:03<00:01, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  76%|███████▌  | 1.43G/1.88G [00:03<00:01, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  79%|███████▊  | 1.48G/1.88G [00:03<00:00, 430MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  81%|████████▏ | 1.53G/1.88G [00:03<00:00, 428MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  84%|████████▍ | 1.58G/1.88G [00:03<00:00, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  87%|████████▋ | 1.64G/1.88G [00:03<00:00, 432MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  90%|████████▉ | 1.69G/1.88G [00:03<00:00, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  93%|█████████▎| 1.74G/1.88G [00:04<00:00, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  95%|█████████▌| 1.79G/1.88G [00:04<00:00, 433MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";:  98%|█████████▊| 1.85G/1.88G [00:04<00:00, 430MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00005-of-00008.bin\";: 100%|██████████| 1.88G/1.88G [00:04<00:00, 427MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  62%|██████▎   | 5/8 [00:24<00:14,  4.86s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:   0%|          | 0.00/1.88G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:   2%|▏         | 31.5M/1.88G [00:00<00:06, 287MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:   4%|▍         | 83.9M/1.88G [00:00<00:04, 377MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:   7%|▋         | 136M/1.88G [00:00<00:04, 406MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  10%|█         | 189M/1.88G [00:00<00:04, 422MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  13%|█▎        | 241M/1.88G [00:00<00:03, 430MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  16%|█▌        | 294M/1.88G [00:00<00:03, 439MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  18%|█▊        | 346M/1.88G [00:00<00:03, 440MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  21%|██        | 398M/1.88G [00:00<00:03, 443MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  24%|██▍       | 451M/1.88G [00:01<00:03, 446MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  27%|██▋       | 503M/1.88G [00:01<00:03, 448MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  30%|██▉       | 556M/1.88G [00:01<00:02, 452MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  32%|███▏      | 608M/1.88G [00:01<00:02, 455MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  35%|███▌      | 661M/1.88G [00:01<00:02, 455MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  38%|███▊      | 713M/1.88G [00:01<00:02, 452MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  41%|████      | 765M/1.88G [00:01<00:02, 447MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  44%|████▎     | 818M/1.88G [00:01<00:02, 442MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  46%|████▋     | 870M/1.88G [00:01<00:02, 443MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  49%|████▉     | 923M/1.88G [00:02<00:02, 445MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  52%|█████▏    | 975M/1.88G [00:02<00:02, 439MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  55%|█████▍    | 1.03G/1.88G [00:02<00:01, 430MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  57%|█████▋    | 1.08G/1.88G [00:02<00:01, 427MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  60%|██████    | 1.13G/1.88G [00:02<00:01, 426MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  63%|██████▎   | 1.18G/1.88G [00:02<00:01, 425MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  66%|██████▌   | 1.24G/1.88G [00:02<00:01, 424MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  69%|██████▊   | 1.29G/1.88G [00:02<00:01, 423MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  71%|███████▏  | 1.34G/1.88G [00:03<00:01, 425MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  74%|███████▍  | 1.39G/1.88G [00:03<00:01, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  77%|███████▋  | 1.45G/1.88G [00:03<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  79%|███████▉  | 1.49G/1.88G [00:03<00:00, 413MB/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  81%|████████▏ | 1.53G/1.88G [00:03<00:00, 404MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  84%|████████▍ | 1.58G/1.88G [00:03<00:00, 415MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  87%|████████▋ | 1.64G/1.88G [00:03<00:00, 423MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  90%|████████▉ | 1.69G/1.88G [00:03<00:00, 422MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  93%|█████████▎| 1.74G/1.88G [00:04<00:00, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  95%|█████████▌| 1.79G/1.88G [00:04<00:00, 436MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";:  98%|█████████▊| 1.85G/1.88G [00:04<00:00, 440MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00006-of-00008.bin\";: 100%|██████████| 1.88G/1.88G [00:04<00:00, 432MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  75%|███████▌  | 6/8 [00:29<00:09,  4.75s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:   0%|          | 0.00/1.07G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:   3%|▎         | 31.5M/1.07G [00:00<00:03, 274MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:   7%|▋         | 73.4M/1.07G [00:00<00:02, 350MB/s]\u001b[0m\n",
      "\u001b[34m#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  11%|█         | 115M/1.07G [00:00<00:02, 380MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  15%|█▍        | 157M/1.07G [00:00<00:02, 394MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  19%|█▊        | 199M/1.07G [00:00<00:02, 400MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  23%|██▎       | 252M/1.07G [00:00<00:02, 409MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  28%|██▊       | 304M/1.07G [00:00<00:01, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  32%|███▏      | 346M/1.07G [00:00<00:01, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  36%|███▌      | 388M/1.07G [00:00<00:01, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  40%|████      | 430M/1.07G [00:01<00:01, 411MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  44%|████▍     | 472M/1.07G [00:01<00:01, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  48%|████▊     | 514M/1.07G [00:01<00:01, 411MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  52%|█████▏    | 556M/1.07G [00:01<00:01, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  56%|█████▌    | 598M/1.07G [00:01<00:01, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  60%|█████▉    | 640M/1.07G [00:01<00:01, 410MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  64%|██████▍   | 692M/1.07G [00:01<00:00, 414MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  68%|██████▊   | 734M/1.07G [00:01<00:00, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  72%|███████▏  | 776M/1.07G [00:01<00:00, 407MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  76%|███████▌  | 818M/1.07G [00:02<00:00, 405MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  80%|████████  | 860M/1.07G [00:02<00:00, 406MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  85%|████████▍ | 912M/1.07G [00:02<00:00, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  90%|████████▉ | 965M/1.07G [00:02<00:00, 419MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";:  95%|█████████▍| 1.02G/1.07G [00:02<00:00, 424MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";: 100%|█████████▉| 1.07G/1.07G [00:02<00:00, 425MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00007-of-00008.bin\";: 100%|██████████| 1.07G/1.07G [00:02<00:00, 408MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards:  88%|████████▊ | 7/8 [00:32<00:04,  4.23s/it]\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:   0%|          | 0.00/1.07G [00:00<?, ?B/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:   3%|▎         | 31.5M/1.07G [00:00<00:03, 285MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:   8%|▊         | 83.9M/1.07G [00:00<00:02, 379MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  13%|█▎        | 136M/1.07G [00:00<00:02, 403MB/s] #033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  18%|█▊        | 189M/1.07G [00:00<00:02, 413MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  23%|██▎       | 241M/1.07G [00:00<00:01, 419MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  27%|██▋       | 294M/1.07G [00:00<00:01, 421MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  32%|███▏      | 346M/1.07G [00:00<00:01, 425MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  37%|███▋      | 398M/1.07G [00:00<00:01, 428MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  42%|████▏     | 451M/1.07G [00:01<00:01, 431MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  47%|████▋     | 503M/1.07G [00:01<00:01, 432MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  52%|█████▏    | 556M/1.07G [00:01<00:01, 434MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  57%|█████▋    | 608M/1.07G [00:01<00:01, 433MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  62%|██████▏   | 661M/1.07G [00:01<00:00, 429MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  67%|██████▋   | 713M/1.07G [00:01<00:00, 424MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  72%|███████▏  | 765M/1.07G [00:01<00:00, 427MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  76%|███████▋  | 818M/1.07G [00:01<00:00, 421MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  81%|████████▏ | 870M/1.07G [00:02<00:00, 416MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  86%|████████▋ | 923M/1.07G [00:02<00:00, 422MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  91%|█████████ | 975M/1.07G [00:02<00:00, 424MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";:  96%|█████████▌| 1.03G/1.07G [00:02<00:00, 424MB/s]#033[A\u001b[0m\n",
      "\u001b[34mDownloading (…)00008-of-00008.bin\";: 100%|██████████| 1.07G/1.07G [00:02<00:00, 421MB/s]\u001b[0m\n",
      "\u001b[34mDownloading shards: 100%|██████████| 8/8 [00:34<00:00,  3.75s/it]\u001b[0m\n",
      "\u001b[34mDownloading shards: 100%|██████████| 8/8 [00:34<00:00,  4.37s/it]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:   0%|          | 0/8 [00:00<?, ?it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  12%|█▎        | 1/8 [00:00<00:05,  1.21it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  25%|██▌       | 2/8 [00:01<00:05,  1.15it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  38%|███▊      | 3/8 [00:02<00:04,  1.10it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  50%|█████     | 4/8 [00:03<00:03,  1.11it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  62%|██████▎   | 5/8 [00:04<00:02,  1.12it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  75%|███████▌  | 6/8 [00:05<00:01,  1.12it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards:  88%|████████▊ | 7/8 [00:05<00:00,  1.28it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards: 100%|██████████| 8/8 [00:06<00:00,  1.43it/s]\u001b[0m\n",
      "\u001b[34mLoading checkpoint shards: 100%|██████████| 8/8 [00:06<00:00,  1.25it/s]\u001b[0m\n",
      "\u001b[34mlen(dataset)=52002\u001b[0m\n",
      "\u001b[34mYou are adding a <class 'transformers.integrations.TensorBoardCallback'> to the callbacks of this Trainer, but there is already one. The currentlist of callbacks is\u001b[0m\n",
      "\u001b[34m:DefaultFlowCallback\u001b[0m\n",
      "\u001b[34mTensorBoardCallback\u001b[0m\n",
      "\u001b[34mYou are adding a <class 'transformers.integrations.TensorBoardCallback'> to the callbacks of this Trainer, but there is already one. The currentlist of callbacks is\u001b[0m\n",
      "\u001b[34m:DefaultFlowCallback\u001b[0m\n",
      "\u001b[34mTensorBoardCallback\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/transformers/optimization.py:391: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning\n",
      "  warnings.warn(\u001b[0m\n",
      "\u001b[34m0%|          | 0/1000 [00:00<?, ?it/s]\u001b[0m\n",
      "\u001b[34m[2023-06-20 09:33:01.440: W smdistributed/modelparallel/torch/nn/predefined_hooks.py:78] Found unsupported HuggingFace version 4.27.1 for automated tensor parallelism. HuggingFace modules will not be automatically distributed. You can use smp.tp_register_with_module API to register desired modules for tensor parallelism, or directly instantiate an smp.nn.DistributedModule. Supported HuggingFace transformers versions for automated tensor parallelism: ['4.17.0', '4.20.1', '4.21.0']\u001b[0m\n",
      "\u001b[34mINFO:root:Using NamedTuple = typing._NamedTuple instead.\u001b[0m\n",
      "\u001b[34m[2023-06-20 09:33:01.467 algo-1:249 INFO utils.py:28] RULE_JOB_STOP_SIGNAL_FILENAME: None\u001b[0m\n",
      "\u001b[34m[2023-06-20 09:33:01.487 algo-1:249 INFO profiler_config_parser.py:111] Unable to find config at /opt/ml/input/config/profilerconfig.json. Profiler is disabled.\u001b[0m\n",
