ab_predict_neutralising_final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a95dd8f0-30d4-44e7-bd86-c951366f30cc",
   "metadata": {},
   "source": [
    "# Install Librarires"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e37ca52f-ba23-4fbe-8d85-f677c7817b27",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow in /usr/local/lib/python3.11/site-packages (2.15.0.post1)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.1.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.6.3)\n",
      "Requirement already satisfied: flatbuffers>=23.5.26 in /usr/local/lib/python3.11/site-packages (from tensorflow) (23.5.26)\n",
      "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.5.4)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.2.0)\n",
      "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (3.10.0)\n",
      "Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (16.0.6)\n",
      "Requirement already satisfied: ml-dtypes~=0.2.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.2.0)\n",
      "Requirement already satisfied: numpy<2.0.0,>=1.23.5 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.26.4)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.11/site-packages (from tensorflow) (3.3.0)\n",
      "Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from tensorflow) (23.2)\n",
      "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in /usr/local/lib/python3.11/site-packages (from tensorflow) (4.25.3)\n",
      "Requirement already satisfied: setuptools in /usr/local/lib/python3.11/site-packages (from tensorflow) (65.5.1)\n",
      "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.16.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.4.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.11/site-packages (from tensorflow) (4.10.0)\n",
      "Requirement already satisfied: wrapt<1.15,>=1.11.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.14.1)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.36.0)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.62.0)\n",
      "Requirement already satisfied: tensorboard<2.16,>=2.15 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.15.2)\n",
      "Requirement already satisfied: tensorflow-estimator<2.16,>=2.15.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.15.0)\n",
      "Requirement already satisfied: keras<2.16,>=2.15.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.15.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.11/site-packages (from astunparse>=1.6.0->tensorflow) (0.42.0)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (2.28.1)\n",
      "Requirement already satisfied: google-auth-oauthlib<2,>=0.5 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (1.2.0)\n",
      "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (3.5.2)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (2.31.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (0.7.2)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (3.0.1)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (5.3.3)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (0.3.0)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.11/site-packages (from google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow) (1.3.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (2.0.7)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (2024.2.2)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.11/site-packages (from werkzeug>=1.0.1->tensorboard<2.16,>=2.15->tensorflow) (2.1.5)\n",
      "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /usr/local/lib/python3.11/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (0.5.1)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.11/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow) (3.2.2)\n",
      "Requirement already satisfied: ablang in /usr/local/lib/python3.11/site-packages (0.3.1)\n",
      "Requirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from ablang) (2.31.0)\n",
      "Requirement already satisfied: torch>=1.6 in /usr/local/lib/python3.11/site-packages (from ablang) (2.2.1)\n",
      "Requirement already satisfied: numpy in /usr/local/lib/python3.11/site-packages (from ablang) (1.26.4)\n",
      "Requirement already satisfied: pandas in /usr/local/lib/python3.11/site-packages (from ablang) (2.2.1)\n",
      "Requirement already satisfied: numba in /usr/local/lib/python3.11/site-packages (from ablang) (0.59.0)\n",
      "Requirement already satisfied: filelock in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (3.13.1)\n",
      "Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (4.10.0)\n",
      "Requirement already satisfied: sympy in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (1.12)\n",
      "Requirement already satisfied: networkx in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (3.2.1)\n",
      "Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (3.1.3)\n",
      "Requirement already satisfied: fsspec in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (2024.2.0)\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (8.9.2.26)\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (12.1.3.1)\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (11.0.2.54)\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (10.3.2.106)\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (11.4.5.107)\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (12.1.0.106)\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.19.3 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (2.19.3)\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (12.1.105)\n",
      "Requirement already satisfied: triton==2.2.0 in /usr/local/lib/python3.11/site-packages (from torch>=1.6->ablang) (2.2.0)\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /usr/local/lib/python3.11/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch>=1.6->ablang) (12.3.101)\n",
      "Requirement already satisfied: llvmlite<0.43,>=0.42.0dev0 in /usr/local/lib/python3.11/site-packages (from numba->ablang) (0.42.0)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from pandas->ablang) (2.8.2)\n",
      "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.11/site-packages (from pandas->ablang) (2024.1)\n",
      "Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.11/site-packages (from pandas->ablang) (2024.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests->ablang) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->ablang) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->ablang) (2.0.7)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->ablang) (2024.2.2)\n",
      "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.11/site-packages (from python-dateutil>=2.8.2->pandas->ablang) (1.16.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2->torch>=1.6->ablang) (2.1.5)\n",
      "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.11/site-packages (from sympy->torch>=1.6->ablang) (1.3.0)\n",
      "Requirement already satisfied: ablang2 in /usr/local/lib/python3.11/site-packages (0.1.0)\n",
      "Requirement already satisfied: torch>1.9 in /usr/local/lib/python3.11/site-packages (from ablang2) (2.2.1)\n",
      "Requirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from ablang2) (2.31.0)\n",
      "Requirement already satisfied: einops in /usr/local/lib/python3.11/site-packages (from ablang2) (0.7.0)\n",
      "Requirement already satisfied: rotary-embedding-torch in /usr/local/lib/python3.11/site-packages (from ablang2) (0.5.3)\n",
      "Requirement already satisfied: numpy in /usr/local/lib/python3.11/site-packages (from ablang2) (1.26.4)\n",
      "Requirement already satisfied: filelock in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (3.13.1)\n",
      "Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (4.10.0)\n",
      "Requirement already satisfied: sympy in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (1.12)\n",
      "Requirement already satisfied: networkx in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (3.2.1)\n",
      "Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (3.1.3)\n",
      "Requirement already satisfied: fsspec in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (2024.2.0)\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (8.9.2.26)\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (12.1.3.1)\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (11.0.2.54)\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (10.3.2.106)\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (11.4.5.107)\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (12.1.0.106)\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.19.3 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (2.19.3)\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (12.1.105)\n",
      "Requirement already satisfied: triton==2.2.0 in /usr/local/lib/python3.11/site-packages (from torch>1.9->ablang2) (2.2.0)\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /usr/local/lib/python3.11/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch>1.9->ablang2) (12.3.101)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests->ablang2) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->ablang2) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->ablang2) (2.0.7)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->ablang2) (2024.2.2)\n",
      "Requirement already satisfied: beartype in /usr/local/lib/python3.11/site-packages (from rotary-embedding-torch->ablang2) (0.17.2)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2->torch>1.9->ablang2) (2.1.5)\n",
      "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.11/site-packages (from sympy->torch>1.9->ablang2) (1.3.0)\n",
      "Requirement already satisfied: tensorflow in /usr/local/lib/python3.11/site-packages (2.15.0.post1)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.1.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.6.3)\n",
      "Requirement already satisfied: flatbuffers>=23.5.26 in /usr/local/lib/python3.11/site-packages (from tensorflow) (23.5.26)\n",
      "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.5.4)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.2.0)\n",
      "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (3.10.0)\n",
      "Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (16.0.6)\n",
      "Requirement already satisfied: ml-dtypes~=0.2.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.2.0)\n",
      "Requirement already satisfied: numpy<2.0.0,>=1.23.5 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.26.4)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.11/site-packages (from tensorflow) (3.3.0)\n",
      "Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from tensorflow) (23.2)\n",
      "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in /usr/local/lib/python3.11/site-packages (from tensorflow) (4.25.3)\n",
      "Requirement already satisfied: setuptools in /usr/local/lib/python3.11/site-packages (from tensorflow) (65.5.1)\n",
      "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.16.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.4.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.11/site-packages (from tensorflow) (4.10.0)\n",
      "Requirement already satisfied: wrapt<1.15,>=1.11.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.14.1)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.11/site-packages (from tensorflow) (0.36.0)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.11/site-packages (from tensorflow) (1.62.0)\n",
      "Requirement already satisfied: tensorboard<2.16,>=2.15 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.15.2)\n",
      "Requirement already satisfied: tensorflow-estimator<2.16,>=2.15.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.15.0)\n",
      "Requirement already satisfied: keras<2.16,>=2.15.0 in /usr/local/lib/python3.11/site-packages (from tensorflow) (2.15.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.11/site-packages (from astunparse>=1.6.0->tensorflow) (0.42.0)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (2.28.1)\n",
      "Requirement already satisfied: google-auth-oauthlib<2,>=0.5 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (1.2.0)\n",
      "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (3.5.2)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (2.31.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (0.7.2)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.11/site-packages (from tensorboard<2.16,>=2.15->tensorflow) (3.0.1)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (5.3.3)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (0.3.0)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.11/site-packages (from google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow) (1.3.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (2.0.7)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow) (2024.2.2)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.11/site-packages (from werkzeug>=1.0.1->tensorboard<2.16,>=2.15->tensorflow) (2.1.5)\n",