      "\u001b[34m/opt/conda/lib/python3.9/site-packages/bitsandbytes/autograd/_functions.py:298: UserWarning: MatMul8bitLt: inputs will be cast from torch.float32 to float16 during quantization\n",
      "  warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\u001b[0m\n",
      "\u001b[34m0%|          | 1/1000 [00:03<54:14,  3.26s/it]\u001b[0m\n",
      "\u001b[34m0%|          | 2/1000 [00:04<38:31,  2.32s/it]\u001b[0m\n",
      "\u001b[34m0%|          | 3/1000 [00:06<30:26,  1.83s/it]\u001b[0m\n",
      "\u001b[34m0%|          | 4/1000 [00:07<26:35,  1.60s/it]\u001b[0m\n",
      "\u001b[34m0%|          | 5/1000 [00:09<29:43,  1.79s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 6/1000 [00:11<28:40,  1.73s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 7/1000 [00:12<28:02,  1.69s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 8/1000 [00:14<30:20,  1.84s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 9/1000 [00:16<27:09,  1.64s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 10/1000 [00:17<26:00,  1.58s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 11/1000 [00:19<28:44,  1.74s/it]\u001b[0m\n",
      "\u001b[34m1%|          | 12/1000 [00:21<27:12,  1.65s/it]\u001b[0m\n",
      "\u001b[34m1%|▏         | 13/1000 [00:22<26:01,  1.58s/it]\u001b[0m\n",
      "\u001b[34m1%|▏         | 14/1000 [00:23<24:02,  1.46s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 15/1000 [00:25<24:38,  1.50s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 16/1000 [00:27<27:43,  1.69s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 17/1000 [00:28<26:15,  1.60s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 18/1000 [00:30<27:22,  1.67s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 19/1000 [00:32<26:09,  1.60s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 20/1000 [00:33<25:26,  1.56s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 21/1000 [00:35<24:43,  1.52s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 22/1000 [00:37<27:42,  1.70s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 23/1000 [00:38<26:14,  1.61s/it]\u001b[0m\n",
      "\u001b[34m2%|▏         | 24/1000 [00:40<25:35,  1.57s/it]\u001b[0m\n",
      "\u001b[34m2%|▎         | 25/1000 [00:41<25:52,  1.59s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 26/1000 [00:43<25:44,  1.59s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 27/1000 [00:45<28:23,  1.75s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 28/1000 [00:47<29:35,  1.83s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 29/1000 [00:48<27:46,  1.72s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 30/1000 [00:50<28:54,  1.79s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 31/1000 [00:52<27:25,  1.70s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 32/1000 [00:53<26:22,  1.63s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 33/1000 [00:55<27:21,  1.70s/it]\u001b[0m\n",
      "\u001b[34m3%|▎         | 34/1000 [00:57<26:14,  1.63s/it]\u001b[0m\n",
      "\u001b[34m4%|▎         | 35/1000 [00:58<26:03,  1.62s/it]\u001b[0m\n",
      "\u001b[34m4%|▎         | 36/1000 [00:59<24:22,  1.52s/it]\u001b[0m\n",
      "\u001b[34m4%|▎         | 37/1000 [01:01<22:55,  1.43s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 38/1000 [01:02<21:02,  1.31s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 39/1000 [01:04<23:30,  1.47s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 40/1000 [01:06<26:31,  1.66s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 41/1000 [01:07<26:07,  1.63s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 42/1000 [01:09<28:29,  1.78s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 43/1000 [01:11<26:53,  1.69s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 44/1000 [01:13<27:39,  1.74s/it]\u001b[0m\n",
      "\u001b[34m4%|▍         | 45/1000 [01:15<29:29,  1.85s/it]\u001b[0m\n",
      "\u001b[34m5%|▍         | 46/1000 [01:16<27:38,  1.74s/it]\u001b[0m\n",
      "\u001b[34m5%|▍         | 47/1000 [01:18<26:09,  1.65s/it]\u001b[0m\n",
      "\u001b[34m5%|▍         | 48/1000 [01:19<25:21,  1.60s/it]\u001b[0m\n",
      "\u001b[34m5%|▍         | 49/1000 [01:21<26:09,  1.65s/it]\u001b[0m\n",
      "\u001b[34m5%|▌         | 50/1000 [01:22<24:57,  1.58s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.9953, 'learning_rate': 9.53e-05, 'epoch': 0.01}\u001b[0m\n",
      "\u001b[34m5%|▌         | 50/1000 [01:22<24:57,  1.58s/it]\u001b[0m\n",
      "\u001b[34m5%|▌         | 51/1000 [01:24<25:24,  1.61s/it]\u001b[0m\n",
      "\u001b[34m5%|▌         | 52/1000 [01:25<24:25,  1.55s/it]\u001b[0m\n",
      "\u001b[34m5%|▌         | 53/1000 [01:28<27:07,  1.72s/it]\u001b[0m\n",
      "\u001b[34m5%|▌         | 54/1000 [01:30<28:58,  1.84s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 55/1000 [01:32<29:42,  1.89s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 56/1000 [01:34<30:47,  1.96s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 57/1000 [01:35<27:31,  1.75s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 58/1000 [01:37<26:07,  1.66s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 59/1000 [01:38<25:47,  1.64s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 60/1000 [01:40<27:09,  1.73s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 61/1000 [01:42<28:03,  1.79s/it]\u001b[0m\n",
      "\u001b[34m6%|▌         | 62/1000 [01:44<27:09,  1.74s/it]\u001b[0m\n",
      "\u001b[34m6%|▋         | 63/1000 [01:45<23:21,  1.50s/it]\u001b[0m\n",
      "\u001b[34m6%|▋         | 64/1000 [01:47<26:19,  1.69s/it]\u001b[0m\n",
      "\u001b[34m6%|▋         | 65/1000 [01:49<27:05,  1.74s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 66/1000 [01:50<26:27,  1.70s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 67/1000 [01:52<28:27,  1.83s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 68/1000 [01:54<29:46,  1.92s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 69/1000 [01:56<29:09,  1.88s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 70/1000 [01:58<27:43,  1.79s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 71/1000 [02:00<28:23,  1.83s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 72/1000 [02:02<29:41,  1.92s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 73/1000 [02:04<29:51,  1.93s/it]\u001b[0m\n",
      "\u001b[34m7%|▋         | 74/1000 [02:06<30:43,  1.99s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 75/1000 [02:07<27:04,  1.76s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 76/1000 [02:09<28:43,  1.87s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 77/1000 [02:10<25:09,  1.64s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 78/1000 [02:12<24:22,  1.59s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 79/1000 [02:13<23:40,  1.54s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 80/1000 [02:15<23:11,  1.51s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 81/1000 [02:16<21:01,  1.37s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 82/1000 [02:17<22:10,  1.45s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 83/1000 [02:19<25:14,  1.65s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 84/1000 [02:22<27:22,  1.79s/it]\u001b[0m\n",
      "\u001b[34m8%|▊         | 85/1000 [02:23<26:35,  1.74s/it]\u001b[0m\n",
      "\u001b[34m9%|▊         | 86/1000 [02:25<27:05,  1.78s/it]\u001b[0m\n",
      "\u001b[34m9%|▊         | 87/1000 [02:27<27:47,  1.83s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 88/1000 [02:29<29:05,  1.91s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 89/1000 [02:31<26:51,  1.77s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 90/1000 [02:32<25:33,  1.69s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 91/1000 [02:34<27:33,  1.82s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 92/1000 [02:36<28:53,  1.91s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 93/1000 [02:38<29:51,  1.98s/it]\u001b[0m\n",
      "\u001b[34m9%|▉         | 94/1000 [02:40<27:34,  1.83s/it]\u001b[0m\n",
      "\u001b[34m10%|▉         | 95/1000 [02:41<25:55,  1.72s/it]\u001b[0m\n",
      "\u001b[34m10%|▉         | 96/1000 [02:43<26:09,  1.74s/it]\u001b[0m\n",
      "\u001b[34m10%|▉         | 97/1000 [02:45<27:50,  1.85s/it]\u001b[0m\n",
      "\u001b[34m10%|▉         | 98/1000 [02:47<29:02,  1.93s/it]\u001b[0m\n",
      "\u001b[34m10%|▉         | 99/1000 [02:49<26:02,  1.73s/it]\u001b[0m\n",
      "\u001b[34m10%|█         | 100/1000 [02:50<23:53,  1.59s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6312, 'learning_rate': 9.030000000000001e-05, 'epoch': 0.01}\u001b[0m\n",
      "\u001b[34m10%|█         | 100/1000 [02:50<23:53,  1.59s/it]\u001b[0m\n",
      "\u001b[34m10%|█         | 101/1000 [02:51<22:19,  1.49s/it]\u001b[0m\n",
      "\u001b[34m10%|█         | 102/1000 [02:53<22:15,  1.49s/it]\u001b[0m\n",
      "\u001b[34m10%|█         | 103/1000 [02:54<21:09,  1.41s/it]\u001b[0m\n",
      "\u001b[34m10%|█         | 104/1000 [02:56<22:13,  1.49s/it]\u001b[0m\n",
      "\u001b[34m10%|█         | 105/1000 [02:57<23:32,  1.58s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 106/1000 [02:59<25:55,  1.74s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 107/1000 [03:01<24:40,  1.66s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 108/1000 [03:02<23:37,  1.59s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 109/1000 [03:05<25:58,  1.75s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 110/1000 [03:06<25:08,  1.69s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 111/1000 [03:08<26:25,  1.78s/it]\u001b[0m\n",
      "\u001b[34m11%|█         | 112/1000 [03:10<26:30,  1.79s/it]\u001b[0m\n",
      "\u001b[34m11%|█▏        | 113/1000 [03:12<27:32,  1.86s/it]\u001b[0m\n",
      "\u001b[34m11%|█▏        | 114/1000 [03:13<25:47,  1.75s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 115/1000 [03:15<23:34,  1.60s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 116/1000 [03:16<24:24,  1.66s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 117/1000 [03:18<25:55,  1.76s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 118/1000 [03:20<26:06,  1.78s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 119/1000 [03:22<27:40,  1.88s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 120/1000 [03:24<25:36,  1.75s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 121/1000 [03:25<22:40,  1.55s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 122/1000 [03:27<22:55,  1.57s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 123/1000 [03:29<25:18,  1.73s/it]\u001b[0m\n",
      "\u001b[34m12%|█▏        | 124/1000 [03:30<25:01,  1.71s/it]\u001b[0m\n",
      "\u001b[34m12%|█▎        | 125/1000 [03:32<25:57,  1.78s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 126/1000 [03:34<27:22,  1.88s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 127/1000 [03:36<24:32,  1.69s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 128/1000 [03:38<26:23,  1.82s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 129/1000 [03:40<27:44,  1.91s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 130/1000 [03:42<28:38,  1.98s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 131/1000 [03:44<29:18,  2.02s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 132/1000 [03:46<29:13,  2.02s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 133/1000 [03:48<29:32,  2.04s/it]\u001b[0m\n",
      "\u001b[34m13%|█▎        | 134/1000 [03:50<29:14,  2.03s/it]\u001b[0m\n",
      "\u001b[34m14%|█▎        | 135/1000 [03:52<29:35,  2.05s/it]\u001b[0m\n",
      "\u001b[34m14%|█▎        | 136/1000 [03:54<27:47,  1.93s/it]\u001b[0m\n",
      "\u001b[34m14%|█▎        | 137/1000 [03:55<25:31,  1.77s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 138/1000 [03:57<23:53,  1.66s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 139/1000 [03:59<24:40,  1.72s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 140/1000 [04:00<23:36,  1.65s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 141/1000 [04:02<24:20,  1.70s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 142/1000 [04:04<26:09,  1.83s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 143/1000 [04:06<24:40,  1.73s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 144/1000 [04:07<24:55,  1.75s/it]\u001b[0m\n",
      "\u001b[34m14%|█▍        | 145/1000 [04:09<23:37,  1.66s/it]\u001b[0m\n",
      "\u001b[34m15%|█▍        | 146/1000 [04:10<22:26,  1.58s/it]\u001b[0m\n",
      "\u001b[34m15%|█▍        | 147/1000 [04:11<21:00,  1.48s/it]\u001b[0m\n",
      "\u001b[34m15%|█▍        | 148/1000 [04:13<21:21,  1.50s/it]\u001b[0m\n",
      "\u001b[34m15%|█▍        | 149/1000 [04:14<20:11,  1.42s/it]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 150/1000 [04:16<21:07,  1.49s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6817, 'learning_rate': 8.53e-05, 'epoch': 0.02}\u001b[0m\n",