      "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /usr/local/lib/python3.11/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow) (0.5.1)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.11/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow) (3.2.2)\n",
      "Requirement already satisfied: tensorboardX in /usr/local/lib/python3.11/site-packages (2.6.2.2)\n",
      "Requirement already satisfied: numpy in /usr/local/lib/python3.11/site-packages (from tensorboardX) (1.26.4)\n",
      "Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from tensorboardX) (23.2)\n",
      "Requirement already satisfied: protobuf>=3.20 in /usr/local/lib/python3.11/site-packages (from tensorboardX) (4.25.3)\n",
      "Requirement already satisfied: icn3dpy in /usr/local/lib/python3.11/site-packages (3.30.0)\n",
      "Requirement already satisfied: idisplay in /usr/local/lib/python3.11/site-packages (from icn3dpy) (0.1.2)\n",
      "Requirement already satisfied: jupyter in /usr/local/lib/python3.11/site-packages (from icn3dpy) (1.0.0)\n",
      "Requirement already satisfied: ipython>=1.0 in /usr/local/lib/python3.11/site-packages (from idisplay->icn3dpy) (8.22.1)\n",
      "Requirement already satisfied: notebook in /usr/local/lib/python3.11/site-packages (from jupyter->icn3dpy) (7.1.1)\n",
      "Requirement already satisfied: qtconsole in /usr/local/lib/python3.11/site-packages (from jupyter->icn3dpy) (5.5.1)\n",
      "Requirement already satisfied: jupyter-console in /usr/local/lib/python3.11/site-packages (from jupyter->icn3dpy) (6.6.3)\n",
      "Requirement already satisfied: nbconvert in /usr/local/lib/python3.11/site-packages (from jupyter->icn3dpy) (7.16.1)\n",
      "Requirement already satisfied: ipykernel in /usr/local/lib/python3.11/site-packages (from jupyter->icn3dpy) (6.29.3)\n",
      "Requirement already satisfied: ipywidgets in /usr/local/lib/python3.11/site-packages (from jupyter->icn3dpy) (8.1.2)\n",
      "Requirement already satisfied: decorator in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (5.1.1)\n",
      "Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (0.19.1)\n",
      "Requirement already satisfied: matplotlib-inline in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (0.1.6)\n",
      "Requirement already satisfied: prompt-toolkit<3.1.0,>=3.0.41 in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (3.0.43)\n",
      "Requirement already satisfied: pygments>=2.4.0 in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (2.17.2)\n",
      "Requirement already satisfied: stack-data in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (0.6.3)\n",
      "Requirement already satisfied: traitlets>=5.13.0 in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (5.14.1)\n",
      "Requirement already satisfied: pexpect>4.3 in /usr/local/lib/python3.11/site-packages (from ipython>=1.0->idisplay->icn3dpy) (4.9.0)\n",
      "Requirement already satisfied: comm>=0.1.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (0.2.1)\n",
      "Requirement already satisfied: debugpy>=1.6.5 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (1.8.1)\n",
      "Requirement already satisfied: jupyter-client>=6.1.12 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (8.6.0)\n",
      "Requirement already satisfied: jupyter-core!=5.0.*,>=4.12 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (5.7.1)\n",
      "Requirement already satisfied: nest-asyncio in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (1.6.0)\n",
      "Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (23.2)\n",
      "Requirement already satisfied: psutil in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (5.9.8)\n",
      "Requirement already satisfied: pyzmq>=24 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (25.1.2)\n",
      "Requirement already satisfied: tornado>=6.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyter->icn3dpy) (6.4)\n",
      "Requirement already satisfied: widgetsnbextension~=4.0.10 in /usr/local/lib/python3.11/site-packages (from ipywidgets->jupyter->icn3dpy) (4.0.10)\n",
      "Requirement already satisfied: jupyterlab-widgets~=3.0.10 in /usr/local/lib/python3.11/site-packages (from ipywidgets->jupyter->icn3dpy) (3.0.10)\n",
      "Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (4.12.3)\n",
      "Requirement already satisfied: bleach!=5.0.0 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (6.1.0)\n",
      "Requirement already satisfied: defusedxml in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (0.7.1)\n",
      "Requirement already satisfied: jinja2>=3.0 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (3.1.3)\n",
      "Requirement already satisfied: jupyterlab-pygments in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (0.3.0)\n",
      "Requirement already satisfied: markupsafe>=2.0 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (2.1.5)\n",
      "Requirement already satisfied: mistune<4,>=2.0.3 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (3.0.2)\n",
      "Requirement already satisfied: nbclient>=0.5.0 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (0.9.0)\n",
      "Requirement already satisfied: nbformat>=5.7 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (5.9.2)\n",
      "Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (1.5.1)\n",
      "Requirement already satisfied: tinycss2 in /usr/local/lib/python3.11/site-packages (from nbconvert->jupyter->icn3dpy) (1.2.1)\n",
      "Requirement already satisfied: jupyter-server<3,>=2.4.0 in /usr/local/lib/python3.11/site-packages (from notebook->jupyter->icn3dpy) (2.12.5)\n",
      "Requirement already satisfied: jupyterlab-server<3,>=2.22.1 in /usr/local/lib/python3.11/site-packages (from notebook->jupyter->icn3dpy) (2.25.3)\n",
      "Requirement already satisfied: jupyterlab<4.2,>=4.1.1 in /usr/local/lib/python3.11/site-packages (from notebook->jupyter->icn3dpy) (4.1.2)\n",
      "Requirement already satisfied: notebook-shim<0.3,>=0.2 in /usr/local/lib/python3.11/site-packages (from notebook->jupyter->icn3dpy) (0.2.4)\n",
      "Requirement already satisfied: qtpy>=2.4.0 in /usr/local/lib/python3.11/site-packages (from qtconsole->jupyter->icn3dpy) (2.4.1)\n",
      "Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert->jupyter->icn3dpy) (1.16.0)\n",
      "Requirement already satisfied: webencodings in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert->jupyter->icn3dpy) (0.5.1)\n",
      "Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jedi>=0.16->ipython>=1.0->idisplay->icn3dpy) (0.8.3)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-client>=6.1.12->ipykernel->jupyter->icn3dpy) (2.8.2)\n",
      "Requirement already satisfied: platformdirs>=2.5 in /usr/local/lib/python3.11/site-packages (from jupyter-core!=5.0.*,>=4.12->ipykernel->jupyter->icn3dpy) (4.2.0)\n",
      "Requirement already satisfied: anyio>=3.1.0 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (4.3.0)\n",
      "Requirement already satisfied: argon2-cffi in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (23.1.0)\n",
      "Requirement already satisfied: jupyter-events>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (0.9.0)\n",
      "Requirement already satisfied: jupyter-server-terminals in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (0.5.2)\n",
      "Requirement already satisfied: overrides in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (7.7.0)\n",
      "Requirement already satisfied: prometheus-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (0.20.0)\n",
      "Requirement already satisfied: send2trash>=1.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.8.2)\n",
      "Requirement already satisfied: terminado>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (0.18.0)\n",
      "Requirement already satisfied: websocket-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.7.0)\n",
      "Requirement already satisfied: async-lru>=1.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab<4.2,>=4.1.1->notebook->jupyter->icn3dpy) (2.0.4)\n",
      "Requirement already satisfied: httpx>=0.25.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab<4.2,>=4.1.1->notebook->jupyter->icn3dpy) (0.27.0)\n",
      "Requirement already satisfied: jupyter-lsp>=2.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab<4.2,>=4.1.1->notebook->jupyter->icn3dpy) (2.2.3)\n",
      "Requirement already satisfied: babel>=2.10 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (2.14.0)\n",
      "Requirement already satisfied: json5>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (0.9.17)\n",
      "Requirement already satisfied: jsonschema>=4.18.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (4.21.1)\n",
      "Requirement already satisfied: requests>=2.31 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (2.31.0)\n",
      "Requirement already satisfied: fastjsonschema in /usr/local/lib/python3.11/site-packages (from nbformat>=5.7->nbconvert->jupyter->icn3dpy) (2.19.1)\n",
      "Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.11/site-packages (from pexpect>4.3->ipython>=1.0->idisplay->icn3dpy) (0.7.0)\n",
      "Requirement already satisfied: wcwidth in /usr/local/lib/python3.11/site-packages (from prompt-toolkit<3.1.0,>=3.0.41->ipython>=1.0->idisplay->icn3dpy) (0.2.13)\n",
      "Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.11/site-packages (from beautifulsoup4->nbconvert->jupyter->icn3dpy) (2.5)\n",
      "Requirement already satisfied: executing>=1.2.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=1.0->idisplay->icn3dpy) (2.0.1)\n",
      "Requirement already satisfied: asttokens>=2.1.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=1.0->idisplay->icn3dpy) (2.4.1)\n",
      "Requirement already satisfied: pure-eval in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=1.0->idisplay->icn3dpy) (0.2.2)\n",
      "Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.11/site-packages (from anyio>=3.1.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (3.6)\n",
      "Requirement already satisfied: sniffio>=1.1 in /usr/local/lib/python3.11/site-packages (from anyio>=3.1.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.3.1)\n",
      "Requirement already satisfied: certifi in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab<4.2,>=4.1.1->notebook->jupyter->icn3dpy) (2024.2.2)\n",
      "Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab<4.2,>=4.1.1->notebook->jupyter->icn3dpy) (1.0.4)\n",
      "Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.11/site-packages (from httpcore==1.*->httpx>=0.25.0->jupyterlab<4.2,>=4.1.1->notebook->jupyter->icn3dpy) (0.14.0)\n",
      "Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (23.2.0)\n",
      "Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (2023.12.1)\n",
      "Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (0.33.0)\n",
      "Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (0.18.0)\n",
      "Requirement already satisfied: python-json-logger>=2.0.4 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (2.0.7)\n",
      "Requirement already satisfied: pyyaml>=5.3 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (6.0.1)\n",
      "Requirement already satisfied: rfc3339-validator in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (0.1.4)\n",
      "Requirement already satisfied: rfc3986-validator>=0.1.1 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (0.1.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (3.3.2)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.22.1->notebook->jupyter->icn3dpy) (2.0.7)\n",
      "Requirement already satisfied: argon2-cffi-bindings in /usr/local/lib/python3.11/site-packages (from argon2-cffi->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (21.2.0)\n",
      "Requirement already satisfied: fqdn in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.5.1)\n",
      "Requirement already satisfied: isoduration in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (20.11.0)\n",
      "Requirement already satisfied: jsonpointer>1.13 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (2.4)\n",
      "Requirement already satisfied: uri-template in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.3.0)\n",
      "Requirement already satisfied: webcolors>=1.11 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.13)\n",