      "\u001b[34m15%|█▌        | 150/1000 [04:16<21:07,  1.49s/it]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 151/1000 [04:17<20:47,  1.47s/it]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 152/1000 [04:19<20:37,  1.46s/it]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 153/1000 [04:20<21:24,  1.52s/it]\u001b[0m\n",
      "\u001b[34m15%|█▌        | 154/1000 [04:22<21:06,  1.50s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 155/1000 [04:24<22:14,  1.58s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 156/1000 [04:26<23:49,  1.69s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 157/1000 [04:27<24:07,  1.72s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 158/1000 [04:29<24:41,  1.76s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 159/1000 [04:31<26:15,  1.87s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 160/1000 [04:33<24:30,  1.75s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 161/1000 [04:35<26:02,  1.86s/it]\u001b[0m\n",
      "\u001b[34m16%|█▌        | 162/1000 [04:37<25:00,  1.79s/it]\u001b[0m\n",
      "\u001b[34m16%|█▋        | 163/1000 [04:38<22:39,  1.62s/it]\u001b[0m\n",
      "\u001b[34m16%|█▋        | 164/1000 [04:39<22:04,  1.58s/it]\u001b[0m\n",
      "\u001b[34m16%|█▋        | 165/1000 [04:41<22:20,  1.61s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 166/1000 [04:42<21:47,  1.57s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 167/1000 [04:44<20:33,  1.48s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 168/1000 [04:46<22:06,  1.59s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 169/1000 [04:47<20:28,  1.48s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 170/1000 [04:49<21:58,  1.59s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 171/1000 [04:50<21:29,  1.56s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 172/1000 [04:51<20:57,  1.52s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 173/1000 [04:53<22:13,  1.61s/it]\u001b[0m\n",
      "\u001b[34m17%|█▋        | 174/1000 [04:55<24:12,  1.76s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 175/1000 [04:58<25:39,  1.87s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 176/1000 [05:00<26:41,  1.94s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 177/1000 [05:01<25:25,  1.85s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 178/1000 [05:03<26:30,  1.93s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 179/1000 [05:05<24:22,  1.78s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 180/1000 [05:07<25:45,  1.88s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 181/1000 [05:08<23:10,  1.70s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 182/1000 [05:09<20:44,  1.52s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 183/1000 [05:11<20:21,  1.49s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 184/1000 [05:13<22:21,  1.64s/it]\u001b[0m\n",
      "\u001b[34m18%|█▊        | 185/1000 [05:14<21:16,  1.57s/it]\u001b[0m\n",
      "\u001b[34m19%|█▊        | 186/1000 [05:16<22:23,  1.65s/it]\u001b[0m\n",
      "\u001b[34m19%|█▊        | 187/1000 [05:17<20:02,  1.48s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 188/1000 [05:19<21:29,  1.59s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 189/1000 [05:21<22:35,  1.67s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 190/1000 [05:22<21:44,  1.61s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 191/1000 [05:24<22:21,  1.66s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 192/1000 [05:26<24:10,  1.80s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 193/1000 [05:28<22:32,  1.68s/it]\u001b[0m\n",
      "\u001b[34m19%|█▉        | 194/1000 [05:29<20:52,  1.55s/it]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 195/1000 [05:30<20:33,  1.53s/it]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 196/1000 [05:32<20:11,  1.51s/it]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 197/1000 [05:34<22:37,  1.69s/it]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 198/1000 [05:36<24:19,  1.82s/it]\u001b[0m\n",
      "\u001b[34m20%|█▉        | 199/1000 [05:38<25:30,  1.91s/it]\u001b[0m\n",
      "\u001b[34m20%|██        | 200/1000 [05:40<26:20,  1.98s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.605, 'learning_rate': 8.030000000000001e-05, 'epoch': 0.02}\u001b[0m\n",
      "\u001b[34m20%|██        | 200/1000 [05:40<26:20,  1.98s/it]\u001b[0m\n",
      "\u001b[34m20%|██        | 201/1000 [05:42<26:58,  2.03s/it]\u001b[0m\n",
      "\u001b[34m20%|██        | 202/1000 [05:44<23:52,  1.80s/it]\u001b[0m\n",
      "\u001b[34m20%|██        | 203/1000 [05:45<23:04,  1.74s/it]\u001b[0m\n",
      "\u001b[34m20%|██        | 204/1000 [05:47<22:24,  1.69s/it]\u001b[0m\n",
      "\u001b[34m20%|██        | 205/1000 [05:48<19:55,  1.50s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 206/1000 [05:50<21:51,  1.65s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 207/1000 [05:51<21:06,  1.60s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 208/1000 [05:53<23:08,  1.75s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 209/1000 [05:55<22:24,  1.70s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 210/1000 [05:57<23:55,  1.82s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 211/1000 [05:59<24:34,  1.87s/it]\u001b[0m\n",
      "\u001b[34m21%|██        | 212/1000 [06:01<22:55,  1.75s/it]\u001b[0m\n",
      "\u001b[34m21%|██▏       | 213/1000 [06:02<20:09,  1.54s/it]\u001b[0m\n",
      "\u001b[34m21%|██▏       | 214/1000 [06:03<18:58,  1.45s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 215/1000 [06:05<21:36,  1.65s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 216/1000 [06:07<23:27,  1.80s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 217/1000 [06:09<21:51,  1.67s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 218/1000 [06:10<21:44,  1.67s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 219/1000 [06:12<23:31,  1.81s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 220/1000 [06:14<24:19,  1.87s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 221/1000 [06:16<24:04,  1.85s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 222/1000 [06:18<25:11,  1.94s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 223/1000 [06:19<21:55,  1.69s/it]\u001b[0m\n",
      "\u001b[34m22%|██▏       | 224/1000 [06:21<23:00,  1.78s/it]\u001b[0m\n",
      "\u001b[34m22%|██▎       | 225/1000 [06:23<22:16,  1.72s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 226/1000 [06:24<20:16,  1.57s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 227/1000 [06:26<22:21,  1.73s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 228/1000 [06:28<23:49,  1.85s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 229/1000 [06:31<24:48,  1.93s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 230/1000 [06:32<23:29,  1.83s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 231/1000 [06:34<23:14,  1.81s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 232/1000 [06:36<22:26,  1.75s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 233/1000 [06:37<20:14,  1.58s/it]\u001b[0m\n",
      "\u001b[34m23%|██▎       | 234/1000 [06:38<20:25,  1.60s/it]\u001b[0m\n",
      "\u001b[34m24%|██▎       | 235/1000 [06:40<21:04,  1.65s/it]\u001b[0m\n",
      "\u001b[34m24%|██▎       | 236/1000 [06:42<22:53,  1.80s/it]\u001b[0m\n",
      "\u001b[34m24%|██▎       | 237/1000 [06:44<21:24,  1.68s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 238/1000 [06:45<21:23,  1.68s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 239/1000 [06:47<19:37,  1.55s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 240/1000 [06:48<19:05,  1.51s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 241/1000 [06:49<18:39,  1.47s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 242/1000 [06:50<17:02,  1.35s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 243/1000 [06:52<18:55,  1.50s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 244/1000 [06:54<19:59,  1.59s/it]\u001b[0m\n",
      "\u001b[34m24%|██▍       | 245/1000 [06:56<21:22,  1.70s/it]\u001b[0m\n",
      "\u001b[34m25%|██▍       | 246/1000 [06:58<21:56,  1.75s/it]\u001b[0m\n",
      "\u001b[34m25%|██▍       | 247/1000 [07:00<23:23,  1.86s/it]\u001b[0m\n",
      "\u001b[34m25%|██▍       | 248/1000 [07:02<24:20,  1.94s/it]\u001b[0m\n",
      "\u001b[34m25%|██▍       | 249/1000 [07:04<22:56,  1.83s/it]\u001b[0m\n",
      "\u001b[34m25%|██▌       | 250/1000 [07:05<20:35,  1.65s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6289, 'learning_rate': 7.53e-05, 'epoch': 0.03}\u001b[0m\n",
      "\u001b[34m25%|██▌       | 250/1000 [07:05<20:35,  1.65s/it]\u001b[0m\n",
      "\u001b[34m25%|██▌       | 251/1000 [07:06<19:42,  1.58s/it]\u001b[0m\n",
      "\u001b[34m25%|██▌       | 252/1000 [07:07<17:27,  1.40s/it]\u001b[0m\n",
      "\u001b[34m25%|██▌       | 253/1000 [07:09<17:32,  1.41s/it]\u001b[0m\n",
      "\u001b[34m25%|██▌       | 254/1000 [07:10<16:55,  1.36s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 255/1000 [07:12<19:44,  1.59s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 256/1000 [07:14<21:40,  1.75s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 257/1000 [07:16<21:45,  1.76s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 258/1000 [07:18<21:14,  1.72s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 259/1000 [07:19<20:19,  1.65s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 260/1000 [07:21<21:48,  1.77s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 261/1000 [07:23<22:06,  1.80s/it]\u001b[0m\n",
      "\u001b[34m26%|██▌       | 262/1000 [07:25<22:07,  1.80s/it]\u001b[0m\n",
      "\u001b[34m26%|██▋       | 263/1000 [07:27<21:22,  1.74s/it]\u001b[0m\n",
      "\u001b[34m26%|██▋       | 264/1000 [07:28<21:35,  1.76s/it]\u001b[0m\n",
      "\u001b[34m26%|██▋       | 265/1000 [07:30<21:11,  1.73s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 266/1000 [07:32<22:37,  1.85s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 267/1000 [07:33<19:38,  1.61s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 268/1000 [07:34<18:24,  1.51s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 269/1000 [07:36<19:33,  1.60s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 270/1000 [07:38<18:16,  1.50s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 271/1000 [07:39<18:48,  1.55s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 272/1000 [07:41<19:34,  1.61s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 273/1000 [07:42<17:43,  1.46s/it]\u001b[0m\n",
      "\u001b[34m27%|██▋       | 274/1000 [07:44<20:05,  1.66s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 275/1000 [07:46<21:45,  1.80s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 276/1000 [07:47<18:28,  1.53s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 277/1000 [07:48<17:23,  1.44s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 278/1000 [07:51<19:49,  1.65s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 279/1000 [07:52<19:05,  1.59s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 280/1000 [07:53<17:43,  1.48s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 281/1000 [07:55<18:15,  1.52s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 282/1000 [07:57<20:23,  1.70s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 283/1000 [07:59<20:39,  1.73s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 284/1000 [08:00<20:05,  1.68s/it]\u001b[0m\n",
      "\u001b[34m28%|██▊       | 285/1000 [08:02<19:55,  1.67s/it]\u001b[0m\n",
      "\u001b[34m29%|██▊       | 286/1000 [08:04<21:33,  1.81s/it]\u001b[0m\n",
      "\u001b[34m29%|██▊       | 287/1000 [08:06<20:40,  1.74s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 288/1000 [08:08<21:34,  1.82s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 289/1000 [08:10<22:38,  1.91s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 290/1000 [08:12<23:20,  1.97s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 291/1000 [08:14<21:51,  1.85s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 292/1000 [08:15<20:31,  1.74s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 293/1000 [08:17<20:45,  1.76s/it]\u001b[0m\n",
      "\u001b[34m29%|██▉       | 294/1000 [08:19<21:59,  1.87s/it]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 295/1000 [08:21<21:08,  1.80s/it]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 296/1000 [08:22<21:14,  1.81s/it]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 297/1000 [08:24<19:59,  1.71s/it]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 298/1000 [08:26<21:25,  1.83s/it]\u001b[0m\n",
      "\u001b[34m30%|██▉       | 299/1000 [08:28<21:27,  1.84s/it]\u001b[0m\n",
      "\u001b[34m30%|███       | 300/1000 [08:29<19:54,  1.71s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5898, 'learning_rate': 7.03e-05, 'epoch': 0.03}\u001b[0m\n",
      "\u001b[34m30%|███       | 300/1000 [08:29<19:54,  1.71s/it]\u001b[0m\n",
      "\u001b[34m30%|███       | 301/1000 [08:31<18:59,  1.63s/it]\u001b[0m\n",
      "\u001b[34m30%|███       | 302/1000 [08:33<20:41,  1.78s/it]\u001b[0m\n",
      "\u001b[34m30%|███       | 303/1000 [08:35<20:43,  1.78s/it]\u001b[0m\n",
      "\u001b[34m30%|███       | 304/1000 [08:36<18:11,  1.57s/it]\u001b[0m\n",
      "\u001b[34m30%|███       | 305/1000 [08:38<19:08,  1.65s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 306/1000 [08:39<18:17,  1.58s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 307/1000 [08:40<17:37,  1.53s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 308/1000 [08:42<18:34,  1.61s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 309/1000 [08:44<20:19,  1.76s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 310/1000 [08:46<21:13,  1.85s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 311/1000 [08:48<21:12,  1.85s/it]\u001b[0m\n",
      "\u001b[34m31%|███       | 312/1000 [08:50<20:25,  1.78s/it]\u001b[0m\n",
      "\u001b[34m31%|███▏      | 313/1000 [08:52<21:36,  1.89s/it]\u001b[0m\n",
      "\u001b[34m31%|███▏      | 314/1000 [08:54<22:23,  1.96s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 315/1000 [08:56<22:57,  2.01s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 316/1000 [08:58<21:25,  1.88s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 317/1000 [09:00<22:13,  1.95s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 318/1000 [09:02<22:16,  1.96s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 319/1000 [09:03<19:45,  1.74s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 320/1000 [09:05<19:51,  1.75s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 321/1000 [09:06<18:07,  1.60s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 322/1000 [09:08<18:42,  1.66s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 323/1000 [09:10<20:16,  1.80s/it]\u001b[0m\n",
      "\u001b[34m32%|███▏      | 324/1000 [09:11<18:24,  1.63s/it]\u001b[0m\n",
      "\u001b[34m32%|███▎      | 325/1000 [09:13<19:42,  1.75s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 326/1000 [09:15<20:56,  1.86s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 327/1000 [09:18<21:47,  1.94s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 328/1000 [09:19<20:51,  1.86s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 329/1000 [09:21<20:08,  1.80s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 330/1000 [09:22<18:45,  1.68s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 331/1000 [09:24<18:33,  1.66s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 332/1000 [09:26<18:23,  1.65s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 333/1000 [09:27<18:46,  1.69s/it]\u001b[0m\n",
      "\u001b[34m33%|███▎      | 334/1000 [09:29<20:08,  1.82s/it]\u001b[0m\n",
      "\u001b[34m34%|███▎      | 335/1000 [09:31<20:41,  1.87s/it]\u001b[0m\n",
      "\u001b[34m34%|███▎      | 336/1000 [09:34<21:31,  1.94s/it]\u001b[0m\n",
      "\u001b[34m34%|███▎      | 337/1000 [09:35<20:58,  1.90s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 338/1000 [09:37<21:15,  1.93s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 339/1000 [09:39<20:19,  1.84s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 340/1000 [09:41<19:26,  1.77s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 341/1000 [09:43<20:33,  1.87s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 342/1000 [09:44<18:59,  1.73s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 343/1000 [09:46<19:15,  1.76s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 344/1000 [09:48<20:22,  1.86s/it]\u001b[0m\n",
      "\u001b[34m34%|███▍      | 345/1000 [09:50<20:18,  1.86s/it]\u001b[0m\n",
      "\u001b[34m35%|███▍      | 346/1000 [09:51<18:20,  1.68s/it]\u001b[0m\n",
      "\u001b[34m35%|███▍      | 347/1000 [09:53<17:26,  1.60s/it]\u001b[0m\n",
      "\u001b[34m35%|███▍      | 348/1000 [09:55<19:08,  1.76s/it]\u001b[0m\n",
      "\u001b[34m35%|███▍      | 349/1000 [09:56<17:17,  1.59s/it]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 350/1000 [09:58<19:01,  1.76s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6355, 'learning_rate': 6.53e-05, 'epoch': 0.04}\u001b[0m\n",
      "\u001b[34m35%|███▌      | 350/1000 [09:58<19:01,  1.76s/it]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 351/1000 [10:00<19:11,  1.77s/it]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 352/1000 [10:02<20:17,  1.88s/it]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 353/1000 [10:04<21:04,  1.95s/it]\u001b[0m\n",
      "\u001b[34m35%|███▌      | 354/1000 [10:06<21:33,  2.00s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 355/1000 [10:07<19:01,  1.77s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 356/1000 [10:10<20:07,  1.88s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 357/1000 [10:12<20:55,  1.95s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 358/1000 [10:14<20:36,  1.93s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 359/1000 [10:16<21:11,  1.98s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 360/1000 [10:17<20:35,  1.93s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 361/1000 [10:19<19:06,  1.79s/it]\u001b[0m\n",
      "\u001b[34m36%|███▌      | 362/1000 [10:21<18:32,  1.74s/it]\u001b[0m\n",
      "\u001b[34m36%|███▋      | 363/1000 [10:23<19:41,  1.86s/it]\u001b[0m\n",
      "\u001b[34m36%|███▋      | 364/1000 [10:24<18:50,  1.78s/it]\u001b[0m\n",
      "\u001b[34m36%|███▋      | 365/1000 [10:26<17:02,  1.61s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 366/1000 [10:27<17:39,  1.67s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 367/1000 [10:29<18:01,  1.71s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 368/1000 [10:31<18:55,  1.80s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 369/1000 [10:33<19:57,  1.90s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 370/1000 [10:35<20:36,  1.96s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 371/1000 [10:37<19:34,  1.87s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 372/1000 [10:38<18:05,  1.73s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 373/1000 [10:40<18:45,  1.79s/it]\u001b[0m\n",
      "\u001b[34m37%|███▋      | 374/1000 [10:42<18:43,  1.80s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 375/1000 [10:44<19:09,  1.84s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 376/1000 [10:45<17:10,  1.65s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 377/1000 [10:47<18:39,  1.80s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 378/1000 [10:50<19:39,  1.90s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 379/1000 [10:51<18:17,  1.77s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 380/1000 [10:53<17:46,  1.72s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 381/1000 [10:55<18:59,  1.84s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 382/1000 [10:56<17:34,  1.71s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 383/1000 [10:58<16:49,  1.64s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 384/1000 [10:59<16:17,  1.59s/it]\u001b[0m\n",
      "\u001b[34m38%|███▊      | 385/1000 [11:01<17:35,  1.72s/it]\u001b[0m\n",
      "\u001b[34m39%|███▊      | 386/1000 [11:03<16:40,  1.63s/it]\u001b[0m\n",
      "\u001b[34m39%|███▊      | 387/1000 [11:05<18:11,  1.78s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 388/1000 [11:07<19:12,  1.88s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 389/1000 [11:08<17:07,  1.68s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 390/1000 [11:09<15:24,  1.52s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 391/1000 [11:11<15:44,  1.55s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 392/1000 [11:12<15:23,  1.52s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 393/1000 [11:14<17:00,  1.68s/it]\u001b[0m\n",
      "\u001b[34m39%|███▉      | 394/1000 [11:16<16:43,  1.66s/it]\u001b[0m\n",
      "\u001b[34m40%|███▉      | 395/1000 [11:17<16:11,  1.61s/it]\u001b[0m\n",
      "\u001b[34m40%|███▉      | 396/1000 [11:19<16:05,  1.60s/it]\u001b[0m\n",
      "\u001b[34m40%|███▉      | 397/1000 [11:20<14:33,  1.45s/it]\u001b[0m\n",
      "\u001b[34m40%|███▉      | 398/1000 [11:22<16:34,  1.65s/it]\u001b[0m\n",
      "\u001b[34m40%|███▉      | 399/1000 [11:23<15:20,  1.53s/it]\u001b[0m\n",
      "\u001b[34m40%|████      | 400/1000 [11:25<15:11,  1.52s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6068, 'learning_rate': 6.03e-05, 'epoch': 0.05}\u001b[0m\n",
      "\u001b[34m40%|████      | 400/1000 [11:25<15:11,  1.52s/it]\u001b[0m\n",
      "\u001b[34m40%|████      | 401/1000 [11:27<15:32,  1.56s/it]\u001b[0m\n",
      "\u001b[34m40%|████      | 402/1000 [11:28<16:20,  1.64s/it]\u001b[0m\n",
      "\u001b[34m40%|████      | 403/1000 [11:30<16:53,  1.70s/it]\u001b[0m\n",
      "\u001b[34m40%|████      | 404/1000 [11:32<17:09,  1.73s/it]\u001b[0m\n",
      "\u001b[34m40%|████      | 405/1000 [11:34<17:22,  1.75s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 406/1000 [11:36<18:12,  1.84s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 407/1000 [11:37<16:51,  1.71s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 408/1000 [11:39<16:06,  1.63s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 409/1000 [11:40<14:54,  1.51s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 410/1000 [11:42<16:40,  1.70s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 411/1000 [11:44<16:00,  1.63s/it]\u001b[0m\n",
      "\u001b[34m41%|████      | 412/1000 [11:45<14:55,  1.52s/it]\u001b[0m\n",
      "\u001b[34m41%|████▏     | 413/1000 [11:47<16:14,  1.66s/it]\u001b[0m\n",
      "\u001b[34m41%|████▏     | 414/1000 [11:49<17:07,  1.75s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 415/1000 [11:51<18:09,  1.86s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 416/1000 [11:53<18:52,  1.94s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 417/1000 [11:55<19:10,  1.97s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 418/1000 [11:57<19:21,  2.00s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 419/1000 [11:59<19:41,  2.03s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 420/1000 [12:01<17:21,  1.80s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 421/1000 [12:02<17:18,  1.79s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 422/1000 [12:04<16:23,  1.70s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 423/1000 [12:05<16:14,  1.69s/it]\u001b[0m\n",
      "\u001b[34m42%|████▏     | 424/1000 [12:07<16:02,  1.67s/it]\u001b[0m\n",
      "\u001b[34m42%|████▎     | 425/1000 [12:08<14:48,  1.54s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 426/1000 [12:10<15:32,  1.62s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 427/1000 [12:12<16:54,  1.77s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 428/1000 [12:14<17:10,  1.80s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 429/1000 [12:16<16:35,  1.74s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 430/1000 [12:17<15:48,  1.66s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 431/1000 [12:19<16:41,  1.76s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 432/1000 [12:20<15:05,  1.59s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 433/1000 [12:22<14:08,  1.50s/it]\u001b[0m\n",
      "\u001b[34m43%|████▎     | 434/1000 [12:23<14:31,  1.54s/it]\u001b[0m\n",
      "\u001b[34m44%|████▎     | 435/1000 [12:25<16:07,  1.71s/it]\u001b[0m\n",
      "\u001b[34m44%|████▎     | 436/1000 [12:28<17:13,  1.83s/it]\u001b[0m\n",
      "\u001b[34m44%|████▎     | 437/1000 [12:30<18:01,  1.92s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 438/1000 [12:31<17:16,  1.84s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 439/1000 [12:33<18:02,  1.93s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 440/1000 [12:35<17:38,  1.89s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 441/1000 [12:36<15:49,  1.70s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 442/1000 [12:39<16:57,  1.82s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 443/1000 [12:40<15:51,  1.71s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 444/1000 [12:42<15:11,  1.64s/it]\u001b[0m\n",
      "\u001b[34m44%|████▍     | 445/1000 [12:44<16:30,  1.78s/it]\u001b[0m\n",
      "\u001b[34m45%|████▍     | 446/1000 [12:46<17:26,  1.89s/it]\u001b[0m\n",
      "\u001b[34m45%|████▍     | 447/1000 [12:47<15:39,  1.70s/it]\u001b[0m\n",
      "\u001b[34m45%|████▍     | 448/1000 [12:49<15:26,  1.68s/it]\u001b[0m\n",
      "\u001b[34m45%|████▍     | 449/1000 [12:50<15:38,  1.70s/it]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 450/1000 [12:52<14:59,  1.64s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6383, 'learning_rate': 5.530000000000001e-05, 'epoch': 0.05}\u001b[0m\n",
      "\u001b[34m45%|████▌     | 450/1000 [12:52<14:59,  1.64s/it]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 451/1000 [12:54<14:59,  1.64s/it]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 452/1000 [12:55<14:54,  1.63s/it]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 453/1000 [12:57<14:59,  1.64s/it]\u001b[0m\n",