      "Requirement already satisfied: cffi>=1.0.1 in /usr/local/lib/python3.11/site-packages (from argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.16.0)\n",
      "Requirement already satisfied: pycparser in /usr/local/lib/python3.11/site-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (2.21)\n",
      "Requirement already satisfied: arrow>=0.15.0 in /usr/local/lib/python3.11/site-packages (from isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (1.3.0)\n",
      "Requirement already satisfied: types-python-dateutil>=2.8.10 in /usr/local/lib/python3.11/site-packages (from arrow>=0.15.0->isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->notebook->jupyter->icn3dpy) (2.8.19.20240106)\n",
      "Requirement already satisfied: jupyterlab in /usr/local/lib/python3.11/site-packages (4.1.2)\n",
      "Requirement already satisfied: async-lru>=1.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.0.4)\n",
      "Requirement already satisfied: httpx>=0.25.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (0.27.0)\n",
      "Requirement already satisfied: ipykernel in /usr/local/lib/python3.11/site-packages (from jupyterlab) (6.29.3)\n",
      "Requirement already satisfied: jinja2>=3.0.3 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (3.1.3)\n",
      "Requirement already satisfied: jupyter-core in /usr/local/lib/python3.11/site-packages (from jupyterlab) (5.7.1)\n",
      "Requirement already satisfied: jupyter-lsp>=2.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.2.3)\n",
      "Requirement already satisfied: jupyter-server<3,>=2.4.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.12.5)\n",
      "Requirement already satisfied: jupyterlab-server<3,>=2.19.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.25.3)\n",
      "Requirement already satisfied: notebook-shim>=0.2 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (0.2.4)\n",
      "Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from jupyterlab) (23.2)\n",
      "Requirement already satisfied: tornado>=6.2.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (6.4)\n",
      "Requirement already satisfied: traitlets in /usr/local/lib/python3.11/site-packages (from jupyterlab) (5.14.1)\n",
      "Requirement already satisfied: anyio in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (4.3.0)\n",
      "Requirement already satisfied: certifi in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (2024.2.2)\n",
      "Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (1.0.4)\n",
      "Requirement already satisfied: idna in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (3.6)\n",
      "Requirement already satisfied: sniffio in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (1.3.1)\n",
      "Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.11/site-packages (from httpcore==1.*->httpx>=0.25.0->jupyterlab) (0.14.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2>=3.0.3->jupyterlab) (2.1.5)\n",
      "Requirement already satisfied: argon2-cffi in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (23.1.0)\n",
      "Requirement already satisfied: jupyter-client>=7.4.4 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (8.6.0)\n",
      "Requirement already satisfied: jupyter-events>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.9.0)\n",
      "Requirement already satisfied: jupyter-server-terminals in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.5.2)\n",
      "Requirement already satisfied: nbconvert>=6.4.4 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (7.16.1)\n",
      "Requirement already satisfied: nbformat>=5.3.0 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (5.9.2)\n",
      "Requirement already satisfied: overrides in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (7.7.0)\n",
      "Requirement already satisfied: prometheus-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.20.0)\n",
      "Requirement already satisfied: pyzmq>=24 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (25.1.2)\n",
      "Requirement already satisfied: send2trash>=1.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (1.8.2)\n",
      "Requirement already satisfied: terminado>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.18.0)\n",
      "Requirement already satisfied: websocket-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (1.7.0)\n",
      "Requirement already satisfied: platformdirs>=2.5 in /usr/local/lib/python3.11/site-packages (from jupyter-core->jupyterlab) (4.2.0)\n",
      "Requirement already satisfied: babel>=2.10 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (2.14.0)\n",
      "Requirement already satisfied: json5>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (0.9.17)\n",
      "Requirement already satisfied: jsonschema>=4.18.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (4.21.1)\n",
      "Requirement already satisfied: requests>=2.31 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (2.31.0)\n",
      "Requirement already satisfied: comm>=0.1.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (0.2.1)\n",
      "Requirement already satisfied: debugpy>=1.6.5 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (1.8.1)\n",
      "Requirement already satisfied: ipython>=7.23.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (8.22.1)\n",
      "Requirement already satisfied: matplotlib-inline>=0.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (0.1.6)\n",
      "Requirement already satisfied: nest-asyncio in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (1.6.0)\n",
      "Requirement already satisfied: psutil in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (5.9.8)\n",
      "Requirement already satisfied: decorator in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (5.1.1)\n",
      "Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (0.19.1)\n",
      "Requirement already satisfied: prompt-toolkit<3.1.0,>=3.0.41 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (3.0.43)\n",
      "Requirement already satisfied: pygments>=2.4.0 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (2.17.2)\n",
      "Requirement already satisfied: stack-data in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (0.6.3)\n",
      "Requirement already satisfied: pexpect>4.3 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (4.9.0)\n",
      "Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (23.2.0)\n",
      "Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (2023.12.1)\n",
      "Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (0.33.0)\n",
      "Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (0.18.0)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-client>=7.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (2.8.2)\n",
      "Requirement already satisfied: python-json-logger>=2.0.4 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.0.7)\n",
      "Requirement already satisfied: pyyaml>=5.3 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (6.0.1)\n",
      "Requirement already satisfied: rfc3339-validator in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (0.1.4)\n",
      "Requirement already satisfied: rfc3986-validator>=0.1.1 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (0.1.1)\n",
      "Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (4.12.3)\n",
      "Requirement already satisfied: bleach!=5.0.0 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (6.1.0)\n",
      "Requirement already satisfied: defusedxml in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.7.1)\n",
      "Requirement already satisfied: jupyterlab-pygments in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.3.0)\n",
      "Requirement already satisfied: mistune<4,>=2.0.3 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (3.0.2)\n",
      "Requirement already satisfied: nbclient>=0.5.0 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.9.0)\n",
      "Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.5.1)\n",
      "Requirement already satisfied: tinycss2 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.2.1)\n",
      "Requirement already satisfied: fastjsonschema in /usr/local/lib/python3.11/site-packages (from nbformat>=5.3.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.19.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.19.0->jupyterlab) (3.3.2)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.19.0->jupyterlab) (2.0.7)\n",
      "Requirement already satisfied: ptyprocess in /usr/local/lib/python3.11/site-packages (from terminado>=0.8.3->jupyter-server<3,>=2.4.0->jupyterlab) (0.7.0)\n",
      "Requirement already satisfied: argon2-cffi-bindings in /usr/local/lib/python3.11/site-packages (from argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (21.2.0)\n",
      "Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.16.0)\n",
      "Requirement already satisfied: webencodings in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.5.1)\n",
      "Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jedi>=0.16->ipython>=7.23.1->ipykernel->jupyterlab) (0.8.3)\n",
      "Requirement already satisfied: fqdn in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.5.1)\n",
      "Requirement already satisfied: isoduration in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (20.11.0)\n",
      "Requirement already satisfied: jsonpointer>1.13 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.4)\n",
      "Requirement already satisfied: uri-template in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.3.0)\n",
      "Requirement already satisfied: webcolors>=1.11 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.13)\n",
      "Requirement already satisfied: wcwidth in /usr/local/lib/python3.11/site-packages (from prompt-toolkit<3.1.0,>=3.0.41->ipython>=7.23.1->ipykernel->jupyterlab) (0.2.13)\n",
      "Requirement already satisfied: cffi>=1.0.1 in /usr/local/lib/python3.11/site-packages (from argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (1.16.0)\n",
      "Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.11/site-packages (from beautifulsoup4->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (2.5)\n",
      "Requirement already satisfied: executing>=1.2.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (2.0.1)\n",
      "Requirement already satisfied: asttokens>=2.1.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (2.4.1)\n",
      "Requirement already satisfied: pure-eval in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (0.2.2)\n",
      "Requirement already satisfied: pycparser in /usr/local/lib/python3.11/site-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (2.21)\n",
      "Requirement already satisfied: arrow>=0.15.0 in /usr/local/lib/python3.11/site-packages (from isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.3.0)\n",
      "Requirement already satisfied: types-python-dateutil>=2.8.10 in /usr/local/lib/python3.11/site-packages (from arrow>=0.15.0->isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.8.19.20240106)\n",
      "/usr/bin/sh: 1: jupyterhub: not found\n",
      "Requirement already satisfied: transformers in /usr/local/lib/python3.11/site-packages (4.38.1)\n",
      "Requirement already satisfied: filelock in /usr/local/lib/python3.11/site-packages (from transformers) (3.13.1)\n",
      "Requirement already satisfied: huggingface-hub<1.0,>=0.19.3 in /usr/local/lib/python3.11/site-packages (from transformers) (0.21.2)\n",
      "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.11/site-packages (from transformers) (1.26.4)\n",
      "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.11/site-packages (from transformers) (23.2)\n",
      "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.11/site-packages (from transformers) (6.0.1)\n",
      "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.11/site-packages (from transformers) (2023.12.25)\n",
      "Requirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from transformers) (2.31.0)\n",
      "Requirement already satisfied: tokenizers<0.19,>=0.14 in /usr/local/lib/python3.11/site-packages (from transformers) (0.15.2)\n",
      "Requirement already satisfied: safetensors>=0.4.1 in /usr/local/lib/python3.11/site-packages (from transformers) (0.4.2)\n",
      "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.11/site-packages (from transformers) (4.66.2)\n",
      "Requirement already satisfied: fsspec>=2023.5.0 in /usr/local/lib/python3.11/site-packages (from huggingface-hub<1.0,>=0.19.3->transformers) (2024.2.0)\n",
      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.11/site-packages (from huggingface-hub<1.0,>=0.19.3->transformers) (4.10.0)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (2.0.7)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (2024.2.2)\n"