      "\u001b[34m45%|████▌     | 454/1000 [12:58<14:26,  1.59s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 455/1000 [13:00<15:53,  1.75s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 456/1000 [13:02<14:58,  1.65s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 457/1000 [13:03<14:19,  1.58s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 458/1000 [13:05<13:48,  1.53s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 459/1000 [13:06<13:33,  1.50s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 460/1000 [13:08<14:42,  1.63s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 461/1000 [13:09<14:06,  1.57s/it]\u001b[0m\n",
      "\u001b[34m46%|████▌     | 462/1000 [13:12<15:32,  1.73s/it]\u001b[0m\n",
      "\u001b[34m46%|████▋     | 463/1000 [13:13<14:46,  1.65s/it]\u001b[0m\n",
      "\u001b[34m46%|████▋     | 464/1000 [13:14<13:39,  1.53s/it]\u001b[0m\n",
      "\u001b[34m46%|████▋     | 465/1000 [13:15<12:22,  1.39s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 466/1000 [13:17<14:17,  1.61s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 467/1000 [13:19<13:14,  1.49s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 468/1000 [13:21<14:54,  1.68s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 469/1000 [13:23<15:16,  1.73s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 470/1000 [13:25<15:59,  1.81s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 471/1000 [13:27<16:48,  1.91s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 472/1000 [13:28<15:58,  1.82s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 473/1000 [13:31<16:44,  1.91s/it]\u001b[0m\n",
      "\u001b[34m47%|████▋     | 474/1000 [13:32<15:28,  1.77s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 475/1000 [13:34<16:22,  1.87s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 476/1000 [13:35<14:47,  1.69s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 477/1000 [13:37<15:52,  1.82s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 478/1000 [13:40<16:37,  1.91s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 479/1000 [13:41<14:54,  1.72s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 480/1000 [13:42<14:40,  1.69s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 481/1000 [13:44<14:05,  1.63s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 482/1000 [13:46<15:19,  1.78s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 483/1000 [13:48<16:04,  1.87s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 484/1000 [13:50<15:17,  1.78s/it]\u001b[0m\n",
      "\u001b[34m48%|████▊     | 485/1000 [13:51<13:55,  1.62s/it]\u001b[0m\n",
      "\u001b[34m49%|████▊     | 486/1000 [13:52<13:25,  1.57s/it]\u001b[0m\n",
      "\u001b[34m49%|████▊     | 487/1000 [13:54<12:33,  1.47s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 488/1000 [13:56<13:33,  1.59s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 489/1000 [13:57<12:21,  1.45s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 490/1000 [13:59<13:39,  1.61s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 491/1000 [14:01<14:58,  1.76s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 492/1000 [14:03<15:23,  1.82s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 493/1000 [14:04<14:44,  1.75s/it]\u001b[0m\n",
      "\u001b[34m49%|████▉     | 494/1000 [14:06<13:31,  1.60s/it]\u001b[0m\n",
      "\u001b[34m50%|████▉     | 495/1000 [14:07<13:26,  1.60s/it]\u001b[0m\n",
      "\u001b[34m50%|████▉     | 496/1000 [14:09<13:02,  1.55s/it]\u001b[0m\n",
      "\u001b[34m50%|████▉     | 497/1000 [14:10<12:16,  1.46s/it]\u001b[0m\n",
      "\u001b[34m50%|████▉     | 498/1000 [14:12<13:55,  1.66s/it]\u001b[0m\n",
      "\u001b[34m50%|████▉     | 499/1000 [14:14<14:36,  1.75s/it]\u001b[0m\n",
      "\u001b[34m50%|█████     | 500/1000 [14:15<12:52,  1.55s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6301, 'learning_rate': 5.03e-05, 'epoch': 0.06}\u001b[0m\n",
      "\u001b[34m50%|█████     | 500/1000 [14:15<12:52,  1.55s/it]\u001b[0m\n",
      "\u001b[34m50%|█████     | 501/1000 [14:17<14:18,  1.72s/it]\u001b[0m\n",
      "\u001b[34m50%|█████     | 502/1000 [14:19<13:56,  1.68s/it]\u001b[0m\n",
      "\u001b[34m50%|█████     | 503/1000 [14:21<15:01,  1.81s/it]\u001b[0m\n",
      "\u001b[34m50%|█████     | 504/1000 [14:23<15:20,  1.86s/it]\u001b[0m\n",
      "\u001b[34m50%|█████     | 505/1000 [14:24<13:46,  1.67s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 506/1000 [14:26<14:47,  1.80s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 507/1000 [14:28<13:59,  1.70s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 508/1000 [14:30<15:00,  1.83s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 509/1000 [14:31<13:58,  1.71s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 510/1000 [14:33<13:11,  1.61s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 511/1000 [14:34<13:10,  1.62s/it]\u001b[0m\n",
      "\u001b[34m51%|█████     | 512/1000 [14:36<14:23,  1.77s/it]\u001b[0m\n",
      "\u001b[34m51%|█████▏    | 513/1000 [14:38<13:37,  1.68s/it]\u001b[0m\n",
      "\u001b[34m51%|█████▏    | 514/1000 [14:40<14:34,  1.80s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 515/1000 [14:41<12:28,  1.54s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 516/1000 [14:43<13:50,  1.72s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 517/1000 [14:44<13:01,  1.62s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 518/1000 [14:46<13:31,  1.68s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 519/1000 [14:48<12:54,  1.61s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 520/1000 [14:49<11:31,  1.44s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 521/1000 [14:50<11:49,  1.48s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 522/1000 [14:52<12:34,  1.58s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 523/1000 [14:54<13:50,  1.74s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▏    | 524/1000 [14:56<13:11,  1.66s/it]\u001b[0m\n",
      "\u001b[34m52%|█████▎    | 525/1000 [14:57<13:38,  1.72s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 526/1000 [14:59<14:16,  1.81s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 527/1000 [15:01<12:49,  1.63s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 528/1000 [15:02<12:50,  1.63s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 529/1000 [15:04<13:54,  1.77s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 530/1000 [15:06<13:34,  1.73s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 531/1000 [15:08<13:11,  1.69s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 532/1000 [15:10<13:46,  1.77s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 533/1000 [15:12<14:32,  1.87s/it]\u001b[0m\n",
      "\u001b[34m53%|█████▎    | 534/1000 [15:14<14:47,  1.90s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▎    | 535/1000 [15:16<14:57,  1.93s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▎    | 536/1000 [15:17<12:55,  1.67s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▎    | 537/1000 [15:19<13:55,  1.80s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 538/1000 [15:20<13:09,  1.71s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 539/1000 [15:22<12:37,  1.64s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 540/1000 [15:23<12:32,  1.64s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 541/1000 [15:25<12:51,  1.68s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 542/1000 [15:27<12:11,  1.60s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 543/1000 [15:28<12:15,  1.61s/it]\u001b[0m\n",
      "\u001b[34m54%|█████▍    | 544/1000 [15:30<13:01,  1.71s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▍    | 545/1000 [15:31<11:57,  1.58s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▍    | 546/1000 [15:33<10:54,  1.44s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▍    | 547/1000 [15:34<11:12,  1.48s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▍    | 548/1000 [15:36<12:14,  1.63s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▍    | 549/1000 [15:38<12:13,  1.63s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 550/1000 [15:39<12:13,  1.63s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6027, 'learning_rate': 4.53e-05, 'epoch': 0.06}\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 550/1000 [15:39<12:13,  1.63s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 551/1000 [15:41<12:05,  1.62s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 552/1000 [15:43<12:08,  1.63s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 553/1000 [15:44<11:59,  1.61s/it]\u001b[0m\n",
      "\u001b[34m55%|█████▌    | 554/1000 [15:45<10:46,  1.45s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 555/1000 [15:47<11:01,  1.49s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 556/1000 [15:49<12:24,  1.68s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 557/1000 [15:50<11:54,  1.61s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 558/1000 [15:52<11:54,  1.62s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 559/1000 [15:54<12:59,  1.77s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 560/1000 [15:56<12:14,  1.67s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 561/1000 [15:58<13:13,  1.81s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▌    | 562/1000 [16:00<13:53,  1.90s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▋    | 563/1000 [16:02<14:01,  1.93s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▋    | 564/1000 [16:04<13:20,  1.84s/it]\u001b[0m\n",
      "\u001b[34m56%|█████▋    | 565/1000 [16:05<12:21,  1.71s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 566/1000 [16:07<12:54,  1.78s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 567/1000 [16:08<12:13,  1.69s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 568/1000 [16:10<12:05,  1.68s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 569/1000 [16:11<11:12,  1.56s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 570/1000 [16:13<11:41,  1.63s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 571/1000 [16:15<11:43,  1.64s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 572/1000 [16:17<12:42,  1.78s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 573/1000 [16:18<11:24,  1.60s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▋    | 574/1000 [16:20<11:28,  1.62s/it]\u001b[0m\n",
      "\u001b[34m57%|█████▊    | 575/1000 [16:21<10:36,  1.50s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 576/1000 [16:23<11:53,  1.68s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 577/1000 [16:25<12:47,  1.81s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 578/1000 [16:27<12:50,  1.83s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 579/1000 [16:29<12:24,  1.77s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 580/1000 [16:30<12:24,  1.77s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 581/1000 [16:32<12:01,  1.72s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 582/1000 [16:33<11:20,  1.63s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 583/1000 [16:35<10:33,  1.52s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 584/1000 [16:36<09:56,  1.43s/it]\u001b[0m\n",
      "\u001b[34m58%|█████▊    | 585/1000 [16:38<11:21,  1.64s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▊    | 586/1000 [16:40<11:19,  1.64s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▊    | 587/1000 [16:42<12:16,  1.78s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 588/1000 [16:44<12:55,  1.88s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 589/1000 [16:46<12:42,  1.86s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 590/1000 [16:47<11:27,  1.68s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 591/1000 [16:49<11:17,  1.66s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 592/1000 [16:51<12:03,  1.77s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 593/1000 [16:52<11:43,  1.73s/it]\u001b[0m\n",
      "\u001b[34m59%|█████▉    | 594/1000 [16:54<12:29,  1.85s/it]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 595/1000 [16:56<11:13,  1.66s/it]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 596/1000 [16:58<12:03,  1.79s/it]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 597/1000 [16:59<11:40,  1.74s/it]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 598/1000 [17:01<11:25,  1.71s/it]\u001b[0m\n",
      "\u001b[34m60%|█████▉    | 599/1000 [17:02<10:56,  1.64s/it]\u001b[0m\n",
      "\u001b[34m60%|██████    | 600/1000 [17:04<10:58,  1.65s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6437, 'learning_rate': 4.0300000000000004e-05, 'epoch': 0.07}\u001b[0m\n",
      "\u001b[34m60%|██████    | 600/1000 [17:04<10:58,  1.65s/it]\u001b[0m\n",
      "\u001b[34m60%|██████    | 601/1000 [17:06<11:53,  1.79s/it]\u001b[0m\n",
      "\u001b[34m60%|██████    | 602/1000 [17:08<11:16,  1.70s/it]\u001b[0m\n",
      "\u001b[34m60%|██████    | 603/1000 [17:10<11:31,  1.74s/it]\u001b[0m\n",
      "\u001b[34m60%|██████    | 604/1000 [17:11<10:10,  1.54s/it]\u001b[0m\n",
      "\u001b[34m60%|██████    | 605/1000 [17:12<10:35,  1.61s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 606/1000 [17:15<11:35,  1.76s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 607/1000 [17:16<10:55,  1.67s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 608/1000 [17:17<09:50,  1.51s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 609/1000 [17:18<09:14,  1.42s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 610/1000 [17:20<09:36,  1.48s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 611/1000 [17:22<10:49,  1.67s/it]\u001b[0m\n",
      "\u001b[34m61%|██████    | 612/1000 [17:23<10:18,  1.59s/it]\u001b[0m\n",
      "\u001b[34m61%|██████▏   | 613/1000 [17:25<10:16,  1.59s/it]\u001b[0m\n",
      "\u001b[34m61%|██████▏   | 614/1000 [17:26<09:55,  1.54s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 615/1000 [17:28<10:23,  1.62s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 616/1000 [17:30<10:58,  1.72s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 617/1000 [17:32<11:05,  1.74s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 618/1000 [17:34<10:46,  1.69s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 619/1000 [17:35<10:59,  1.73s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 620/1000 [17:37<11:02,  1.74s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 621/1000 [17:39<11:43,  1.86s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 622/1000 [17:41<10:58,  1.74s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 623/1000 [17:42<10:42,  1.70s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▏   | 624/1000 [17:44<09:51,  1.57s/it]\u001b[0m\n",
      "\u001b[34m62%|██████▎   | 625/1000 [17:45<09:32,  1.53s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 626/1000 [17:47<10:38,  1.71s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 627/1000 [17:49<10:20,  1.66s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 628/1000 [17:51<11:09,  1.80s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 629/1000 [17:53<10:52,  1.76s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 630/1000 [17:54<10:36,  1.72s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 631/1000 [17:56<10:30,  1.71s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 632/1000 [17:58<11:15,  1.84s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 633/1000 [17:59<10:24,  1.70s/it]\u001b[0m\n",
      "\u001b[34m63%|██████▎   | 634/1000 [18:01<10:18,  1.69s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▎   | 635/1000 [18:02<09:45,  1.60s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▎   | 636/1000 [18:05<10:40,  1.76s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▎   | 637/1000 [18:07<11:19,  1.87s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 638/1000 [18:08<10:05,  1.67s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 639/1000 [18:09<09:20,  1.55s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 640/1000 [18:11<09:23,  1.56s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 641/1000 [18:13<10:21,  1.73s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 642/1000 [18:15<11:01,  1.85s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 643/1000 [18:16<09:57,  1.67s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 644/1000 [18:18<10:31,  1.77s/it]\u001b[0m\n",
      "\u001b[34m64%|██████▍   | 645/1000 [18:19<09:03,  1.53s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▍   | 646/1000 [18:20<08:33,  1.45s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▍   | 647/1000 [18:23<09:43,  1.65s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▍   | 648/1000 [18:25<10:31,  1.79s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▍   | 649/1000 [18:26<09:55,  1.70s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 650/1000 [18:28<09:46,  1.68s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5585, 'learning_rate': 3.53e-05, 'epoch': 0.07}\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 650/1000 [18:28<09:46,  1.68s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 651/1000 [18:29<09:38,  1.66s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 652/1000 [18:31<10:12,  1.76s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 653/1000 [18:33<09:56,  1.72s/it]\u001b[0m\n",
      "\u001b[34m65%|██████▌   | 654/1000 [18:35<10:08,  1.76s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 655/1000 [18:37<10:15,  1.79s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 656/1000 [18:39<10:16,  1.79s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 657/1000 [18:40<10:02,  1.76s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 658/1000 [18:42<09:51,  1.73s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 659/1000 [18:43<09:15,  1.63s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 660/1000 [18:45<09:45,  1.72s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 661/1000 [18:47<10:24,  1.84s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▌   | 662/1000 [18:49<09:40,  1.72s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▋   | 663/1000 [18:51<10:20,  1.84s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▋   | 664/1000 [18:53<10:47,  1.93s/it]\u001b[0m\n",
      "\u001b[34m66%|██████▋   | 665/1000 [18:54<09:23,  1.68s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 666/1000 [18:55<08:26,  1.52s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 667/1000 [18:56<07:41,  1.39s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 668/1000 [18:58<08:52,  1.60s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 669/1000 [19:01<09:42,  1.76s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 670/1000 [19:03<10:12,  1.86s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 671/1000 [19:04<09:27,  1.73s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 672/1000 [19:06<10:04,  1.84s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 673/1000 [19:08<10:30,  1.93s/it]\u001b[0m\n",
      "\u001b[34m67%|██████▋   | 674/1000 [19:10<10:01,  1.84s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 675/1000 [19:11<08:55,  1.65s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 676/1000 [19:13<09:31,  1.76s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 677/1000 [19:14<08:41,  1.61s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 678/1000 [19:16<08:41,  1.62s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 679/1000 [19:18<09:01,  1.69s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 680/1000 [19:20<09:41,  1.82s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 681/1000 [19:22<10:09,  1.91s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 682/1000 [19:24<09:23,  1.77s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 683/1000 [19:25<08:45,  1.66s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 684/1000 [19:27<08:26,  1.60s/it]\u001b[0m\n",
      "\u001b[34m68%|██████▊   | 685/1000 [19:28<08:31,  1.62s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▊   | 686/1000 [19:30<08:15,  1.58s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▊   | 687/1000 [19:32<09:07,  1.75s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 688/1000 [19:33<08:39,  1.67s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 689/1000 [19:35<08:32,  1.65s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 690/1000 [19:36<07:52,  1.52s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 691/1000 [19:38<08:15,  1.60s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 692/1000 [19:39<07:41,  1.50s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 693/1000 [19:41<07:33,  1.48s/it]\u001b[0m\n",
      "\u001b[34m69%|██████▉   | 694/1000 [19:43<08:31,  1.67s/it]\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 695/1000 [19:45<09:12,  1.81s/it]\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 696/1000 [19:47<09:16,  1.83s/it]\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 697/1000 [19:49<09:41,  1.92s/it]\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 698/1000 [19:51<09:43,  1.93s/it]\u001b[0m\n",
      "\u001b[34m70%|██████▉   | 699/1000 [19:52<08:21,  1.67s/it]\u001b[0m\n",
      "\u001b[34m70%|███████   | 700/1000 [19:53<08:13,  1.64s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6212, 'learning_rate': 3.03e-05, 'epoch': 0.08}\u001b[0m\n",
      "\u001b[34m70%|███████   | 700/1000 [19:53<08:13,  1.64s/it]\u001b[0m\n",
      "\u001b[34m70%|███████   | 701/1000 [19:55<08:27,  1.70s/it]\u001b[0m\n",
      "\u001b[34m70%|███████   | 702/1000 [19:56<07:33,  1.52s/it]\u001b[0m\n",
      "\u001b[34m70%|███████   | 703/1000 [19:58<07:40,  1.55s/it]\u001b[0m\n",
      "\u001b[34m70%|███████   | 704/1000 [20:00<07:44,  1.57s/it]\u001b[0m\n",
      "\u001b[34m70%|███████   | 705/1000 [20:01<07:28,  1.52s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 706/1000 [20:03<08:05,  1.65s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 707/1000 [20:05<08:46,  1.80s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 708/1000 [20:07<08:49,  1.81s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 709/1000 [20:09<08:51,  1.83s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 710/1000 [20:11<08:50,  1.83s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 711/1000 [20:12<08:43,  1.81s/it]\u001b[0m\n",
      "\u001b[34m71%|███████   | 712/1000 [20:14<08:22,  1.75s/it]\u001b[0m\n",
      "\u001b[34m71%|███████▏  | 713/1000 [20:16<08:10,  1.71s/it]\u001b[0m\n",
      "\u001b[34m71%|███████▏  | 714/1000 [20:17<07:49,  1.64s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 715/1000 [20:19<07:34,  1.59s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 716/1000 [20:20<07:47,  1.65s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 717/1000 [20:23<08:27,  1.79s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 718/1000 [20:24<07:56,  1.69s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 719/1000 [20:26<08:32,  1.82s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 720/1000 [20:28<08:12,  1.76s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 721/1000 [20:30<08:16,  1.78s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 722/1000 [20:31<07:50,  1.69s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 723/1000 [20:32<07:09,  1.55s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▏  | 724/1000 [20:34<07:12,  1.57s/it]\u001b[0m\n",
      "\u001b[34m72%|███████▎  | 725/1000 [20:35<06:43,  1.47s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 726/1000 [20:37<07:12,  1.58s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 727/1000 [20:39<07:42,  1.69s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 728/1000 [20:41<07:37,  1.68s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 729/1000 [20:42<07:47,  1.72s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 730/1000 [20:44<07:19,  1.63s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 731/1000 [20:46<07:57,  1.78s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 732/1000 [20:47<07:15,  1.63s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 733/1000 [20:49<07:18,  1.64s/it]\u001b[0m\n",
      "\u001b[34m73%|███████▎  | 734/1000 [20:51<07:34,  1.71s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▎  | 735/1000 [20:52<07:26,  1.68s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▎  | 736/1000 [20:54<07:07,  1.62s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▎  | 737/1000 [20:56<07:46,  1.77s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 738/1000 [20:58<08:13,  1.88s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 739/1000 [20:59<07:21,  1.69s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 740/1000 [21:01<07:00,  1.62s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 741/1000 [21:03<07:26,  1.72s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 742/1000 [21:04<07:06,  1.65s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 743/1000 [21:05<06:18,  1.47s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 744/1000 [21:07<07:07,  1.67s/it]\u001b[0m\n",
      "\u001b[34m74%|███████▍  | 745/1000 [21:10<07:39,  1.80s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▍  | 746/1000 [21:12<08:01,  1.90s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▍  | 747/1000 [21:14<08:05,  1.92s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▍  | 748/1000 [21:16<08:05,  1.93s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▍  | 749/1000 [21:17<07:56,  1.90s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 750/1000 [21:19<07:48,  1.87s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5444, 'learning_rate': 2.5300000000000002e-05, 'epoch': 0.09}\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 750/1000 [21:19<07:48,  1.87s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 751/1000 [21:21<07:53,  1.90s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 752/1000 [21:23<07:18,  1.77s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 753/1000 [21:24<06:33,  1.60s/it]\u001b[0m\n",