     ]
    }
   ],
   "source": [
    "!pip install tensorflow\n",
    "!pip install ablang\n",
    "!pip install ablang2\n",
    "!pip install tensorflow\n",
    "!pip install tensorboardX\n",
    "!pip install icn3dpy\n",
    "!pip install jupyterlab\n",
    "!jupyterhub labextension install jupyterlab_3dmol\n",
    "!pip install transformers"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ee4ac2e8-099f-4886-936f-e9fb5f5d75f6",
   "metadata": {},
   "source": [
    "# General Steps \n",
    "To import and prepare your data for analysis with machine learning models, focusing on VH (variable heavy chain) and VL (variable light chain) sequences along with their corresponding labels, follow these structured steps:\n",
    "\n",
    "1. Import Libraries: Include necessary libraries for data manipulation (e.g., pandas), machine learning, and any specific libraries for handling VH and VL sequences, such as ablang and ablang2 for embedding generation.\n",
    "2. Load Your Data: Use pandas or a similar library to load your dataset from a CSV file or another data source. This dataset should include VH and VL sequences and their corresponding labels indicating antigen neutralization.\n",
    "3. Preprocess Data: Prepare the sequence data according to the input requirements of your pretrained models (ablang and ablang2). This might involve sequence cleaning, encoding, or formatting.\n",
    "4. Load Pretrained Models: Initialize ablang and ablang2 models with pretrained weights, ready to generate embeddings for your sequences.\n",
    "5. Generate Embeddings: Apply the pretrained models to your preprocessed VH and VL sequences to produce embeddings. These embeddings transform the sequence data into a numerical format suitable for machine learning analysis.\n",
    "6. Prepare Final Dataset: Combine the generated embeddings with the corresponding labels into a final dataset. This dataset will serve as the input for subsequent machine learning tasks, such as classification or clustering.\n",
    "7. Machine Learning Analysis: Use the prepared dataset to train machine learning models, evaluate their performance, and perform predictions or exploratory data analysis.\n",
    "\n",
    "## VH (Variable Heavy) and VL (Variable Light)\n",
    "Antibodies, also known as immunoglobulins, are Y-shaped molecules composed of two identical heavy chains and two identical light chains. Each chain consists of variable (V) and constant (C) regions. The variable regions of the heavy and light chains (VH and VL, respectively) are critical for the antibody's specificity because they contain the antigen-binding site. This site binds to specific parts of an antigen, called epitopes, allowing the antibody to recognize and bind to a vast array of pathogens.\n",
    "\n",
    "## VHH (Single Domain Antibodies)\n",
    "VHH refers to a special type of antibody fragment derived from heavy-chain-only antibodies (HcAbs) found in camelids (camels, llamas, and alpacas). Unlike conventional antibodies, HcAbs do not have light chains, so their antigen-binding site consists solely of the variable domain of the heavy chain (VHH). VHH antibodies, also known as nanobodies, are much smaller than regular antibodies, which allows them to reach and bind to hidden epitopes that are inaccessible to conventional antibodies. Their small size, high stability, solubility, and ease of production in microbial systems make VHH antibodies particularly attractive for research, diagnostic, and therapeutic applications.\n",
    "\n",
    "VHorVHH seems to contain both heavy and light weight variables. ablang2_model requires both VH and VL as input sequences to conduct embedding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "15e490a7-d3a8-40ca-a593-226a81a2b713",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-03-01 14:48:32.136780: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2024-03-01 14:48:32.136848: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2024-03-01 14:48:32.138236: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2024-03-01 14:48:32.145716: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2024-03-01 14:48:32.998498: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "# Core libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "# Visualization libraries\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Machine Learning / Data Processing libraries\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.metrics import classification_report, confusion_matrix, accuracy_score, f1_score\n",
    "from sklearn.model_selection import GridSearchCV, KFold, StratifiedKFold, train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.utils import class_weight\n",
    "\n",
    "# Deep Learning libraries\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import regularizers\n",
    "import keras  \n",
    "\n",
    "# Specialized libraries (assuming these are either third-party or local)\n",
    "import ablang  \n",
    "import ablang2  \n",
    "import icn3dpy  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8bb91438-f176-4bf2-a8ba-72ed62f982a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Ab or Nb</th>\n",
       "      <th>Binds to</th>\n",
       "      <th>Doesn't Bind to</th>\n",
       "      <th>Neutralising Vs</th>\n",
       "      <th>Not Neutralising Vs</th>\n",
       "      <th>Protein + Epitope</th>\n",
       "      <th>Origin</th>\n",
       "      <th>VHorVHH</th>\n",
       "      <th>VL</th>\n",
       "      <th>...</th>\n",
       "      <th>Light J Gene</th>\n",
       "      <th>CDRH3</th>\n",
       "      <th>CDRL3</th>\n",
       "      <th>Structures</th>\n",
       "      <th>ABB Homology Model (if no structure)</th>\n",
       "      <th>Sources</th>\n",
       "      <th>Date Added</th>\n",
       "      <th>Last Updated</th>\n",
       "      <th>Update Description</th>\n",
       "      <th>Notes/Following Up?</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>002-S21F2</td>\n",
       "      <td>Ab</td>\n",
       "      <td>SARS-CoV2_WT;SARS-CoV2_Alpha;SARS-CoV2_Beta;SA...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SARS-CoV2_WT;SARS-CoV2_Alpha;SARS-CoV2_Beta;SA...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S; RBD</td>\n",
       "      <td>B-cells; SARS-CoV2_WT Human Patient</td>\n",
       "      <td>ND</td>\n",
       "      <td>ND</td>\n",
       "      <td>...</td>\n",
       "      <td>ND</td>\n",
       "      <td>ARGEMTAVFGDY</td>\n",
       "      <td>QQYKILLTWT</td>\n",
       "      <td>Expected (7U0P+7UPL)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Sanjeev Kumar et al., 2022 (https://www.biorxi...</td>\n",
       "      <td>Jun 14, 2022</td>\n",
       "      <td>Jun 14, 2022</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Awaiting release of structures in the PDB to g...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0304-2F8</td>\n",
       "      <td>Ab</td>\n",
       "      <td>SARS-CoV2_WT (weak)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>S; Unk</td>\n",
       "      <td>B-cells; SARS-CoV2_WT Human Patient</td>\n",
       "      <td>EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLE...</td>\n",
       "      <td>EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPG...</td>\n",
       "      <td>...</td>\n",
       "      <td>IGKJ3 (Human)</td>\n",
       "      <td>ARRGDGLYYYGMDV</td>\n",
       "      <td>MQALQTPQT</td>\n",
       "      <td>ND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Xiangyang Chi et al., 2020 (https://science.sc...</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Complete</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0304-3H3</td>\n",
       "      <td>Ab</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S; S2</td>\n",
       "      <td>B-cells; SARS-CoV2_WT Human Patient</td>\n",
       "      <td>EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLE...</td>\n",
       "      <td>DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRL...</td>\n",
       "      <td>...</td>\n",
       "      <td>IGKJ1 (Human)</td>\n",
       "      <td>ARDRIAPVGKFFGWYFDL</td>\n",
       "      <td>QQYNKWPPWT</td>\n",
       "      <td>ND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Xiangyang Chi et al., 2020 (https://science.sc...</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Complete</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0304-4A10</td>\n",
       "      <td>Ab</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>S; S2</td>\n",
       "      <td>B-cells; SARS-CoV2_WT Human Patient</td>\n",
       "      <td>EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGLE...</td>\n",
       "      <td>EIVLTQSPDSLAVSLGERATINCRSSQSVLYSSNNKNYLAWYQQKP...</td>\n",
       "      <td>...</td>\n",
       "      <td>IGKJ3 (Human)</td>\n",
       "      <td>ARSSSRGFDY</td>\n",
       "      <td>QQYYSSPYA</td>\n",
       "      <td>ND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Xiangyang Chi et al., 2020 (https://science.sc...</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Complete</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0304-4A2</td>\n",
       "      <td>Ab</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SARS-CoV2_WT</td>\n",
       "      <td>S; S1 non-RBD</td>\n",
       "      <td>B-cells; SARS-CoV2_WT Human Patient</td>\n",
       "      <td>EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGKG...</td>\n",
       "      <td>EIVLTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKP...</td>\n",
       "      <td>...</td>\n",
       "      <td>IGKJ1 (Human)</td>\n",
       "      <td>ARELFTAVAGKGGIDY</td>\n",
       "      <td>HQYYNTPRT</td>\n",
       "      <td>ND</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Xiangyang Chi et al., 2020 (https://science.sc...</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>Jun 23, 2020</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Complete</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Name Ab or Nb                                           Binds to  \\\n",
       "0  002-S21F2       Ab  SARS-CoV2_WT;SARS-CoV2_Alpha;SARS-CoV2_Beta;SA...   \n",
       "1   0304-2F8       Ab                                SARS-CoV2_WT (weak)   \n",
       "2   0304-3H3       Ab                                       SARS-CoV2_WT   \n",
       "3  0304-4A10       Ab                                       SARS-CoV2_WT   \n",
       "4   0304-4A2       Ab                                       SARS-CoV2_WT   \n",
       "\n",
       "  Doesn't Bind to                                    Neutralising Vs  \\\n",
       "0             NaN  SARS-CoV2_WT;SARS-CoV2_Alpha;SARS-CoV2_Beta;SA...   \n",
       "1             NaN                                                NaN   \n",
       "2             NaN                                       SARS-CoV2_WT   \n",
       "3             NaN                                                NaN   \n",
       "4             NaN                                                NaN   \n",
       "\n",
       "  Not Neutralising Vs Protein + Epitope                               Origin  \\\n",
       "0                 NaN            S; RBD  B-cells; SARS-CoV2_WT Human Patient   \n",
       "1        SARS-CoV2_WT            S; Unk  B-cells; SARS-CoV2_WT Human Patient   \n",
       "2                 NaN             S; S2  B-cells; SARS-CoV2_WT Human Patient   \n",
       "3        SARS-CoV2_WT             S; S2  B-cells; SARS-CoV2_WT Human Patient   \n",
       "4        SARS-CoV2_WT     S; S1 non-RBD  B-cells; SARS-CoV2_WT Human Patient   \n",
       "\n",
       "                                             VHorVHH  \\\n",
       "0                                                 ND   \n",
       "1  EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLE...   \n",
       "2  EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLE...   \n",
       "3  EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGLE...   \n",
       "4  EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGKG...   \n",
       "\n",
       "                                                  VL  ...   Light J Gene  \\\n",
       "0                                                 ND  ...             ND   \n",
       "1  EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPG...  ...  IGKJ3 (Human)   \n",
       "2  DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRL...  ...  IGKJ1 (Human)   \n",
       "3  EIVLTQSPDSLAVSLGERATINCRSSQSVLYSSNNKNYLAWYQQKP...  ...  IGKJ3 (Human)   \n",
       "4  EIVLTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKP...  ...  IGKJ1 (Human)   \n",
       "\n",