      "\u001b[34m75%|███████▌  | 754/1000 [21:25<06:04,  1.48s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 755/1000 [21:26<05:58,  1.46s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 756/1000 [21:28<06:20,  1.56s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 757/1000 [21:29<05:44,  1.42s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 758/1000 [21:31<05:56,  1.47s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 759/1000 [21:32<05:55,  1.48s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 760/1000 [21:34<06:20,  1.59s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 761/1000 [21:36<06:36,  1.66s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▌  | 762/1000 [21:38<07:09,  1.80s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▋  | 763/1000 [21:40<06:45,  1.71s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▋  | 764/1000 [21:42<07:14,  1.84s/it]\u001b[0m\n",
      "\u001b[34m76%|███████▋  | 765/1000 [21:43<06:44,  1.72s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 766/1000 [21:45<06:35,  1.69s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 767/1000 [21:46<05:58,  1.54s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 768/1000 [21:48<06:04,  1.57s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 769/1000 [21:49<05:42,  1.48s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 770/1000 [21:50<05:26,  1.42s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 771/1000 [21:52<06:13,  1.63s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 772/1000 [21:54<05:59,  1.57s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 773/1000 [21:55<05:46,  1.52s/it]\u001b[0m\n",
      "\u001b[34m77%|███████▋  | 774/1000 [21:57<05:25,  1.44s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 775/1000 [21:59<06:09,  1.64s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 776/1000 [22:00<05:53,  1.58s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 777/1000 [22:02<06:28,  1.74s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 778/1000 [22:04<06:52,  1.86s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 779/1000 [22:06<06:58,  1.89s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 780/1000 [22:08<06:51,  1.87s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 781/1000 [22:10<07:07,  1.95s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 782/1000 [22:12<06:28,  1.78s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 783/1000 [22:13<06:13,  1.72s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 784/1000 [22:15<06:26,  1.79s/it]\u001b[0m\n",
      "\u001b[34m78%|███████▊  | 785/1000 [22:17<06:46,  1.89s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▊  | 786/1000 [22:19<06:25,  1.80s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▊  | 787/1000 [22:20<05:25,  1.53s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 788/1000 [22:22<06:02,  1.71s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 789/1000 [22:24<06:16,  1.79s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 790/1000 [22:26<06:08,  1.75s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 791/1000 [22:27<05:49,  1.67s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 792/1000 [22:28<05:31,  1.59s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 793/1000 [22:30<05:20,  1.55s/it]\u001b[0m\n",
      "\u001b[34m79%|███████▉  | 794/1000 [22:32<05:54,  1.72s/it]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 795/1000 [22:33<05:25,  1.59s/it]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 796/1000 [22:35<05:29,  1.62s/it]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 797/1000 [22:37<05:50,  1.73s/it]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 798/1000 [22:38<05:28,  1.63s/it]\u001b[0m\n",
      "\u001b[34m80%|███████▉  | 799/1000 [22:40<05:45,  1.72s/it]\u001b[0m\n",
      "\u001b[34m80%|████████  | 800/1000 [22:42<06:07,  1.84s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5517, 'learning_rate': 2.0300000000000002e-05, 'epoch': 0.09}\u001b[0m\n",
      "\u001b[34m80%|████████  | 800/1000 [22:42<06:07,  1.84s/it]\u001b[0m\n",
      "\u001b[34m80%|████████  | 801/1000 [22:44<05:39,  1.71s/it]\u001b[0m\n",
      "\u001b[34m80%|████████  | 802/1000 [22:46<06:01,  1.83s/it]\u001b[0m\n",
      "\u001b[34m80%|████████  | 803/1000 [22:48<06:17,  1.92s/it]\u001b[0m\n",
      "\u001b[34m80%|████████  | 804/1000 [22:49<05:38,  1.72s/it]\u001b[0m\n",
      "\u001b[34m80%|████████  | 805/1000 [22:51<05:30,  1.69s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 806/1000 [22:53<05:53,  1.82s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 807/1000 [22:55<05:53,  1.83s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 808/1000 [22:57<05:40,  1.77s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 809/1000 [22:58<05:31,  1.73s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 810/1000 [23:00<05:22,  1.70s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 811/1000 [23:02<05:44,  1.82s/it]\u001b[0m\n",
      "\u001b[34m81%|████████  | 812/1000 [23:04<05:42,  1.82s/it]\u001b[0m\n",
      "\u001b[34m81%|████████▏ | 813/1000 [23:05<05:17,  1.70s/it]\u001b[0m\n",
      "\u001b[34m81%|████████▏ | 814/1000 [23:07<05:09,  1.66s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 815/1000 [23:08<05:05,  1.65s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 816/1000 [23:10<04:41,  1.53s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 817/1000 [23:12<05:03,  1.66s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 818/1000 [23:14<05:21,  1.77s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 819/1000 [23:16<05:39,  1.87s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 820/1000 [23:18<05:41,  1.90s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 821/1000 [23:19<05:23,  1.81s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 822/1000 [23:21<05:19,  1.80s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 823/1000 [23:23<05:35,  1.90s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▏ | 824/1000 [23:25<05:20,  1.82s/it]\u001b[0m\n",
      "\u001b[34m82%|████████▎ | 825/1000 [23:27<05:34,  1.91s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 826/1000 [23:28<04:59,  1.72s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 827/1000 [23:29<04:33,  1.58s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 828/1000 [23:31<04:35,  1.60s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 829/1000 [23:33<05:01,  1.76s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 830/1000 [23:35<04:42,  1.66s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 831/1000 [23:36<04:46,  1.69s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 832/1000 [23:38<04:50,  1.73s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 833/1000 [23:40<05:04,  1.82s/it]\u001b[0m\n",
      "\u001b[34m83%|████████▎ | 834/1000 [23:42<05:08,  1.86s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▎ | 835/1000 [23:44<05:06,  1.86s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▎ | 836/1000 [23:46<04:51,  1.78s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▎ | 837/1000 [23:47<04:41,  1.73s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 838/1000 [23:49<04:24,  1.63s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 839/1000 [23:51<04:46,  1.78s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 840/1000 [23:53<05:00,  1.88s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 841/1000 [23:55<04:45,  1.79s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 842/1000 [23:56<04:33,  1.73s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 843/1000 [23:58<04:49,  1.84s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 844/1000 [24:00<04:44,  1.82s/it]\u001b[0m\n",
      "\u001b[34m84%|████████▍ | 845/1000 [24:02<04:32,  1.76s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▍ | 846/1000 [24:03<04:33,  1.77s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▍ | 847/1000 [24:05<04:32,  1.78s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▍ | 848/1000 [24:07<04:32,  1.79s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▍ | 849/1000 [24:08<04:06,  1.63s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 850/1000 [24:10<04:26,  1.78s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.6048, 'learning_rate': 1.53e-05, 'epoch': 0.1}\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 850/1000 [24:10<04:26,  1.78s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 851/1000 [24:12<04:19,  1.74s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 852/1000 [24:14<04:05,  1.66s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 853/1000 [24:15<03:39,  1.49s/it]\u001b[0m\n",
      "\u001b[34m85%|████████▌ | 854/1000 [24:16<03:19,  1.36s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 855/1000 [24:17<03:29,  1.44s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 856/1000 [24:19<03:29,  1.46s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 857/1000 [24:20<03:33,  1.49s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 858/1000 [24:23<03:58,  1.68s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 859/1000 [24:24<04:03,  1.73s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 860/1000 [24:26<04:18,  1.85s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 861/1000 [24:28<04:02,  1.74s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▌ | 862/1000 [24:29<03:48,  1.66s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▋ | 863/1000 [24:31<03:43,  1.63s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▋ | 864/1000 [24:32<03:34,  1.57s/it]\u001b[0m\n",
      "\u001b[34m86%|████████▋ | 865/1000 [24:35<03:55,  1.74s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 866/1000 [24:36<03:49,  1.71s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 867/1000 [24:38<04:00,  1.81s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 868/1000 [24:40<04:11,  1.90s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 869/1000 [24:42<03:59,  1.83s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 870/1000 [24:43<03:21,  1.55s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 871/1000 [24:44<03:14,  1.51s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 872/1000 [24:46<03:36,  1.69s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 873/1000 [24:48<03:39,  1.73s/it]\u001b[0m\n",
      "\u001b[34m87%|████████▋ | 874/1000 [24:50<03:52,  1.85s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 875/1000 [24:52<03:55,  1.88s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 876/1000 [24:54<04:02,  1.95s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 877/1000 [24:56<03:34,  1.74s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 878/1000 [24:57<03:16,  1.61s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 879/1000 [24:59<03:23,  1.69s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 880/1000 [25:01<03:34,  1.79s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 881/1000 [25:03<03:44,  1.88s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 882/1000 [25:04<03:20,  1.70s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 883/1000 [25:06<03:33,  1.83s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 884/1000 [25:09<03:41,  1.91s/it]\u001b[0m\n",
      "\u001b[34m88%|████████▊ | 885/1000 [25:10<03:23,  1.77s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▊ | 886/1000 [25:12<03:33,  1.87s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▊ | 887/1000 [25:14<03:35,  1.90s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 888/1000 [25:16<03:31,  1.89s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 889/1000 [25:18<03:25,  1.85s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 890/1000 [25:20<03:32,  1.93s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 891/1000 [25:22<03:25,  1.88s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 892/1000 [25:23<03:19,  1.85s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 893/1000 [25:25<03:24,  1.91s/it]\u001b[0m\n",
      "\u001b[34m89%|████████▉ | 894/1000 [25:27<03:14,  1.83s/it]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 895/1000 [25:28<03:00,  1.72s/it]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 896/1000 [25:29<02:34,  1.49s/it]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 897/1000 [25:31<02:42,  1.58s/it]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 898/1000 [25:33<02:49,  1.67s/it]\u001b[0m\n",
      "\u001b[34m90%|████████▉ | 899/1000 [25:35<02:41,  1.60s/it]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 900/1000 [25:37<02:55,  1.76s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5152, 'learning_rate': 1.03e-05, 'epoch': 0.1}\u001b[0m\n",