       "                CDRH3       CDRL3            Structures  \\\n",
       "0        ARGEMTAVFGDY  QQYKILLTWT  Expected (7U0P+7UPL)   \n",
       "1      ARRGDGLYYYGMDV   MQALQTPQT                    ND   \n",
       "2  ARDRIAPVGKFFGWYFDL  QQYNKWPPWT                    ND   \n",
       "3          ARSSSRGFDY   QQYYSSPYA                    ND   \n",
       "4    ARELFTAVAGKGGIDY   HQYYNTPRT                    ND   \n",
       "\n",
       "  ABB Homology Model (if no structure)  \\\n",
       "0                                  NaN   \n",
       "1                                  NaN   \n",
       "2                                  NaN   \n",
       "3                                  NaN   \n",
       "4                                  NaN   \n",
       "\n",
       "                                             Sources    Date Added  \\\n",
       "0  Sanjeev Kumar et al., 2022 (https://www.biorxi...  Jun 14, 2022   \n",
       "1  Xiangyang Chi et al., 2020 (https://science.sc...  Jun 23, 2020   \n",
       "2  Xiangyang Chi et al., 2020 (https://science.sc...  Jun 23, 2020   \n",
       "3  Xiangyang Chi et al., 2020 (https://science.sc...  Jun 23, 2020   \n",
       "4  Xiangyang Chi et al., 2020 (https://science.sc...  Jun 23, 2020   \n",
       "\n",
       "   Last Updated Update Description  \\\n",
       "0  Jun 14, 2022                NaN   \n",
       "1  Jun 23, 2020                NaN   \n",
       "2  Jun 23, 2020                NaN   \n",
       "3  Jun 23, 2020                NaN   \n",
       "4  Jun 23, 2020                NaN   \n",
       "\n",
       "                                 Notes/Following Up?  \n",
       "0  Awaiting release of structures in the PDB to g...  \n",
       "1                                           Complete  \n",
       "2                                           Complete  \n",
       "3                                           Complete  \n",
       "4                                           Complete  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"data/covabdab_all.csv\")# 3k rows\n",
    "# df = pd.read_csv(\"data/CoV-AbDab_080224.csv\") # 12K rows\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "31ecc31d-c04f-4112-93e6-04f8c34ec371",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Lines: 3000\n",
      "['EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLEWMGIIYPGDSDTKYTPSFQGQVTISTDKSINTAYLQWSSLKASDTAMYYCARRGDGLYYYGMDVWGQGTTVTVSS', 'EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLEWIGYIYYSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARDRIAPVGKFFGWYFDLWGRGTLVTVSS']\n",
      "['EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPGQSPQLLIYLGSNRASGVPDRFSGSGSGTDFTLKISRVEAEDVGVYYCMQALQTPQTFGQGTKVDIK', 'DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRLLIYGASTRATGIPARFSGSGSGTEFTLTISSLQSEDFAVYYCQQYNKWPPWTFGQGTKVDIK']\n",
      "[0, 1]\n",
      "['', '']\n"
     ]
    }
   ],
   "source": [
    "# get the heavy chain sequence \"seqs\", the neutralising data (0 or 1) \"neutralising\", and PDB IDs \"pdbids\"\n",
    "seqsvh = []\n",
    "seqsvl = []\n",
    "neutralising = []\n",
    "pdbids = []\n",
    "\n",
    "frame_size = len(df.index)\n",
    "\n",
    "# There are some problems in some sequences,let's try the first 3000 records for now\n",
    "#MAXCNT = frame_size\n",
    "MAXCNT = len(df)\n",
    "MAXCNT = 3000\n",
    "cnt = 0\n",
    "for i in range(frame_size):\n",
    "    if (cnt < MAXCNT and str(df.iloc[i]['Origin']).lower().find('human') != -1 and df.iloc[i]['VHorVHH'] != 'ND' and df.iloc[i]['VL'] != 'ND' \n",
    "       and not pd.isnull(df.iloc[i]['VL'])):\n",
    "        seqsvh.append(df.iloc[i]['VHorVHH'])\n",
    "        seqsvl.append(df.iloc[i]['VL'])\n",
    "        # somehow the empty string became \"nan\"\n",
    "        if(str(df.iloc[i]['Neutralising Vs']) != \"nan\"):\n",
    "            neutralising.append(1)\n",
    "        else:\n",
    "            neutralising.append(0)\n",
    "            \n",
    "        if(str(df.iloc[i]['Structures']).find('PDB entry') != -1):\n",
    "            pdbids.append(str(df.iloc[i]['Structures'])[10:14])\n",
    "        else:\n",
    "            pdbids.append('')\n",
    "            \n",
    "        cnt = cnt + 1\n",
    "\n",
    "print (\"Lines:\", len(seqsvh))\n",
    "print (seqsvh[0:2])\n",
    "print (seqsvl[0:2])\n",
    "print (neutralising[0:2])\n",
    "print (pdbids[0:2])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c84e6687-70cd-4ec1-8064-14b666af8662",
   "metadata": {},
   "source": [
    "# Data exploration\n",
    "First we can compare the distibution of those data related to binidng and non binding and see if there is a different between them\n",
    "We can use simple box plot or even PC to check it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f195d6c2-5fc0-4096-b8f2-94d5414ee484",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>seqsvh</th>\n",
       "      <th>seqsvl</th>\n",
       "      <th>neutralising</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLE...</td>\n",
       "      <td>EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPG...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLE...</td>\n",
       "      <td>DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRL...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGLE...</td>\n",
       "      <td>EIVLTQSPDSLAVSLGERATINCRSSQSVLYSSNNKNYLAWYQQKP...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGKG...</td>\n",
       "      <td>EIVLTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKP...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>QVQLVQSGGGVVQPGRSLRLSCAAPGFTFSSYGMHWVRQAPGKGLE...</td>\n",
       "      <td>DIVMTQSPATLSLSPGERATLSCRASQSVSSYLAWYQQKPGQAPRL...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              seqsvh  \\\n",
       "0  EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLE...   \n",
       "1  EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLE...   \n",
       "2  EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGLE...   \n",
       "3  EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGKG...   \n",
       "4  QVQLVQSGGGVVQPGRSLRLSCAAPGFTFSSYGMHWVRQAPGKGLE...   \n",
       "\n",
       "                                              seqsvl  neutralising  \n",
       "0  EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPG...             0  \n",
       "1  DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRL...             1  \n",
       "2  EIVLTQSPDSLAVSLGERATINCRSSQSVLYSSNNKNYLAWYQQKP...             0  \n",
       "3  EIVLTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKP...             0  \n",
       "4  DIVMTQSPATLSLSPGERATLSCRASQSVSSYLAWYQQKPGQAPRL...             0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame({'seqsvh': seqsvh, 'seqsvl':seqsvl, 'neutralising': neutralising})\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1cf6bd16-4cde-48ca-9e3e-30be854283cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    3000.000000\n",
      "mean        0.236000\n",
      "std         0.424693\n",
      "min         0.000000\n",
      "25%         0.000000\n",
      "50%         0.000000\n",
      "75%         0.000000\n",
      "max         1.000000\n",
      "Name: neutralising, dtype: float64\n",
      "neutralising\n",
      "0    2292\n",
      "1     708\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjAAAAGdCAYAAAAMm0nCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAi20lEQVR4nO3deXDU9f3H8Veu3RBlEyLmqhEBRzkEwVBDVDxDwlG8mFGEIlqUqsEZTYuKIqc1irc26nhiZ0DUjloLDGaFIopBNJIRAakHFh3YUEEIh2w25Pv7w8n+XDlk093v8obnYyYz3e9+9ruffZOYZ/eAJMdxHAEAABiSnOgNAAAARIuAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDmpid5AvLS0tGjjxo1q3769kpKSEr0dAABwCBzH0Y4dO1RQUKDk5AM/z3LEBszGjRtVWFiY6G0AAIA2+Pbbb3XCCScc8PojNmDat28v6acB+Hy+mJ03FAqppqZGZWVlSktLi9l5sS9m7Q7m7A7m7A7m7I54zrmxsVGFhYXh3+MHcsQGTOvLRj6fL+YBk5GRIZ/Pxw9HnDFrdzBndzBndzBnd7gx5197+wdv4gUAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMSU30Bqw6berbCu49+D/1fTj55r6hid4CAAAxwzMwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5kQVMFVVVfrtb3+r9u3bKycnR5deeqnWrVsXsWbPnj2qqKjQcccdp2OPPVbDhw9XQ0NDxJoNGzZo6NChysjIUE5OjiZMmKDm5uaINUuWLNEZZ5whr9erk08+WbNmzWrbIwQAAEecqALm3XffVUVFhZYvXy6/369QKKSysjLt2rUrvObWW2/VP//5T7322mt69913tXHjRl1++eXh6/fu3auhQ4eqqalJH3zwgV566SXNmjVLkydPDq9Zv369hg4dqgsuuED19fW65ZZbdN111+ntt9+OwUMGAADWpUazeOHChRGXZ82apZycHNXV1encc8/V9u3b9fzzz2vOnDm68MILJUkvvviiunfvruXLl6t///6qqanRmjVr9M477yg3N1d9+vTRjBkzdPvtt2vq1KnyeDx6+umn1blzZz300EOSpO7du+v999/XI488ovLy8hg9dAAAYFVUAfNL27dvlyRlZ2dLkurq6hQKhVRaWhpe061bN5144omqra1V//79VVtbq169eik3Nze8pry8XDfeeKNWr16tvn37qra2NuIcrWtuueWWA+4lGAwqGAyGLzc2NkqSQqGQQqHQ//IwI7Sey5vsxOycbojlDNzSumeLe7eEObuDObuDObsjnnM+1HO2OWBaWlp0yy236Oyzz9Zpp50mSQoEAvJ4PMrKyopYm5ubq0AgEF7z83hpvb71uoOtaWxs1I8//qh27drts5+qqipNmzZtn+M1NTXKyMho24M8iBn9WmJ+znhasGBBorfQZn6/P9FbOCowZ3cwZ3cwZ3fEY867d+8+pHVtDpiKigp99tlnev/999t6ipiaOHGiKisrw5cbGxtVWFiosrIy+Xy+mN1PKBSS3+/X3R8nK9iSFLPzxttnU+299NY664EDByotLS3R2zliMWd3MGd3MGd3xHPOra+g/Jo2Bcz48eM1b948LV26VCeccEL4eF5enpqamrRt27aIZ2EaGhqUl5cXXrNixYqI87V+Sunna375yaWGhgb5fL79PvsiSV6vV16vd5/jaWlpcfkmDrYkKbjXTsBY/kGO158hIjFndzBndzBnd8Rjzod6vqg+heQ4jsaPH6833nhDixcvVufOnSOuLyoqUlpamhYtWhQ+tm7dOm3YsEElJSWSpJKSEq1atUqbN28Or/H7/fL5fOrRo0d4zc/P0bqm9RwAAODoFtUzMBUVFZozZ47+8Y9/qH379uH3rGRmZqpdu3bKzMzU2LFjVVlZqezsbPl8Pt18880qKSlR//79JUllZWXq0aOHRo8erZkzZyoQCGjSpEmqqKgIP4Nyww036K9//atuu+02/eEPf9DixYv16quvav78+TF++AAAwKKonoF56qmntH37dp1//vnKz88Pf73yyivhNY888oh+97vfafjw4Tr33HOVl5en119/PXx9SkqK5s2bp5SUFJWUlOj3v/+9rr76ak2fPj28pnPnzpo/f778fr9OP/10PfTQQ3ruuef4CDUAAJAU5TMwjvPrHx1OT09XdXW1qqurD7imU6dOv/qpmPPPP18rV66MZnsAAOAowb+FBAAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOQQMAAAwh4ABAADmEDAAAMAcAgYAAJhDwAAAAHMIGAAAYA4BAwAAzCFgAACAOVEHzNKlSzVs2DAVFBQoKSlJb775ZsT111xzjZKSkiK+Bg0aFLFm69atGjVqlHw+n7KysjR27Fjt3LkzYs2nn36qAQMGKD09XYWFhZo5c2b0jw4AAByRog6YXbt26fTTT1d1dfUB1wwaNEibNm0Kf7388ssR148aNUqrV6+W3+/XvHnztHTpUo0bNy58fWNjo8rKytSpUyfV1dXpgQce0NSpU/XMM89Eu10AAHAESo32BoMHD9bgwYMPusbr9SovL2+/161du1YLFy7URx99pH79+kmSnnjiCQ0ZMkQPPvigCgoKNHv2bDU1NemFF16Qx+NRz549VV9fr4cffjgidAAAwNEp6oA5FEuWLFFOTo46dOigCy+8UPfcc4+OO+44SVJtba2ysrLC8SJJpaWlSk5O1ocffqjLLrtMtbW1Ovfcc+XxeMJrysvLdf/99+uHH35Qhw4d9rnPYDCoYDAYvtzY2ChJCoVCCoVCMXtsrefyJjsxO6cbYjkDt7Tu2eLeLWHO7mDO7mDO7ojnnA/1nDEPmEGDBunyyy9X586d9dVXX+nOO+/U4MGDVVtbq5SUFAUCAeXk5ERuIjVV2dnZCgQCkqRAIKDOnTtHrMnNzQ1ft7+Aqaqq0rRp0/Y5XlNTo4yMjFg9vLAZ/Vpifs54WrBgQaK30GZ+vz/RWzgqMGd3MGd3MGd3xGPOu3fvPqR1MQ+YESNGhP93r1691Lt3b3Xt2lVLlizRRRddFOu7C5s4caIqKyvDlxsbG1VYWKiysjL5fL6Y3U8oFJLf79fdHycr2JIUs/PG22dTyxO9hai1znrgwIFKS0tL9HaOWMzZHczZHczZHfGcc+srKL8mLi8h/VyXLl3UsWNHffnll7rooouUl5enzZs3R6xpbm7W1q1bw++bycvLU0NDQ8Sa1ssHem+N1+uV1+vd53haWlpcvomDLUkK7rUTMJZ/kOP1Z4hIzNkdzNkdzNkd8ZjzoZ4v7n8PzHfffactW7YoPz9fklRSUqJt27aprq4uvGbx4sVqaWlRcXFxeM3SpUsjXgfz+/069dRT9/vyEQAAOLpEHTA7d+5UfX296uvrJUnr169XfX29NmzYoJ07d2rChAlavny5vvnmGy1atEiXXHKJTj75ZJWX//QSRvfu3TVo0CBdf/31WrFihZYtW6bx48drxIgRKigokCSNHDlSHo9HY8eO1erVq/XKK6/osccei3iJCAAAHL2iDpiPP/5Yffv2Vd++fSVJlZWV6tu3ryZPnqyUlBR9+umnuvjii3XKKado7NixKioq0nvvvRfx8s7s2bPVrVs3XXTRRRoyZIjOOeeciL/jJTMzUzU1NVq/fr2Kior0pz/9SZMnT+Yj1AAAQFIb3gNz/vnny3EO/BHit99++1fPkZ2drTlz5hx0Te/evfXee+9Fuz0AAHAU4N9CAgAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAnKgDZunSpRo2bJgKCgqUlJSkN998M+J6x3E0efJk5efnq127diotLdUXX3wRsWbr1q0aNWqUfD6fsrKyNHbsWO3cuTNizaeffqoBAwYoPT1dhYWFmjlzZvSPDgAAHJGiDphdu3bp9NNPV3V19X6vnzlzph5//HE9/fTT+vDDD3XMMceovLxce/bsCa8ZNWqUVq9eLb/fr3nz5mnp0qUaN25c+PrGxkaVlZWpU6dOqqur0wMPPKCpU6fqmWeeacNDBAAAR5rUaG8wePBgDR48eL/XOY6jRx99VJMmTdIll1wiSfrb3/6m3NxcvfnmmxoxYoTWrl2rhQsX6qOPPlK/fv0kSU888YSGDBmiBx98UAUFBZo9e7aampr0wgsvyOPxqGfPnqqvr9fDDz8cEToAAODoFHXAHMz69esVCARUWloaPpaZmani4mLV1tZqxIgRqq2tVVZWVjheJKm0tFTJycn68MMPddlll6m2tlbnnnuuPB5PeE15ebnuv/9+/fDDD+rQocM+9x0MBhUMBsOXGxsbJUmhUEihUChmj7H1XN5kJ2bndEMsZ+CW1j1b3LslzNkdzNkdzNkd8ZzzoZ4zpgETCAQkSbm5uRHHc3Nzw9cFAgHl5OREbiI1VdnZ2RFrOnfuvM85Wq/bX8BUVVVp2rRp+xyvqalRRkZGGx/Rgc3o1xLzc8bTggULEr2FNvP7/YnewlGBObuDObuDObsjHnPevXv3Ia2LacAk0sSJE1VZWRm+3NjYqMLCQpWVlcnn88XsfkKhkPx+v+7+OFnBlqSYnTfePptanugtRK111gMHDlRaWlqit3PEYs7uYM7uYM7uiOecW19B+TUxDZi8vDxJUkNDg/Lz88PHGxoa1KdPn/CazZs3R9yuublZW7duDd8+Ly9PDQ0NEWtaL7eu+SWv1yuv17vP8bS0tLh8EwdbkhTcaydgLP8gx+vPEJGYszuYszuYszviMedDPV9M/x6Yzp07Ky8vT4sWLQofa2xs1IcffqiSkhJJUklJibZt26a6urrwmsWLF6ulpUXFxcXhNUuXLo14Hczv9+vUU0/d78tHAADg6BJ1wOzcuVP19fWqr6+X9NMbd+vr67VhwwYlJSXplltu0T333KO33npLq1at0tVXX62CggJdeumlkqTu3btr0KBBuv7667VixQotW7ZM48eP14gRI1RQUCBJGjlypDwej8aOHavVq1frlVde0WOPPRbxEhEAADh6Rf0S0scff6wLLrggfLk1KsaMGaNZs2bptttu065duzRu3Dht27ZN55xzjhYuXKj09PTwbWbPnq3x48froosuUnJysoYPH67HH388fH1mZqZqampUUVGhoqIidezYUZMnT+Yj1AAAQFIbAub888+X4xz4I8RJSUmaPn26pk+ffsA12dnZmjNnzkHvp3fv3nrvvfei3R4AADgK8G8hAQAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOakJnoDAAAc7U66Y36itxAVb4qjmWcmdg88AwMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkxD5ipU6cqKSkp4qtbt27h6/fs2aOKigodd9xxOvbYYzV8+HA1NDREnGPDhg0aOnSoMjIylJOTowkTJqi5uTnWWwUAAEalxuOkPXv21DvvvPP/d5L6/3dz6623av78+XrttdeUmZmp8ePH6/LLL9eyZcskSXv37tXQoUOVl5enDz74QJs2bdLVV1+ttLQ03XvvvfHYLgAAMCYuAZOamqq8vLx9jm/fvl3PP/+85syZowsvvFCS9OKLL6p79+5avny5+vfvr5qaGq1Zs0bvvPOOcnNz1adPH82YMUO33367pk6dKo/HE48tAwAAQ+ISMF988YUKCgqUnp6ukpISVVVV6cQTT1RdXZ1CoZBKS0vDa7t166YTTzxRtbW16t+/v2pra9WrVy/l5uaG15SXl+vGG2/U6tWr1bdv3/3eZzAYVDAYDF9ubGyUJIVCIYVCoZg9ttZzeZOdmJ3TDbGcgVta92xx75YwZ3cwZ3dYnbM3xdbvlNbfgfGY86GeM+YBU1xcrFmzZunUU0/Vpk2bNG3aNA0YMECfffaZAoGAPB6PsrKyIm6Tm5urQCAgSQoEAhHx0np963UHUlVVpWnTpu1zvKamRhkZGf/jo9rXjH4tMT9nPC1YsCDRW2gzv9+f6C0cFZizO5izO6zNeeaZid5B28Rjzrt37z6kdTEPmMGDB4f/d+/evVVcXKxOnTrp1VdfVbt27WJ9d2ETJ05UZWVl+HJjY6MKCwtVVlYmn88Xs/sJhULy+/26++NkBVuSYnbeePtsanmitxC11lkPHDhQaWlpid7OEYs5u4M5u8PqnE+b+naitxAVb7KjGf1a4jLn1ldQfk1cXkL6uaysLJ1yyin68ssvNXDgQDU1NWnbtm0Rz8I0NDSE3zOTl5enFStWRJyj9VNK+3tfTSuv1yuv17vP8bS0tLh8EwdbkhTcaydgLP0g/1K8/gwRiTm7gzm7w9qcLf0++bl4zPlQzxf3vwdm586d+uqrr5Sfn6+ioiKlpaVp0aJF4evXrVunDRs2qKSkRJJUUlKiVatWafPmzeE1fr9fPp9PPXr0iPd2AQCAATF/BubPf/6zhg0bpk6dOmnjxo2aMmWKUlJSdNVVVykzM1Njx45VZWWlsrOz5fP5dPPNN6ukpET9+/eXJJWVlalHjx4aPXq0Zs6cqUAgoEmTJqmiomK/z7AAAICjT8wD5rvvvtNVV12lLVu26Pjjj9c555yj5cuX6/jjj5ckPfLII0pOTtbw4cMVDAZVXl6uJ598Mnz7lJQUzZs3TzfeeKNKSkp0zDHHaMyYMZo+fXqstwoAAIyKecDMnTv3oNenp6erurpa1dXVB1zTqVMn05+aAQAA8cW/hQQAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDkEDAAAMIeAAQAA5hAwAADAHAIGAACYQ8AAAABzCBgAAGAOAQMAAMwhYAAAgDmHdcBUV1frpJNOUnp6uoqLi7VixYpEbwkAABwGDtuAeeWVV1RZWakpU6bok08+0emnn67y8nJt3rw50VsDAAAJdtgGzMMPP6zrr79e1157rXr06KGnn35aGRkZeuGFFxK9NQAAkGCpid7A/jQ1Namurk4TJ04MH0tOTlZpaalqa2v3e5tgMKhgMBi+vH37dknS1q1bFQqFYra3UCik3bt3KzWUrL0tSTE7b7xt2bIl0VuIWuust2zZorS0tERv54jFnN3BnN1hdc6pzbsSvYWopLY42r27JS5z3rFjhyTJcZyD7yGm9xoj33//vfbu3avc3NyI47m5ufr888/3e5uqqipNmzZtn+OdO3eOyx6t6fhQoncAADiSjIzz+Xfs2KHMzMwDXn9YBkxbTJw4UZWVleHLLS0t2rp1q4477jglJcXumZLGxkYVFhbq22+/lc/ni9l5sS9m7Q7m7A7m7A7m7I54ztlxHO3YsUMFBQUHXXdYBkzHjh2VkpKihoaGiOMNDQ3Ky8vb7228Xq+8Xm/EsaysrHhtUT6fjx8OlzBrdzBndzBndzBnd8Rrzgd75qXVYfkmXo/Ho6KiIi1atCh8rKWlRYsWLVJJSUkCdwYAAA4Hh+UzMJJUWVmpMWPGqF+/fjrzzDP16KOPateuXbr22msTvTUAAJBgh23AXHnllfrvf/+ryZMnKxAIqE+fPlq4cOE+b+x1m9fr1ZQpU/Z5uQqxx6zdwZzdwZzdwZzdcTjMOcn5tc8pAQAAHGYOy/fAAAAAHAwBAwAAzCFgAACAOQQMAAAwh4DZj+rqap100klKT09XcXGxVqxYcdD1r732mrp166b09HT16tVLCxYscGmn9kUz62effVYDBgxQhw4d1KFDB5WWlv7qnw1+Eu33dKu5c+cqKSlJl156aXw3eISIds7btm1TRUWF8vPz5fV6dcopp/Dfj0MQ7ZwfffRRnXrqqWrXrp0KCwt16623as+ePS7t1qalS5dq2LBhKigoUFJSkt58881fvc2SJUt0xhlnyOv16uSTT9asWbPiu0kHEebOnet4PB7nhRdecFavXu1cf/31TlZWltPQ0LDf9cuWLXNSUlKcmTNnOmvWrHEmTZrkpKWlOatWrXJ55/ZEO+uRI0c61dXVzsqVK521a9c611xzjZOZmel89913Lu/clmjn3Gr9+vXOb37zG2fAgAHOJZdc4s5mDYt2zsFg0OnXr58zZMgQ5/3333fWr1/vLFmyxKmvr3d557ZEO+fZs2c7Xq/XmT17trN+/Xrn7bffdvLz851bb73V5Z3bsmDBAueuu+5yXn/9dUeS88Ybbxx0/ddff+1kZGQ4lZWVzpo1a5wnnnjCSUlJcRYuXBi3PRIwv3DmmWc6FRUV4ct79+51CgoKnKqqqv2uv+KKK5yhQ4dGHCsuLnb++Mc/xnWfR4JoZ/1Lzc3NTvv27Z2XXnopXls8IrRlzs3Nzc5ZZ53lPPfcc86YMWMImEMQ7Zyfeuopp0uXLk5TU5NbWzwiRDvniooK58ILL4w4VllZ6Zx99tlx3eeR5FAC5rbbbnN69uwZcezKK690ysvL47YvXkL6maamJtXV1am0tDR8LDk5WaWlpaqtrd3vbWprayPWS1J5efkB1+MnbZn1L+3evVuhUEjZ2dnx2qZ5bZ3z9OnTlZOTo7Fjx7qxTfPaMue33npLJSUlqqioUG5urk477TTde++92rt3r1vbNqctcz7rrLNUV1cXfpnp66+/1oIFCzRkyBBX9ny0SMTvwsP2b+JNhO+//1579+7d52/7zc3N1eeff77f2wQCgf2uDwQCcdvnkaAts/6l22+/XQUFBfv80OD/tWXO77//vp5//nnV19e7sMMjQ1vm/PXXX2vx4sUaNWqUFixYoC+//FI33XSTQqGQpkyZ4sa2zWnLnEeOHKnvv/9e55xzjhzHUXNzs2644Qbdeeedbmz5qHGg34WNjY368ccf1a5du5jfJ8/AwKT77rtPc+fO1RtvvKH09PREb+eIsWPHDo0ePVrPPvusOnbsmOjtHNFaWlqUk5OjZ555RkVFRbryyit111136emnn0701o4oS5Ys0b333qsnn3xSn3zyiV5//XXNnz9fM2bMSPTW8D/iGZif6dixo1JSUtTQ0BBxvKGhQXl5efu9TV5eXlTr8ZO2zLrVgw8+qPvuu0/vvPOOevfuHc9tmhftnL/66it98803GjZsWPhYS0uLJCk1NVXr1q1T165d47tpg9ry/Zyfn6+0tDSlpKSEj3Xv3l2BQEBNTU3yeDxx3bNFbZnz3XffrdGjR+u6666TJPXq1Uu7du3SuHHjdNdddyk5mf8fHwsH+l3o8/ni8uyLxDMwETwej4qKirRo0aLwsZaWFi1atEglJSX7vU1JSUnEekny+/0HXI+ftGXWkjRz5kzNmDFDCxcuVL9+/dzYqmnRzrlbt25atWqV6uvrw18XX3yxLrjgAtXX16uwsNDN7ZvRlu/ns88+W19++WU4ECXp3//+t/Lz84mXA2jLnHfv3r1PpLRGo8M/BRgzCfldGLe3Bxs1d+5cx+v1OrNmzXLWrFnjjBs3zsnKynICgYDjOI4zevRo54477givX7ZsmZOamuo8+OCDztq1a50pU6bwMepDFO2s77vvPsfj8Th///vfnU2bNoW/duzYkaiHYEK0c/4lPoV0aKKd84YNG5z27ds748ePd9atW+fMmzfPycnJce65555EPQQTop3zlClTnPbt2zsvv/yy8/XXXzs1NTVO165dnSuuuCJRD8GEHTt2OCtXrnRWrlzpSHIefvhhZ+XKlc5//vMfx3Ec54477nBGjx4dXt/6MeoJEyY4a9eudaqrq/kYdSI88cQTzoknnuh4PB7nzDPPdJYvXx6+7rzzznPGjBkTsf7VV191TjnlFMfj8Tg9e/Z05s+f7/KO7Ypm1p06dXIk7fM1ZcoU9zduTLTf0z9HwBy6aOf8wQcfOMXFxY7X63W6dOni/OUvf3Gam5td3rU90cw5FAo5U6dOdbp27eqkp6c7hYWFzk033eT88MMP7m/ckH/961/7/e9t62zHjBnjnHfeefvcpk+fPo7H43G6dOnivPjii3HdY5Lj8BwaAACwhffAAAAAcwgYAABgDgEDAADMIWAAAIA5BAwAADCHgAEAAOYQMAAAwBwCBgAAmEPAAAAAcwgYAABgDgEDAADMIWAAAIA5/wfqxV5X+65P3AAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Summary statistics for the neutralising column\n",
    "print(df['neutralising'].describe())\n",
    "\n",
    "# Count the distribution of the 'neutralising' labels\n",
    "print(df['neutralising'].value_counts())\n",
    "df['neutralising'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "680de0be-b441-42d7-9533-a91a96ac851b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<Axes: title={'center': 'seqsvh_length'}>,\n",
       "        <Axes: title={'center': 'seqsvl_length'}>]], dtype=object)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAigAAAGzCAYAAAAFROyYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAABE7UlEQVR4nO3df1xVdYL/8fcFLyDghSABGZVYrZTSNCq5mzWl/Mhhmkx2ysY1bF2dIXBGmXX8MmP+wFKzZnRyUGcaF5sfbuVs2WpmXM10S/wRjTumPSjLok0vlC6iEpcrnO8fPbjjFTD5de8BXs/Hg8fD8zmfzzmfz4fj8e35ca/FMAxDAAAAJhLg7w4AAABcioACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4CCdtuwYYMsFoveeeedTtvWJ5980vGOdaFPPvlEFotFTz/9tL+7AvQab775piwWi958880ubeMvFotFeXl5/u6G6RBQgBZs27ZNixYt8nc3APQQe/fu1aJFi1RdXe3vrnQbBBSgBdu2bdPixYv93Q0APcTevXu1ePFiAkobEFAAAIDpEFD87OzZs5o9e7auueYaBQcHKyYmRmlpaXr33Xc9dfbv36977rlHERERCg0N1be//W29/fbbzbb11ltv6dZbb1VISIiGDBmi3/72t1q0aJEsFotXPYfDobFjxyoyMlLh4eG6/vrr9fOf/1ySVFlZqT59+rR49aC8vFwWi0W/+c1vvMpdLpfy8/PVv39/hYWF6f7779cXX3zRGdOj1157TXfccYfCwsLUr18/ZWZm6siRI151pk2bpvDwcH3++eeaOHGiwsPD1b9/f/3bv/2bGhoavOqeOnVKU6dOlc1mU2RkpLKzs/U///M/slgs2rBhg2d7RUVFkr6+N9z0c6nf/e53GjJkiIKDg3Xrrbfq4MGDnTJmoCv0hHNNZ7iSMTaN5dixY5o2bZoiIyMVERGhRx55RLW1tV51v/rqK/34xz/W1VdfrX79+ul73/uePv/8c1ksFs9t4kWLFmnu3LmSpMTERM855dJn7jZv3qwbb7xRwcHBuuGGG7R9+/ZOH3930sffHejtfvSjH+kvf/mL8vLylJSUpFOnTumtt97S+++/r5tvvllvvPGGJkyYoOTkZC1cuFABAQEqLi7WuHHj9N///d+67bbbJEmHDx9Wenq6+vfvr0WLFunChQtauHChYmNjvfZ35MgRffe739XIkSNVWFio4OBgHTt2zPMXNDY2Vt/+9rf14osvauHChV5tX3jhBQUGBur73/++V/msWbN01VVXaeHChfrkk0+0atUq5eXl6YUXXujQ3Pzxj39Udna2MjIy9OSTT6q2tlZr167V2LFj9de//lXXXHONp25DQ4MyMjI0ZswYPf3009qxY4d++ctfasiQIcrJyZEkNTY26t5779WBAweUk5OjYcOG6ZVXXlF2drbXfn/4wx/qxIkTcjgc+uMf/9hi3zZu3KizZ8/qhz/8oSwWi1asWKFJkybp448/ltVq7dC4ga7QE841HXWlY2zywAMPKDExUcuWLdO7776r3//+94qJidGTTz7pqTNt2jS9+OKLmjp1qlJSUrR7925lZmZ6bWfSpEn64IMP9B//8R9auXKlrr76aklS//79PXXeeustvfTSS3r00UfVr18/PfPMM8rKylJFRYWio6M7dR66DQN+FRERYeTm5ra4rrGx0bj22muNjIwMo7Gx0VNeW1trJCYmGmlpaZ6yiRMnGiEhIcann37qKTt69KgRGBhoXPxrXrlypSHJ+OKLL1rt029/+1tDknH48GGv8qSkJGPcuHGe5eLiYkOSkZqa6tW/OXPmGIGBgUZ1dfUVzID3to4fP24YhmGcPXvWiIyMNGbMmOFVz+l0GhEREV7l2dnZhiSjsLDQq+7o0aON5ORkz/J//ud/GpKMVatWecoaGhqMcePGGZKM4uJiT3lubq7R0l+P48ePG5KM6Oho4/Tp057yV155xZBkbNmy5YrHDPhSdz7X7Nq1y5Bk7Nq164rHe2mbtoxx4cKFhiTjX/7lX7y2ef/99xvR0dGe5bKyMkOSMXv2bK9606ZNMyQZCxcu9JQ99dRTXue4i0kygoKCjGPHjnnK/ud//seQZKxevfqKx9zTcIvHzyIjI7V//36dOHGi2bpDhw7pww8/1A9+8AOdOnVKX375pb788kudP39e48eP1549e9TY2KiGhga9/vrrmjhxogYPHuxpP3z4cGVkZDTbnyS98soramxsbLFPkyZNUp8+fbyugLz33ns6evSoHnzwwWb1Z86c6XVp94477lBDQ4M+/fTTNs3FxRwOh6qrq/XQQw95xv3ll18qMDBQY8aM0a5du5q1+dGPfuS1fMcdd+jjjz/2LG/fvl1Wq1UzZszwlAUEBCg3N7fN/XvwwQd11VVXee1Lktf+ADPpCeeajrjSMV6spXPKqVOnVFNTI0meWzCPPvqoV71Zs2a1uX+pqakaMmSIZ3nkyJGy2Wy9+pxCQPGzFStW6L333tOgQYN02223adGiRZ4D8sMPP5QkZWdnq3///l4/v//97+VyuXTmzBl98cUX+uqrr3Tttdc22/7111/vtfzggw/q9ttv17/+678qNjZWkydP1osvvuj1F/Pqq6/W+PHj9eKLL3rKXnjhBfXp00eTJk1qto+LT1SSPP9w/9///V87Z+XvYx83blyzsZeUlKiqqsqrfkhIiNfl0qZ+XNyHTz/9VAMGDFBoaKhXvaFDh7a5f10xZqAr9YRzTUdc6Rgv9k1/zz/99FMFBAQoMTHRq15nnFOa9tebzyk8g+JnDzzwgO644w69/PLLKikp0VNPPaUnn3xSL730kucv8lNPPaVRo0a12D48PFwul+uK99e3b1/t2bNHu3bt0quvvqrt27frhRde0Lhx41RSUqLAwEBJ0uTJk/XII4/o0KFDGjVqlF588UWNHz/ec+/0Yk1tLmUYxhX361JNY//jH/+ouLi4Zuv79PE+dFvrQ1fpijEDXaknnGs64krHeDFf/j3nnNIcAcUEBgwYoEcffVSPPvqoqqqqdPPNN+uJJ57QypUrJUk2m02pqamttu/fv7/69u3r+R/CxcrLy5uVBQQEaPz48Ro/frx+9atfaenSpfrFL36hXbt2efYzceJE/fCHP/Rcev3ggw9UUFDQGcO9Ik2XOmNiYi479rZISEjQrl27VFtb63UV5dixY83qtvTWDtDd9eZzTdM55ZvG2BYJCQlqbGzU8ePHva4qcU7pHNzi8aOGhoZmlxRjYmIUHx8vl8ul5ORkDRkyRE8//bTOnTvXrH3Tq7yBgYHKyMjQ5s2bVVFR4Vn//vvv6/XXX/dqc/r06WbbafrfxMX/O4qMjFRGRoZefPFFPf/88woKCtLEiRPbO9Q2y8jIkM1m09KlS+V2u5utb89rzBkZGXK73Xr22Wc9ZY2NjZ5Xii8WFhYmSXyoEnoEzjW64jG2RdNzN2vWrPEqX716dbO6nFPajisofnT27FkNHDhQ//RP/6SbbrpJ4eHh2rFjhw4ePKhf/vKXCggI0O9//3tNmDBBN9xwgx555BF961vf0ueff65du3bJZrNpy5YtkqTFixdr+/btuuOOO/Too4/qwoULWr16tW644Qb97W9/8+yzsLBQe/bsUWZmphISElRVVaU1a9Zo4MCBGjt2rFf/HnzwQf3zP/+z1qxZo4yMDM9Db75gs9m0du1aTZ06VTfffLMmT56s/v37q6KiQq+++qpuv/32Nn9GwsSJE3Xbbbfppz/9qY4dO6Zhw4bpv/7rvzwn0ov/h5OcnCxJ+vGPf6yMjAwFBgZq8uTJnTdAwIc416hNY7xSycnJysrK0qpVq3Tq1CnPa8YffPCBpJbPKb/4xS80efJkWa1W3XvvvZ7gghb49yWi3s3lchlz5841brrpJqNfv35GWFiYcdNNNxlr1qzxqvfXv/7VmDRpkhEdHW0EBwcbCQkJxgMPPGDs3LnTq97u3buN5ORkIygoyPiHf/gHY926dZ7X5Zrs3LnTuO+++4z4+HgjKCjIiI+PNx566CHjgw8+aNa/mpoao2/fvoYk409/+lOz9U2vBh88eNCrvD2vBF76mvHF28rIyDAiIiKMkJAQY8iQIca0adOMd955x1MnOzvbCAsLa7bNS8duGIbxxRdfGD/4wQ+Mfv36GREREca0adOMt99+25BkPP/88556Fy5cMGbNmmX079/fsFgsnu00vWb81FNPNdufLnmtEDCL7n6u6YzXjNsyxqaxXPqKdEvnqfPnzxu5ublGVFSUER4ebkycONEoLy83JBnLly/3ar9kyRLjW9/6lhEQEOC1HUktvgKekJBgZGdnX/GYexqLYfTiJ3B6gUWLFmnx4sW9+kGrb7J582bdf//9euutt3T77bf7uztAt8S55u8OHTqk0aNH609/+pOmTJni7+50WzyDgl7lq6++8lpuaGjQ6tWrZbPZdPPNN/upVwC6q0vPKZK0atUqBQQE6M477/RDj3oOnkFBlzp37lyLD6RdrH///j57TXjWrFn66quvZLfb5XK59NJLL2nv3r1aunSp+vbt65M+AGi/r776qtkDv5eKiopSUFCQT/qzYsUKlZWV6e6771afPn302muv6bXXXtPMmTM1aNAgn/ShpyKgoEs9/fTTLX4Z2MWOHz/u9b06XWncuHH65S9/qa1bt6qurk5Dhw7V6tWrlZeX55P9A+iYF154QY888shl6+zatUt33XWXT/rzj//4j3I4HFqyZInOnTunwYMHa9GiRfrFL37hk/33ZDyDgi718ccff+NHNY8dO1YhISE+6hGA7uzkyZPNvtH8UsnJyV5fRYHuiYACAABMh4dkAQCA6XTLZ1AaGxt14sQJ9evXj48PBjqZYRg6e/as4uPjFRDQ+/4Pw/kF6DptOb90y4By4sQJno4Guthnn32mgQMH+rsbPsf5Beh6V3J+6ZYBpV+/fpK+HqDNZmvXNtxut0pKSpSeni6r1dqZ3et2mAtvvX0+ampqNGjQIM/fs96mM84v/tbbj+EmzIP55qAt55duGVCaLrvabLYOBZTQ0FDZbDZT/NL8ibnwxnx8rbfe3uiM84u/cQx/jXkw7xxcyfml991gBgAApkdAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAApkNAAQAAptPH3x2AOa10fNDqujlp1/mwJ+hNrrnmGn366afNyh999FEVFRWprq5OP/3pT/X888/L5XIpIyNDa9asUWxsrKduRUWFcnJytGvXLoWHhys7O1vLli1Tnz6c7syktXMM5xc04QoKANM4ePCgTp486flxOBySpO9///uSpDlz5mjLli3atGmTdu/erRMnTmjSpEme9g0NDcrMzFR9fb327t2r5557Ths2bNCCBQv8Mh4A7cd/KQCYRv/+/b2Wly9friFDhujb3/62zpw5o/Xr12vjxo0aN26cJKm4uFjDhw/Xvn37lJKSopKSEh09elQ7duxQbGysRo0apSVLlmjevHlatGiRgoKC/DGsHo+rIegKBBQAplRfX68//elPys/Pl8ViUVlZmdxut1JTUz11hg0bpsGDB6u0tFQpKSkqLS3ViBEjvG75ZGRkKCcnR0eOHNHo0aOb7cflcsnlcnmWa2pqJElut1tut7sLR9h1mvrtq/5bjIbL9qOz2rSVr+fBjMw2B23pBwEFgClt3rxZ1dXVmjZtmiTJ6XQqKChIkZGRXvViY2PldDo9dS4OJ03rm9a1ZNmyZVq8eHGz8pKSEoWGhnZwFP7VdIusqyW2Ur5tW+vPsrWnTXv5ah7MzCxzUFtbe8V1CSgATGn9+vWaMGGC4uPju3Q/BQUFys/P9yzX1NRo0KBBSk9Pl81m69J9dxW32y2Hw6G0tDRZrdYu31/RrmMtlufePbRT27SVr+fBjMw2B01XKK8EAQWA6Xz66afasWOHXnrpJU9ZXFyc6uvrVV1d7XUVpbKyUnFxcZ46Bw4c8NpWZWWlZ11LgoODFRwc3KzcarWa4oTeEb4ag2EJbHX/ndmmvXrC77KjzDIHbekDb/EAMJ3i4mLFxMQoMzPTU5acnCyr1aqdO3d6ysrLy1VRUSG73S5JstvtOnz4sKqqqjx1HA6HbDabkpKSfDcAAB3GFRQAptLY2Kji4mJlZ2d7fXZJRESEpk+frvz8fEVFRclms2nWrFmy2+1KSUmRJKWnpyspKUlTp07VihUr5HQ6NX/+fOXm5rZ4lQSAeRFQAJjKjh07VFFRoX/5l39ptm7lypUKCAhQVlaW1we1NQkMDNTWrVuVk5Mju92usLAwZWdnq7Cw0JdDANAJCCgATCU9PV2GYbS4LiQkREVFRSoqKmq1fUJCgrZt29ZV3QPgIzyDAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATIeAAgAATKdDAWX58uWyWCyaPXu2p6yurk65ubmKjo5WeHi4srKyVFlZ6dWuoqJCmZmZCg0NVUxMjObOnasLFy50pCsAAKAHaXdAOXjwoH77299q5MiRXuVz5szRli1btGnTJu3evVsnTpzQpEmTPOsbGhqUmZmp+vp67d27V88995w2bNigBQsWtH8UAACgR2lXQDl37pymTJmiZ599VldddZWn/MyZM1q/fr1+9atfady4cUpOTlZxcbH27t2rffv2SZJKSkp09OhR/elPf9KoUaM0YcIELVmyREVFRaqvr++cUQEAgG6tT3sa5ebmKjMzU6mpqXr88cc95WVlZXK73UpNTfWUDRs2TIMHD1ZpaalSUlJUWlqqESNGKDY21lMnIyNDOTk5OnLkiEaPHt1sfy6XSy6Xy7NcU1MjSXK73XK73e0Zgqdde9v3JC3NhcVo+Mb6PVVvPzZ667gBmEubA8rzzz+vd999VwcPHmy2zul0KigoSJGRkV7lsbGxcjqdnjoXh5Om9U3rWrJs2TItXry4WXlJSYlCQ0PbOgQvDoejQ+17kovnIvEy9bZt+6DrO2MCvfXYqK2t9XcXAKBtAeWzzz7TT37yEzkcDoWEhHRVn5opKChQfn6+Z7mmpkaDBg1Senq6bDZbu7bpdrvlcDiUlpYmq9XaWV3tllqai6Jdx1qtn3v3UF91zS96+7HRdIUSAPypTQGlrKxMVVVVuvnmmz1lDQ0N2rNnj37zm9/o9ddfV319vaqrq72uolRWViouLk6SFBcXpwMHDnhtt+ktn6Y6lwoODlZwcHCzcqvV2uF/QDpjGz3FxXNhWAIvW6836K3Hhj/H/Pnnn2vevHl67bXXVFtbq6FDh6q4uFi33HKLJMkwDC1cuFDPPvusqqurdfvtt2vt2rW69tprPds4ffq0Zs2apS