      "\u001b[34m90%|█████████ | 900/1000 [25:37<02:55,  1.76s/it]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 901/1000 [25:39<03:04,  1.87s/it]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 902/1000 [25:40<02:56,  1.80s/it]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 903/1000 [25:43<03:03,  1.89s/it]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 904/1000 [25:44<02:48,  1.76s/it]\u001b[0m\n",
      "\u001b[34m90%|█████████ | 905/1000 [25:46<02:54,  1.84s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 906/1000 [25:48<03:00,  1.92s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 907/1000 [25:49<02:38,  1.71s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 908/1000 [25:51<02:40,  1.75s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 909/1000 [25:53<02:39,  1.76s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 910/1000 [25:55<02:34,  1.72s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 911/1000 [25:56<02:36,  1.76s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████ | 912/1000 [25:58<02:25,  1.65s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████▏| 913/1000 [26:00<02:27,  1.69s/it]\u001b[0m\n",
      "\u001b[34m91%|█████████▏| 914/1000 [26:01<02:17,  1.60s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 915/1000 [26:02<02:06,  1.48s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 916/1000 [26:04<02:03,  1.47s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 917/1000 [26:05<02:05,  1.52s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 918/1000 [26:07<02:06,  1.55s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 919/1000 [26:08<01:58,  1.46s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 920/1000 [26:10<02:13,  1.66s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 921/1000 [26:12<02:01,  1.54s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 922/1000 [26:13<02:02,  1.57s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 923/1000 [26:15<02:13,  1.74s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▏| 924/1000 [26:17<02:06,  1.66s/it]\u001b[0m\n",
      "\u001b[34m92%|█████████▎| 925/1000 [26:19<02:10,  1.75s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 926/1000 [26:20<02:05,  1.70s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 927/1000 [26:22<02:13,  1.83s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 928/1000 [26:24<02:05,  1.75s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 929/1000 [26:26<02:12,  1.86s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 930/1000 [26:28<02:15,  1.94s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 931/1000 [26:30<02:07,  1.85s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 932/1000 [26:32<02:04,  1.83s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 933/1000 [26:34<02:08,  1.92s/it]\u001b[0m\n",
      "\u001b[34m93%|█████████▎| 934/1000 [26:35<02:00,  1.82s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▎| 935/1000 [26:38<02:04,  1.91s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▎| 936/1000 [26:39<01:52,  1.76s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▎| 937/1000 [26:41<01:57,  1.87s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 938/1000 [26:43<02:00,  1.95s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 939/1000 [26:44<01:43,  1.69s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 940/1000 [26:46<01:43,  1.73s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 941/1000 [26:48<01:39,  1.69s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 942/1000 [26:50<01:45,  1.82s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 943/1000 [26:52<01:46,  1.87s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 944/1000 [26:53<01:34,  1.68s/it]\u001b[0m\n",
      "\u001b[34m94%|█████████▍| 945/1000 [26:55<01:37,  1.77s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▍| 946/1000 [26:57<01:30,  1.68s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▍| 947/1000 [26:58<01:33,  1.77s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▍| 948/1000 [27:01<01:37,  1.88s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▍| 949/1000 [27:03<01:39,  1.95s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 950/1000 [27:04<01:33,  1.87s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5755, 'learning_rate': 5.3e-06, 'epoch': 0.11}\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 950/1000 [27:04<01:33,  1.87s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 951/1000 [27:07<01:35,  1.95s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 952/1000 [27:08<01:23,  1.73s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 953/1000 [27:10<01:24,  1.80s/it]\u001b[0m\n",
      "\u001b[34m95%|█████████▌| 954/1000 [27:12<01:27,  1.90s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 955/1000 [27:13<01:16,  1.70s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 956/1000 [27:15<01:11,  1.63s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 957/1000 [27:16<01:08,  1.58s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 958/1000 [27:17<01:02,  1.48s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 959/1000 [27:19<01:02,  1.52s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 960/1000 [27:21<01:04,  1.61s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 961/1000 [27:23<01:08,  1.76s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▌| 962/1000 [27:24<01:05,  1.72s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▋| 963/1000 [27:26<01:02,  1.70s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▋| 964/1000 [27:27<00:56,  1.56s/it]\u001b[0m\n",
      "\u001b[34m96%|█████████▋| 965/1000 [27:29<00:58,  1.68s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 966/1000 [27:31<00:54,  1.60s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 967/1000 [27:33<00:57,  1.75s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 968/1000 [27:34<00:53,  1.67s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 969/1000 [27:36<00:51,  1.66s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 970/1000 [27:38<00:53,  1.79s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 971/1000 [27:39<00:48,  1.68s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 972/1000 [27:41<00:48,  1.74s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 973/1000 [27:43<00:45,  1.69s/it]\u001b[0m\n",
      "\u001b[34m97%|█████████▋| 974/1000 [27:45<00:45,  1.77s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 975/1000 [27:46<00:39,  1.57s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 976/1000 [27:48<00:39,  1.63s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 977/1000 [27:49<00:34,  1.52s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 978/1000 [27:51<00:35,  1.62s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 979/1000 [27:52<00:31,  1.51s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 980/1000 [27:54<00:31,  1.60s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 981/1000 [27:56<00:33,  1.75s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 982/1000 [27:58<00:30,  1.72s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 983/1000 [28:00<00:31,  1.84s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 984/1000 [28:02<00:30,  1.89s/it]\u001b[0m\n",
      "\u001b[34m98%|█████████▊| 985/1000 [28:03<00:27,  1.80s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▊| 986/1000 [28:05<00:24,  1.74s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▊| 987/1000 [28:06<00:21,  1.64s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 988/1000 [28:08<00:19,  1.63s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 989/1000 [28:09<00:17,  1.56s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 990/1000 [28:11<00:16,  1.69s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 991/1000 [28:14<00:16,  1.82s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 992/1000 [28:15<00:14,  1.77s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 993/1000 [28:17<00:11,  1.68s/it]\u001b[0m\n",
      "\u001b[34m99%|█████████▉| 994/1000 [28:18<00:09,  1.61s/it]\u001b[0m\n",
      "\u001b[34m100%|█████████▉| 995/1000 [28:19<00:07,  1.46s/it]\u001b[0m\n",
      "\u001b[34m100%|█████████▉| 996/1000 [28:20<00:05,  1.31s/it]\u001b[0m\n",
      "\u001b[34m100%|█████████▉| 997/1000 [28:22<00:04,  1.35s/it]\u001b[0m\n",
      "\u001b[34m100%|█████████▉| 998/1000 [28:23<00:03,  1.50s/it]\u001b[0m\n",
      "\u001b[34m100%|█████████▉| 999/1000 [28:26<00:01,  1.69s/it]\u001b[0m\n",
      "\u001b[34m100%|██████████| 1000/1000 [28:27<00:00,  1.67s/it]\u001b[0m\n",
      "\u001b[34m{'loss': 1.5623, 'learning_rate': 3.0000000000000004e-07, 'epoch': 0.12}\u001b[0m\n",
      "\u001b[34m100%|██████████| 1000/1000 [28:27<00:00,  1.67s/it]\u001b[0m\n",
      "\u001b[34m{'train_runtime': 1707.7623, 'train_samples_per_second': 3.513, 'train_steps_per_second': 0.586, 'train_loss': 1.621132553100586, 'epoch': 0.12}\u001b[0m\n",
      "\u001b[34m100%|██████████| 1000/1000 [28:27<00:00,  1.67s/it]\u001b[0m\n",
      "\u001b[34m100%|██████████| 1000/1000 [28:27<00:00,  1.71s/it]\u001b[0m\n",
      "\u001b[34mcp /tmp/output/README.md s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/README.md\u001b[0m\n",
      "\u001b[34mcp /tmp/output/adapter_config.json s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/adapter_config.json\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/scheduler.pt s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/scheduler.pt\u001b[0m\n",
      "\u001b[34mcp /tmp/output/runs/Jun20_09-32-01_algo-1/1687253581.3341768/events.out.tfevents.1687253581.algo-1.249.2 s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/runs/Jun20_09-32-01_algo-1/1687253581.3341768/events.out.tfevents.1687253581.algo-1.249.2\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/training_args.bin s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/training_args.bin\u001b[0m\n",
      "\u001b[34mcp /tmp/output/runs/Jun20_09-32-01_algo-1/events.out.tfevents.1687253581.algo-1.249.1 s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/runs/Jun20_09-32-01_algo-1/events.out.tfevents.1687253581.algo-1.249.1\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/scaler.pt s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/scaler.pt\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/trainer_state.json s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/trainer_state.json\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/rng_state.pth s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/rng_state.pth\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/adapter_model.bin s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/adapter_model.bin\u001b[0m\n",
      "\u001b[34mcp /tmp/output/adapter_model.bin s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/adapter_model.bin\u001b[0m\n",
      "\u001b[34mcp /tmp/output/checkpoint-1000/optimizer.pt s3://sagemaker-us-west-2-687912291502/models/chatglm-lora/output/2023-06-20-10-01-30/checkpoint-1000/optimizer.pt\u001b[0m\n",
      "\u001b[34m2023-06-20 10:01:30,662 sagemaker-training-toolkit INFO     Waiting for the process to finish and give a return code.\u001b[0m\n",
      "\u001b[34m2023-06-20 10:01:30,662 sagemaker-training-toolkit INFO     Done waiting for a return code. Received 0 from exiting process.\u001b[0m\n",
      "\u001b[34m2023-06-20 10:01:30,662 sagemaker-training-toolkit INFO     Reporting training SUCCESS\u001b[0m\n",
      "\n",
      "2023-06-20 10:01:40 Uploading - Uploading generated training model\n",
      "2023-06-20 10:01:40 Completed - Training job completed\n",
      "Training seconds: 2124\n",
      "Billable seconds: 2124\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "from sagemaker.estimator import Estimator\n",
    "\n",
    "environment = {\n",
    "              'MODEL_S3_BUCKET': bucket # The bucket to store pretrained model and fine-tune model\n",
    "}\n",
    "\n",
    "base_job_name = 'chatglm-lora-demo'         \n",
    "\n",
    "instance_type = 'ml.g5.12xlarge'\n",
    "\n",
    "estimator = Estimator(role=role,\n",
    "                      entry_point='train.sh',\n",
    "                      source_dir='./ChatGLM-Tuning/',\n",
    "                      base_job_name=base_job_name,\n",
    "                      instance_count=1,\n",
    "                      instance_type=instance_type,\n",
    "                      image_uri=image_uri,\n",
    "                      environment=environment,\n",
    "                      disable_profiler=True,\n",
    "                      debugger_hook_config=False,\n",
    "                      max_run=24*60*60*2)\n",
    "\n",
    "estimator.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e6740ffb-6951-46a2-a9da-8677261f9147",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cbea5eca-ff85-4226-95fb-3d0f979f5e83",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_amazonei_pytorch_latest_p37",
   "language": "python",
   "name": "conda_amazonei_pytorch_latest_p37"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}