1btiggIEBZWVn69a9/rfDwcH8NC0A7tOkh2fHjx+vw4cM6dOiQ5+eWW27RlClTPH+2Wq3auXOnp015ebkqKipkt9slSXa7XYcPH1ZVVZWnjsPhkM1mU1JSUicNC0B383//93+6/fbbZbVa9dprr+no0aP65S9/6fUg/ooVK/TMM89o3bp12r9/v8LCwpSRkaG6ujpPnSlTpujIkSNyOBzaunWr9uzZo5kzZ/pjSAA6oE1XUPr166cbb7zRqywsLEzR0dGe8unTpys/P19RUVGy2WyaNWuW7Ha7UlJSJEnp6elKSkrS1KlTtWLFCjmdTs2fP1+5ubktXiUB0Ds8+eSTGjRokIqLiz1liYl/fxrKMAytWrVK8+fP13333SdJ+sMf/qDY2Fht3rxZkydP1vvvv6/t27fr4MGDnqsuq1ev1ne+8x09/fTTio+P9+2gALRbu97iuZyVK1d6Lqu6XC5lZGRozZo1nvWBgYHaunWrcnJyZLfbFRYWpuzsbBUWFnZ2VwB0I//1X/+ljIwMff/739fu3bv1rW99S48++qhmzJghSTp+/LicTqfXW4IREREaM2aMSktLNXnyZJWWlioyMtITTiQpNTVVAQEB2r9/v+6///5m++2KtwT9zddvorX21t/l9t+eNm3V29/Ik8w3B23pR4cDyptvvum1HBISoqKiIhUVFbXaJiEhQdu2bevorgH0IB9//LHWrl2r/Px8/fznP9fBgwf14x//WEFBQcrOzva85dfSW4AXvyUYExPjtb5Pnz6Kioryy1uC/uarN9Fae+vvcm/8tadNe/XWN/IuZpY5aMtbgp1+BQUA2qOxsVG33HKLli5dKkkaPXq03nvvPa1bt07Z2dldtt+ueEvQ33z9Jlprb/1d7o2/9rRpq97+Rp5kvjloy1uCBBQApjBgwIBmD8oPHz5c//mf/ynp72/5VVZWasCAAZ46lZWVGjVqlKfOxQ/gS9KFCxd0+vRpv7wl6G++GkNrb/1dbt/tadNePeF32VFmmYO29IFvMwZgCrfffrvKy8u9yj744AMlJCRI+vqB2bi4OK+3BGtqarR//36vtwSrq6tVVlbmqfPGG2+osbFRY8aM8cEoAHQWrqAAMIU5c+boH//xH7V06VI98MADOnDggH73u9/pd7/7nSR5vjn98ccf17XXXqvExEQ99thjio+P18SJEyV9fcXlnnvu0YwZM7Ru3Tq53W7l5eVp8uTJvMEDdDMEFACmcOutt+rll19WQUGBCgsLlZiYqFWrVmnKlCmeOj/72c90/vx5zZw5U9XV1Ro7dqy2b9/u9cnWf/7zn5WXl6fx48d73ih85pln/DEkAB1AQAFgGt/97nf13e9+t9X1FotFhYWFl/1YgqioKG3cuLErugfAh3gGBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBYBpLFq0SBaLxetn2LBhnvV1dXXKzc1VdHS0wsPDlZWVpcrKSq9tVFRUKDMzU6GhoYqJidHcuXN14cIFXw8FQAf18XcHAOBiN9xwg3bs2OFZ7tPn76epOXPm6NVXX9WmTZsUERGhvLw8TZo0SW+//bYkqaGhQZmZmYqLi9PevXt18uRJPfzww7JarVq6dKnPxwKg/QgoAEylT58+iouLa1Z+5swZrV+/Xhs3btS4ceMkScXFxRo+fLj27dunlJQUlZSU6OjRo9qxY4diY2M1atQoLVmyRPPmzdOiRYsUFBTk6+EAaCcCCgBT+fDDDxUfH6+QkBDZ7XYtW7ZMgwcPVllZmdxut1JTUz11hw0bpsGDB6u0tFQpKSkqLS3ViBEjFBsb66mTkZGhnJwcHTlyRKNHj262P5fLJZfL5VmuqamRJLndbrnd7i4caddp6rev+m8xGi7bj85q01a+ngczMtsctKUfBBQApjFmzBht2LBB119/vU6ePKnFixfrjjvu0HvvvSen06mgoCBFRkZ6tYmNjZXT6ZQkOZ1Or3DStL5pXUuWLVumxYsXNysvKSlRaGhoJ4zKfxwOh0/2k9hK+bZtH3Rqm/by1TyYmVnmoLa29orrElDQZisdLZ9A5qRd5+OeoKeZMGGC588jR47UmDFjlJCQoBdffFF9+/btkn0WFBQoPz/fs1xTU6NBgwYpPT1dNputS/bZ1dxutxwOh9LS0mS1Wrt8f0W7jrVYnnv30E5t01a+ngczMtscNF2hvBIEFACmFRkZqeuuu07Hjh1TWlqa6uvrVV1d7XUVpbKy0vPMSlxcnA4cOOC1jaa3fFp6rkWSgoODFRwc3KzcarWa4oTeEb4ag2EJbHX/ndmmvXrC77KjzDIHbekDrxkDMK1z587po48+0oABA5ScnCyr1aqdO3d61peXl6uiokJ2u12SZLfbdfjwYVVVVXnqOBwO2Ww2JSUl+bz/ANqPKygATOPf/u3fdO+99yohIUEnTpzQwoULFRgYqIceekgRERGaPn268vPzFRUVJZvNplmzZslutyslJUWSlJ6erqSkJE2dOlUrVqyQ0+nU/PnzlZub2+JVEgDmRUABYBr/+7//q4ceekinTp1S//79NXbsWO3bt0/9+/eXJK1cuVIBAQHKysqSy+VSRkaG1qxZ42kfGBiorVu3KicnR3a7XWFhYcrOzlZhYaG/hgSgnQgoAEzj+eefv+z6kJAQFRUVqaioqNU6CQkJ2rZtW2d3DYCP8QwKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwnTYFlLVr12rkyJGy2Wyy2Wyy2+167bXXPOvr6uqUm5ur6OhohYeHKysry/NV500qKiqUmZmp0NBQxcTEaO7cubpw4ULnjAYAAPQIbQooAwcO1PLly1VWVqZ33nlH48aN03333acjR45IkubMmaMtW7Zo06ZN2r17t06cOKFJkyZ52jc0NCgzM1P19fXau3evnnvuOW3YsEELFizo3FEBAIBurU1fFnjvvfd6LT/xxBNau3at9u3bp4EDB2r9+vXauHGjxo0bJ0kqLi7W8OHDtW/fPqWkpKikpERHjx7Vjh07FBsbq1GjRmnJkiWaN2+eFi1apKCgoM4bGQAA6Lba/W3GDQ0N2rRpk86fPy+73a6ysjK53W6lpqZ66gwbNkyDBw9WaWmpUlJSVFpaqhEjRig2NtZTJyMjQzk5OTpy5IhGjx7d4r5cLpdcLpdnuaamRpLkdrvldrvb1f+mdu1t35O0NBcWo6Hd2+nuevux0VvHDcBc2hxQDh8+LLvdrrq6OoWHh+vll19WUlKSDh06pKCgIEVGRnrVj42NldPplCQ5nU6vcNK0vmlda5YtW6bFixc3Ky8pKVFoaGhbh+DF4XB0qH1PcvFcJLaj/bZtH3ReZ0ygtx4btbW1/u4CALQ9oFx//fU6dOiQzpw5o7/85S/Kzs7W7t27u6JvHgUFBcrPz/cs19TUaNCgQUpPT5fNZmvXNt1utxwOh9LS0mS1Wjurq91SS3NRtOtYm7eTe/fQzu6aX/T2Y6PpCiUA+FObA0pQUJCGDv36H6Lk5GQdPHhQv/71r/Xggw+qvr5e1dXVXldRKisrFRcXJ0mKi4vTgQMHvLbX9JZPU52WBAcHKzg4uFm51Wrt8D8gnbGNnuLiuTAsge1q35P01mOjN44ZgPl0+HNQGhsb5XK5lJycLKvVqp07d3rWlZeXq6KiQna7XZJkt9t1+PBhVVVVeeo4HA7ZbDYlJSV1tCsAAKCHaNMVlIKCAk2YMEGDBw/W2bNntXHjRr355pt6/fXXFRERoenTpys/P19RUVGy2WyaNWuW7Ha7UlJSJEnp6elKSkrS1KlTtWLFCjmdTs2fP1+5ubktXiEBAAC9U5sCSlVVlR5++GGdPHlSERERGjlypF5//XWlpaVJklauXKmAgABlZWXJ5XIpIyNDa9as8bQPDAzU1q1blZOTI7vdrrCwMGVnZ6uwsLBzRwUAALq1NgWU9evXX3Z9SEiIioqKVFRU1GqdhIQEbdu2rS27BQAAvQzfxQMAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgAIAAEyHgALAlJYvXy6LxaLZs2d7yurq6pSbm6vo6GiFh4crKytLlZWVXu0qKiqUmZmp0NBQxcTEaO7cubpw4YKPew+gowgoAEzn4MGD+u1vf6uRI0d6lc+ZM0dbtmzRpk2btHv3bp04cUKTJk3yrG9oaFBmZqbq6+u1d+9ePffcc9qwYYMWLFjg6yEA6KA+/u4AeoeVjg9aLJ+Tdp2PewKzO3funKZMmaJnn31Wjz/+uKf8zJkzWr9+vTZu3Khx48ZJkoqLizV8+HDt27dPKSkpKikp0dGjR7Vjxw7FxsZq1KhRWrJkiebNm6dFixYpKCjIX8MC0EYEFACmkpubq8zMTKWmpnoFlLKyMrndbqWmpnrKhg0bpsGDB6u0tFQpKSkqLS3ViBEjFBsb66mTkZGhnJwcHTlyRKNHj262P5fLJZfL5VmuqamRJLndbrnd7q4YYpdr6rev+m8xGi7bj85q01a+ngczMtsctKUfBBQApvH888/r3Xff1cGDB5utczqdCgoKUmRkpFd5bGysnE6np87F4aRpfdO6lixbtkyLFy9uVl5SUqLQ0ND2DMM0HA6HT/aT2Er5tm0tXzltb5v28tU8mJlZ5qC2tvaK6xJQAJjCZ599pp/85CdyOBwKCQnx2X4LCgqUn5/vWa6pqdGgQYOUnp4um83ms350JrfbLYfDobS0NFmt1i7fX9GuYy2W5949tFPbtJWv58GMzDYHTVcorwQBBYAplJWVqaqqSjfffLOnrKGhQXv27NFvfvMbvf7666qvr1d1dbXXVZTKykrFxcVJkuLi4nTgwAGv7Ta95dNU51LBwcEKDg5uVm61Wk1xQu8IX43BsAS2uv/ObNNePeF32VFmmYO29IG3eACYwvjx43X48GEdOnTI83PLLbdoypQpnj9brVbt3LnT06a8vFwVFRWy2+2SJLvdrsOHD6uqqspTx+FwyGazKSkpyedjAtB+XEEBYAr9+vXTjTfe6FUWFham6OhoT/n06dOVn5+vqKgo2Ww2zZo1S3a7XSkpKZKk9PR0JSUlaerUqVqxYoWcTqfmz5+v3NzcFq+SADAvAgqAbmPlypUKCAhQVlaWXC6XMjIytGbNGs/6wMBAbd26VTk5ObLb7QoLC1N2drYKCwv92GsA7UFAAWBab775ptdySEiIioqKVFRU1GqbhIQEbdu2rYt7BqCr8QwKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHQIKAAAwHV4zRqdZ6ej8L/kCAPROXEEBAACmQ0ABAACmwy2eXm6l4wNZjAYl6uuvP2/tG0YBAPAlrqAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADTIaAAAADT6ePvDgAA0JWKdh2TYQn0KpuTdp2feoMrxRUUAABgOgQUAABgOgQUAABgOgQUAABgOgQUAABgOgQUAABgOgQUAABgOm0KKMuWLdOtt96qfv36KSYmRhMnTlR5eblXnbq6OuXm5io6Olrh4eHKyspSZWWlV52KigplZmYqNDRUMTExmjt3ri5cuNDx0QAAgB6hTQFl9+7dys3N1b59++RwOOR2u5Wenq7z58976syZM0dbtmzRpk2btHv3bp04cUKTJk3yrG9oaFBmZqbq6+u1d+9ePffcc9qwYYMWLFjQeaMCAADdWps+SXb79u1eyxs2bFBMTIzKysp055136syZM1q/fr02btyocePGSZKKi4s1fPhw7du3TykpKSopKdHRo0e1Y8cOxcbGatSoUVqyZInmzZunRYsWKSgoqPNGBwAAuqUOfdT9mTNnJElRUVGSpLKyMrndbqWmpnrqDBs2TIMHD1ZpaalSUlJUWlqqESNGKDY21lMnIyNDOTk5OnLkiEaPHt1sPy6XSy6Xy7NcU1MjSXK73XK73e3qe1O79rbvKSxGgyxGg+fPvmbG+e/tx4Y/x7127VqtXbtWn3zyiSTphhtu0IIFCzRhwgRJX99C/ulPf6rnn39eLpdLGRkZWrNmjdf5pKKiQjk5Odq1a5fCw8OVnZ2tZcuWqU8fvtkD6E7a/Te2sbFRs2fP1u23364bb7xRkuR0OhUUFKTIyEivurGxsXI6nZ46F59MmtY3rWvJsmXLtHjx4mblJSUlCg0Nbe8QJEkOh6ND7bu7xIv+fE3dRz7f/7ZtH/h8n1eqtx4btbW1ftv3wIEDtXz5cl177bUyDEPPPfec7rvvPv31r3/VDTfcoDlz5ujVV1/Vpk2bFBERoby8PE2aNElvv/22pL/fQo6Li9PevXt18uRJPfzww7JarVq6dKnfxgWg7dodUHJzc/Xee+/prbfe6sz+tKigoED5+fme5ZqaGg0aNEjp6emy2Wzt2qbb7ZbD4VBaWpqsVmtndbXbKdp1TBajQdfUfaRPQoY0+0Itf8q9e6hf9tvbj42mK5T+cO+993otP/HEE1q7dq327dungQMHcgsZ6EXaFVDy8vK0detW7dmzRwMHDvSUx8XFqb6+XtXV1V5XUSorKxUXF+epc+DAAa/tNb3l01TnUsHBwQoODm5WbrVaO/wPSGdsozu7OJAYlkBTBRR//15667FhljE3NDRo06ZNOn/+vOx2e7e6hexvvr5N2drt4cvtvz1t2qppWy3tq7v+btvKbLes29KPNgUUwzA0a9Ysvfzyy3rzzTeVmJjotT45OVlWq1U7d+5UVlaWJKm8vFwVFRWy2+2SJLvdrieeeEJVVVWKiYmR9PWldJvNpqSkpLZ0B0APdPjwYdntdtXV1Sk8PFwvv/yykpKSdOjQoW53C9nffHWbMrGV8svdwm1Pm/Zq6fa1mW8vdwWz3LJuyy3kNgWU3Nxcbdy4Ua+88or69evn+QsfERGhvn37KiIiQtOnT1d+fr6ioqJks9k0a9Ys2e12paSkSJLS09OVlJSkqVOnasWKFXI6nZo/f75yc3NbvEqCjlvp6F1/EdG9XX/99Tp06JDOnDmjv/zlL8rOztbu3bu7bH9dcQvZ33x9m7Jo17EWyy93m7Y9bdqqaR5aun3tr1vIvma2W9ZtuYXcpoCydu1aSdJdd93lVV5cXKxp06ZJklauXKmAgABlZWV5PWXfJDAwUFu3blVOTo7sdrvCwsKUnZ2twsLCtnQFQA8VFBSkoUO//scjOTlZBw8e1K9//Ws9+OCD3e4Wsr/5agyt3Rq+3L7b06a9Wrp93d1/t21lluO5LX1o8y2ebxISEqKioiIVFRW1WichIUHbtm1ry64B9FKNjY1yuVzcQgZ6GT4YAIBpFBQUaMKECRo8eLDOnj2rjRs36s0339Trr7/OLWSglyGgADCNqqoqPfzwwzp58qQiIiI0cuRIvf7660pLS5PELWSgNyGgADCN9evXX3Y9t5CB3oOAAgDoFlp7I3FO2nU+7gl8oU3fZgwAAOALBBQAAGA6BBQAAGA6BBQAAGA6BBQAAGA6BBQAAGA6BBQAAGA6fA4KAMCDzxqBWXAFBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmA4BBQAAmE4ff3cAAAAzWen4oMXyOWnX+bgnvRtXUAAAgOlwBQWm1dr/YiT+JwMAPR1XUAAAgOkQUAAAgOkQUAAAgOnwDEo3w3MZAIDegICCbonXAAGgZ+MWDwAAMB0CCgAAMB0CCgAAMB0CCgAAMB0ekkWPwltOANAzcAUFAACYDgEFgCksW7ZMt956q/r166eYmBhNnDhR5eXlXnXq6uqUm5ur6OhohYeHKysrS5WVlV51KioqlJmZqdDQUMXExGju3Lm6cOGCL4cCoBNwiweAKezevVu5ubm69dZbdeHCBf385z9Xenq6jh49qrCwMEnSnDlz9Oqrr2rTpk2KiIhQXl6eJk2apLfffluS1NDQoMzMTMXFxWnv3r06efKkHn74YVmtVi1dutSfw0MPx2czdT4CCgBT2L59u9fyhg0bFBMTo7KyMt155506c+aM1q9fr40bN2rcuHGSpOLiYg0fPlz79u1TSkqKSkpKdPToUe3YsUOxsbEaNWqUlixZonnz5mnRokUKCgryx9AAtAMBBYApnTlzRpIUFRUlSSorK5Pb7VZqaqqnzrBhwzR48GCVlpYqJSVFpaWlGjFihGJjYz11MjIylJOToyNHjmj06NHN9uNyueRyuTzLNTU1kiS32y23290lY+tqTf1uT/8tRsNlt9md2jSVt9TO333zlY4cC12hLf0goAAwncbGRs2ePVu33367brzxRkmS0+lUUFCQIiMjverGxsbK6XR66lwcTprWN61rybJly7R48eJm5SUlJQoNDe3oUPzK4XC0uU1iK+XbtrX+hpyZ20jSNXUfdfl+2ts3X2nPsdAVamtrr7guAQWA6eTm5uq9997TW2+91eX7KigoUH5+vme5pqZGgwYNUnp6umw2W5fvvyu43W45HA6lpaXJarW2qW3RrmMtlufePbTbtWmah09ChsiwBJqqb77SkWOhKzRdobwSBBQAppKXl6etW7dqz549GjhwoKc8Li5O9fX1qq6u9rqKUllZqbi4OE+dAwcOeG2v6S2fpjqXCg4OVnBwcLNyq9VqihN6R7RnDJf+Q37xtrpjm6Z2l7Y1S998xSzHc1v6wGvGAEzBMAzl5eXp5Zdf1htvvKHERO+L5snJybJardq5c6enrLy8XBUVFbLb7ZIku92uw4cPq6qqylPH4XDIZrMpKSnJNwMB0Cm4ggLAFHJzc7Vx40a98sor6tevn+eZkYiICPXt21cRERGaPn268vPzFRUVJZvNplmzZslutyslJUWSlJ6erqSkJE2dOlUrVqyQ0+nU/PnzlZub2+JVEgDmRUABYApr166VJN11111e5cXFxZo2bZokaeXKlQoICFBWVpZcLpcyMjK0Zs0aT93AwEBt3bpVOTk5stvtCgsLU3Z2tgoLC301DACdhIACwBQMw/jGOiEhISoqKlJRUVGrdRISErRt27bO7BoAP+AZFAAAYDoEFAAAYDoEFAAAYDoEFAAAYDo8JNuDtPZtmgAAdDdcQQEAAKZDQAEAAKZDQAEAAKZDQAEAAKZDQAEAAKZDQAEAAKZDQAEAAKbT5oCyZ88e3XvvvYqPj5fFYtHmzZu91huGoQULFmjAgAHq27evUlNT9eGHH3rVOX36tKZMmSKbzabIyEhNnz5d586d69BAAABAz9HmgHL+/HnddNNNrX6b6IoVK/TMM89o3bp12r9/v8LCwpSRkaG6ujpPnSlTpujIkSNyOBzaunWr9uzZo5kzZ7Z/FAAAoEdp8yfJTpgwQRMmTGhxnWEYWrVqlebPn6/77rtPkvSHP/xBsbGx2rx5syZPnqz3339f27dv18GDB3XLLbdIklavXq3vfOc7evrppxUfH9+B4QAAgJ6gUz/q/vjx43I6nUpNTfWURUREaMyYMSotLdXkyZNVWlqqyMhITziRpNTUVAUEBGj//v26//77m23X5XLJ5XJ5lmtqaiRJbrdbbre7XX1tatfe9v5iMRq6bJtdsW0zudLfdXc9NjpLbx034GuX+3qSOWnX+bAn5tSpAcXpdEqSYmNjvcpjY2M965xOp2JiYrw70aePoqKiPHUutWzZMi1evLhZeUlJiUJDQzvUZ4fD0aH2vpbYhdu+pu6jLty6/23b1rbvKupux0Znqa2t9XcXAKB7fFlgQUGB8vPzPcs1NTUaNGiQ0tPTZbPZ2rVNt9sth8OhtLQ0Wa3WzupqlyvadazTt2kxGnRN3Uf6JGSIDEtgp2/fLHLvHnpF9brrsdFZmq5QAoA/dWpAiYuLkyRVVlZqwIABnvLKykqNGjXKU6eqqsqr3YULF3T69GlP+0sFBwcrODi4WbnVau3wPyCdsQ1f6soAYVgCe3RAaevvubsdG52lN44ZgPl0akBJTExUXFycdu7c6QkkNTU12r9/v3JyciRJdrtd1dXVKisrU3JysiTpjTfeUGNjo8aMGdOZ3QG8tHa/l3u9AGA+bQ4o586d07Fjf7/NcPz4cR06dEhRUVEaPHiwZs+erccff1zXXnutEhMT9dhjjyk+Pl4TJ06UJA0fPlz33HOPZsyYoXXr1sntdisvL0+TJ0/mDR4AACCpHQHlnXfe0d133+1Zbno2JDs7Wxs2bNDPfvYznT9/XjNnzlR1dbXGjh2r7du3KyQkxNPmz3/+s/Ly8jR+/HgFBAQoKytLzzzzTCcMBwAA9ARtDih33XWXDMNodb3FYlFhYaEKCwtbrRMVFaWNGze2ddcAAKCX4Lt4AACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAACA6RBQAJjGnj17dO+99yo+Pl4Wi0WbN2/2Wm8YhhYsWKABAwaob9++Sk1N1YcffuhV5/Tp05oyZYpsNpsiIyM1ffp0nTt3zoejANAZCCgATOP8+fO66aabVFRU1OL6FStW6JlnntG6deu0f/9+hYWFKSMjQ3V1dZ46U6ZM0ZEjR+RwOLR161bt2bNHM2fO9NUQAHSSPv7uAAA0mTBhgiZMmNDiOsMwtGrVKs2fP1/33XefJOkPf/iDYmNjtXnzZk2ePFnvv/++tm/froMHD+qWW26RJK1evVrf+c539PTTTys+Pr7Zdl0ul1wul2e5pqZGkuR2u+V2uzt7iD7R1O/29N9iNFx2m92pTVN5S+383bfLtfmmdm3RkWOhK7SlHwQUAN3C8ePH5XQ6lZqa6imLiIjQmDFjVFpaqsmTJ6u0tFSRkZGecCJJqampCggI0P79+3X//fc32+6yZcu0ePHiZuUlJSUKDQ3tmsH4iMPhaHObxFbKt237oFu2kaRr6j7q8v10Zptvatce7TkWukJtbe0V1yWgAOgWnE6nJCk2NtarPDY21rPO6XQqJibGa32fPn0UFRXlqXOpgoIC5efne5Zramo0aNAgpaeny2azdeYQfMbtdsvhcCgtLU1Wq7VNbYt2HWuxPPfuod2uTdM8fBIyRIYl0FR9u1ybb2rXFh05FrpC0xXKK0FAMamVjs5NzwBaFhwcrODg4GblVqvVFCf0jmjPGC79h/zibXXHNk3tLm1rhr611uab2rWHWY7ntvSBh2QBdAtxcXGSpMrKSq/yyspKz7q4uDhVVVV5rb9w4YJOnz7tqQOgeyCgAOgWEhMTFRcXp507d3rKampqtH//ftntdkmS3W5XdXW1ysrKPHXeeOMNNTY2asyYMT7vM4D24xYPANM4d+6cjh37+33548eP69ChQ4qKitLgwYM1e/ZsPf7447r22muVmJioxx57TPHx8Zo4caIkafjw4brnnns0Y8YMrVu3Tm63W3l5eZo8eXKLb/AAMC8CCgDTeOedd3T33Xd7lpseXs3OztaGDRv0s5/9TOfPn9fMmTNVXV2tsWPHavv27QoJCfG0+fOf/6y8vDyNHz9eAQEBysrK0jPPPOPzsQDoGAIKANO46667ZBhGq+stFosKCwtVWFjYap2oqCht3LixK7oHwId4BgUAAJgOAQUAAJgOAQUAAJgOAQUAAJgOD8mi17v0U3stRoMS9fXHUM9OH+6fTgFAL8cVFAAAYDoEFAAAYDrc4gEuo7UvbZyTdp2PewIAvQtXUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOkQUAAAgOnwUfcA0EMV7TomwxLYrJyvakB3wBUUAABgOgQUAABgOgQUAABgOgQUAABgOgQUAABgOgQUAABgOgQUAABgOnwOCtAOKx0ftLqOz5gAgI7jCgoAADAdAgoAADAdbvH4wOVuBwAAgOYIKG3UWtjguQM04RgBgI4joAAA0E30pgf0eQYFAACYDgEFAACYDgEFAACYDs+gAD7Sm+4do3Nx7KA34goKAAAwHQIKAAAwHb/e4ikqKtJTTz0lp9Opm266SatXr9Ztt93mzy61Gx/Gho7gs1M6n1nPL9yuAa6M3wLKCy+8oPz8fK1bt05jxozRqlWrlJGRofLycsXExPirWwB6AM4vgLeiXcdkWAKblZs5FPstoPzqV7/SjBkz9Mgjj0iS1q1bp1dffVX//u//rv/3//6fv7oFoAfg/AJ0jBmu9PkloNTX16usrEwFBQWesoCAAKWmpqq0tLRZfZfLJZfL5Vk+c+aMJOn06dNyu93t6oPb7VZtba1OnTolq9V6xe1c5860a39mZjEaVOuqlavhTIsJu7cx03ycOnXK5/s8e/asJMkwDJ/vuzP48vzy7J6PW103485/aLH8cueQ1n7fbW3TdH5r7Ri+3HHV2r66Y5vLzYO/+3a5Npdr56tj4XL7Wf7ywRbLWzvmL9am84vhB59//rkhydi7d69X+dy5c43bbrutWf2FCxcakvjhhx8f/nz22We+OiV0Ks4v/PBj/p8rOb90i89BKSgoUH5+vme5sbFRp0+fVnR0tCwWS7u2WVNTo0GDBumzzz6TzWbrrK52S8yFt94+H4Zh6OzZs4qPj/d3V3yiK84v/tbbj+EmzIP55qAt5xe/BJSrr75agYGBqqys9CqvrKxUXFxcs/rBwcEKDg72KouMjOyUvthsNlP80syAufDWm+cjIiLC311oNzOdX/ytNx/DF2MezDUHV3p+8cvnoAQFBSk5OVk7d+70lDU2Nmrnzp2y2+3+6BKAHoLzC9Az+O0WT35+vrKzs3XLLbfotttu06pVq3T+/HnPU/cA0F6cX4Duz28B5cEHH9QXX3yhBQsWyOl0atSoUdq+fbtiY2N9sv/g4GAtXLiw2aXd3oi58MZ8dH/+Pr/4G8fw15iH7j0HFsPopu8SAgCAHovv4gEAAKZDQAEAAKZDQAEAAKZDQAEAAKZDQAEAAKbTowLKnj17dO+99yo+Pl4Wi0WbN2/2Wv/SSy8pPT3d8xHWhw4daraNuro65ebmKjo6WuHh4crKymr2iZTdxeXmw+12a968eRoxYoTCwsIUHx+vhx9+WCdOnPDaxunTpzVlyhTZbDZFRkZq+vTpOnfunI9H0nHfdGwsWrRIw4YNU1hYmK666iqlpqZq//79XnV6ylyg+/qm49gwDC1YsEADBgxQ3759lZqaqg8//NCrzjXXXCOLxeL1s3z5ch+OomM4z3fOHNx1113NjoMf/ehHvhnAFepRAeX8+fO66aabVFRU1Or6sWPH6sknn2x1G3PmzNGWLVu0adMm7d69WydOnNCkSZO6qstd6nLzUVtbq3fffVePPfaY3n33Xb300ksqLy/X9773Pa96U6ZM0ZEjR+RwOLR161bt2bNHM2fO9NUQOs03HRvXXXedfvOb3+jw4cN66623dM011yg9PV1ffPGFp05PmQt0X990HK9YsULPPPOM1q1bp/379yssLEwZGRmqq6vzqldYWKiTJ096fmbNmuWL7ncKzvOdMweSNGPGDK/jYMWKFV3R3fbrjG8PNSNJxssvv9ziuuPHjxuSjL/+9a9e5dXV1YbVajU2bdrkKXv//fcNSUZpaWkX9rbrXW4+mhw4cMCQZHz66aeGYRjG0aNHDUnGwYMHPXVee+01w2KxGJ9//nlXdrdLXclcnDlzxpBk7NixwzCMnjsX6L4uPY4bGxuNuLg446mnnvKUVVdXG8HBwcZ//Md/eMoSEhKMlStX+rCnXYfzfPvmwDAM49vf/rbxk5/8pEv71lE96gpKR5WVlcntdis1NdVTNmzYMA0ePFilpaV+7JlvnDlzRhaLxfNFaaWlpYqMjNQtt9ziqZOamqqAgIBmtz96kvr6ev3ud79TRESEbrrpJkm9dy7QfRw/flxOp9Pr/BUREaExY8Y0O38tX75c0dHRGj16tJ566ilduHDB1931m95+nr/Yn//8Z1199dW68cYbVVBQoNraWn93yYvfPurejJxOp4KCgpp9k2lsbKycTqd/OuUjdXV1mjdvnh566CHPN146nU7FxMR41evTp4+ioqJ65Hxs3bpVkydPVm1trQYMGCCHw6Grr75aUu+bC3Q/TcfhpR/nf+n568c//rFuvvlmRUVFae/evSooKNDJkyf1q1/9yqf99ZfefJ6/2A9+8AMlJCQoPj5ef/vb3zRv3jyVl5frpZde8nfXPAgokNvt1gMPPCDDMLR27Vp/d8dv7r77bh06dEhffvmlnn32WT3wwAPav39/s2ACdGf5+fmeP48cOVJBQUH64Q9/qGXLlnXL72tB+1z8/NyIESM0YMAAjR8/Xh999JGGDBnix579Hbd4LhIXF6f6+npVV1d7lVdWViouLs4/nepiTeHk008/lcPh8Fw9kb6ej6qqKq/6Fy5c0OnTp3vkfISFhWno0KFKSUnR+vXr1adPH61fv15S75sLdD9Nx+Glb6N80/lrzJgxunDhgj755JOu7J5p9Mbz/JUYM2aMJOnYsWN+7snfEVAukpycLKvVqp07d3rKysvLVVFRIbvd7seedY2mcPLhhx9qx44dio6O9lpvt9tVXV2tsrIyT9kbb7yhxsZGz8HckzU2NsrlckliLmB+iYmJiouL8zp/1dTUaP/+/Zc9fx06dEgBAQG95kphbzvPX6mmV5EHDBjg345cpEfd4jl37pxX+jt+/LgOHTqkqKgoDR48WKdPn1ZFRYXnsz7Ky8slfZ2o4+LiFBERoenTpys/P19RUVGy2WyaNWuW7Ha7UlJS/DKmjrjcfAwYMED/9E//pHfffVdbt25VQ0OD5/5rVFSUgoKCNHz4cN1zzz2aMWOG1q1bJ7fbrby8PE2ePFnx8fH+Gla7XG4uoqOj9cQTT+h73/ueBgwYoC+//FJFRUX6/PPP9f3vf1+SetRcoPv6pnPc7Nmz9fjjj+vaa69VYmKiHnvsMcXHx2vixImSvn7Ye//+/br77rvVr18/lZaWas6cOfrnf/5nXXXVVX4aVdtwnu/4HHz00UfauHGjvvOd7yg6Olp/+9vfNGfOHN15550aOXKkX8bUIn+/RtSZdu3aZUhq9pOdnW0YhmEUFxe3uH7hwoWebXz11VfGo48+alx11VVGaGiocf/99xsnT570z4A66HLz0fT6WUs/u3bt8mzj1KlTxkMPPWSEh4cbNpvNeOSRR4yzZ8/6b1DtdLm5+Oqrr4z777/fiI+PN4KCgowBAwYY3/ve94wDBw54baOnzAW6r286xzU2NhqPPfaYERsbawQHBxvjx483ysvLPe3LysqMMWPGGBEREUZISIgxfPhwY+nSpUZdXZ2fRtR2nOc7PgcVFRXGnXfeaURFRRnBwcHG0KFDjblz5xpnzpzx36BaYDEMw+jkzAMAANAhPIMCAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABMh4ACAABM5/8DgQLAptS/PlwAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Calculate and plot the distribution of sequence lengths for VH and VL\n",
    "df['seqsvh_length'] = df['seqsvh'].apply(len)\n",
    "df['seqsvl_length'] = df['seqsvl'].apply(len)\n",
    "\n",
    "# Plotting\n",
    "df[['seqsvh_length', 'seqsvl_length']].hist(bins=30, alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f17ba992-cea8-4906-be30-0fca52a28928",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "S    0.141699\n",
      "G    0.114109\n",
      "T    0.079608\n",
      "Y    0.071577\n",
      "Q    0.043913\n",
      "V    0.081083\n",
      "I    0.037439\n",
      "L    0.066703\n",
      "D    0.040477\n",
      "K    0.043703\n",
      "A    0.055311\n",
      "W    0.032401\n",
      "M    0.027318\n",
      "E    0.034167\n",
      "P    0.033974\n",
      "R    0.039130\n",
      "F    0.026022\n",
      "C    0.016277\n",
      "N    0.022757\n",
      "H    0.008149\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Example: Frequency of amino acids in 'seqsvh'\n",
    "# This is a more complex analysis, depending on your needs. Here's a simplified approach:\n",
    "def amino_acid_frequency(sequence):\n",
    "    return pd.Series(list(sequence)).value_counts(normalize=True)\n",
    "\n",
    "# Apply the function to a small sample for demonstration\n",
    "sample_freqs_vh = df['seqsvh'].head().apply(amino_acid_frequency)\n",
    "print(sample_freqs_vh.mean())  # Averaging frequencies across the sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7d5ca60e-5762-4514-baec-f001d863e9ce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "               seqsvh_length  seqsvl_length  neutralising\n",
      "seqsvh_length       1.000000       0.120148      0.000673\n",
      "seqsvl_length       0.120148       1.000000      0.011824\n",
      "neutralising        0.000673       0.011824      1.000000\n"
     ]
    }
   ],
   "source": [
    "# Simple correlation between sequence length and neutralising status\n",
    "print(df[['seqsvh_length', 'seqsvl_length', 'neutralising']].corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "759f5105-86b1-43dd-acf7-045b715b887c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Sequence Length Distribution')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Class balance\n",
    "sns.countplot(x='neutralising', data=df)\n",
    "plt.title('Class Distribution')\n",
    "\n",
    "# Sequence length distribution\n",
    "sns.histplot(data=df, x='seqsvh_length', color='blue', kde=True, label='VH Length', alpha=0.6)\n",
    "sns.histplot(data=df, x='seqsvl_length', color='red', kde=True, label='VL Length', alpha=0.6)\n",
    "plt.legend()\n",
    "plt.title('Sequence Length Distribution')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30688b8f-460a-4825-b570-957309c2fdc9",
   "metadata": {},
   "source": [
    "# Embedding \n",
    "Loading a pretrained model to perform embedding on antibody sequences involves converting the string representation of these sequences (which consists of amino acid characters) into a numerical format that machine learning models can process. \n",
    "\n",
    "- How Embedding Works\n",
    "Pretrained Models: A pretrained model, such as ablang or ablang2 mentioned earlier, is typically a deep learning model that has been previously trained on a large dataset of antibody sequences. During its training, the model learns to capture the complex patterns and relationships between the amino acids in these sequences, effectively understanding the biological significance behind different sequence motifs and structures.\n",
    "\n",
    "- Sequence to Numerical Representation: When you pass an antibody sequence to such a pretrained model, the model processes the sequence and converts it into a high-dimensional vector (or array) of numbers. This vector is called an \"embedding\". The embedding process captures the essential features of the sequence, such as the presence of certain motifs or the overall sequence structure, in a form that is useful for further computational analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb3e53ee-d83f-448f-9116-e46f3b4afbda",
   "metadata": {},
   "source": [
    "## Ablang "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "70432404-5091-4271-9f57-40e82239a147",
   "metadata": {},
   "outputs": [],
   "source": [
    "# heavy chain sequence\n",
    "# heavy_ablang = ablang.pretrained(\"heavy\")\n",
    "# heavy_ablang.freeze()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "740a98ea-c6da-49e3-9883-7bc31a97a88b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# use the \"Seq-coding\" example to get an array for each sequence\n",
    "# seqcodings = heavy_ablang(seqs, mode='seqcoding')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6734a7e0-81d2-43e1-b408-db186d32f763",
   "metadata": {},
   "source": [
    "## Ablang2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "082c4c96-dbfe-404e-ba0f-b6b47a702b94",
   "metadata": {},
   "source": [
    "Note: VHorVHH seems contain both heavy and light weight variable. ablang2_model reuires both VH and VL as input sequence to conduct embedding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0ffa2992-3f67-4ad2-92a6-6130804c0b5b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# load Ablang2 model to create embedding the sequence\n",
    "ablang2_model = ablang2.pretrained()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "99a293bf-43b9-4d34-8551-9ecf4d4fa3ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ablang2 data format [VH sequence, VL sequence]\n",
    "# Test the model embedding\n",
    "seq1 = [\n",
    "    'EVQLLESGGEVKKPGASVKVSCRASGYTFRNYGLTWVRQAPGQGLEWMGWISAYNGNTNYAQKFQGRVTLTTDTSTSTAYMELRSLRSDDTAVYFCARDVPGHGAAFMDVWGTGTTVTVSS', # VH sequence\n",
    "    'DIQLTQSPLSLPVTLGQPASISCRSSQSLEASDTNIYLSWFQQRPGQSPRRLIYKISNRDSGVPDRFSGSGSGTHFTLRISRVEADDVAVYYCMQGTHWPPAFGQGTKVDIK' # VL sequence\n",
    "]\n",
    "seq2 = [\n",
    "    'EVQLLESGGEVKKPGASVKVSCRASGYTFRNYGLTWVRQAPGQGLEWMGWISAYNGNTNYAQKFQGRVTLTTDTSTSTAYMELRSLRSDDTAVYFCARDVPGHGAAFMDVWGTGTT',\n",
    "    'PVTLGQPASISCRSSQSLEASDTNIYLSWFQQRPGQSPRRLIYKISNRDSGVPDRFSGSGSGTHFTLRISRVEADDVAVYYCMQGTHWPPAFGQGTKVDIK'\n",
    "]\n",
    "seq3 = [\n",
    "    'EVQLLESGGEVKKPGASVKVSCRASGYTFRNYGLTWVRQAPGQGLEWMGWISAYNGNTNYAQKFQGRVTLTTDTSTSTAYMELRSLRSDDTAVYFCARDVPGHGAAFMDVWGTGTTVTVSS',\n",
    "    '' # The VL sequence is not known, so an empty string is left instead. \n",
    "]\n",
    "seq4 = [\n",
    "    '',\n",
    "    'DIQLTQSPLSLPVTLGQPASISCRSSQSLEASDTNIYLSWFQQRPGQSPRRLIYKISNRDSGVPDRFSGSGSGTHFTLRISRVEADDVAVYYCMQGTHWPPAFGQGTKVDIK'\n",
    "]\n",
    "seq5 = [\n",
    "    'EVQ***SGGEVKKPGASVKVSCRASGYTFRNYGLTWVRQAPGQGLEWMGWISAYNGNTNYAQKFQGRVTLTTDTSTSTAYMELRSLRSDDTAVYFCAR**PGHGAAFMDVWGTGTTVTVSS', # (*) is used to mask certain residues\n",
    "    'DIQLTQSPLSLPVTLGQPASISCRSS*SLEASDTNIYLSWFQQRPGQSPRRLIYKI*NRDSGVPDRFSGSGSGTHFTLRISRVEADDVAVYYCMQGTHWPPAFGQGTKVDIK'\n",
    "]\n",
    "\n",
    "all_seqs = [seq1, seq2, seq3, seq4, seq5]\n",
    "only_both_chains_seqs = [seq1, seq2, seq5]\n",
    "encoded = ablang2_model(all_seqs, mode='seqcoding')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "df4ff8c4-f5e6-4d51-97a9-ac889a619b25",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>seqsvh</th>\n",
       "      <th>seqsvl</th>\n",
       "      <th>neutralising</th>\n",
       "      <th>seqsvh_length</th>\n",
       "      <th>seqsvl_length</th>\n",
       "      <th>seq</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLE...</td>\n",
       "      <td>EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPG...</td>\n",
       "      <td>0</td>\n",
       "      <td>121</td>\n",
       "      <td>112</td>\n",
       "      <td>[EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGL...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLE...</td>\n",
       "      <td>DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRL...</td>\n",
       "      <td>1</td>\n",
       "      <td>124</td>\n",
       "      <td>108</td>\n",
       "      <td>[EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGL...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGLE...</td>\n",
       "      <td>EIVLTQSPDSLAVSLGERATINCRSSQSVLYSSNNKNYLAWYQQKP...</td>\n",
       "      <td>0</td>\n",
       "      <td>117</td>\n",
       "      <td>113</td>\n",
       "      <td>[EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGL...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGKG...</td>\n",
       "      <td>EIVLTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKP...</td>\n",
       "      <td>0</td>\n",
       "      <td>124</td>\n",
       "      <td>113</td>\n",
       "      <td>[EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGK...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>QVQLVQSGGGVVQPGRSLRLSCAAPGFTFSSYGMHWVRQAPGKGLE...</td>\n",
       "      <td>DIVMTQSPATLSLSPGERATLSCRASQSVSSYLAWYQQKPGQAPRL...</td>\n",
       "      <td>0</td>\n",
       "      <td>129</td>\n",
       "      <td>107</td>\n",
       "      <td>[QVQLVQSGGGVVQPGRSLRLSCAAPGFTFSSYGMHWVRQAPGKGL...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              seqsvh  \\\n",
       "0  EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGLE...   \n",
       "1  EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGLE...   \n",
       "2  EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGLE...   \n",
       "3  EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGKG...   \n",
       "4  QVQLVQSGGGVVQPGRSLRLSCAAPGFTFSSYGMHWVRQAPGKGLE...   \n",
       "\n",
       "                                              seqsvl  neutralising  \\\n",
       "0  EIVMTQSHTLLPVTPGEPASITCRSSQSLLHSNGYNYLDWYLQKPG...             0   \n",
       "1  DIVMTQSPATLSVSPEERATLSCRASQSVSSNLAWYQQKPGQAPRL...             1   \n",
       "2  EIVLTQSPDSLAVSLGERATINCRSSQSVLYSSNNKNYLAWYQQKP...             0   \n",
       "3  EIVLTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKP...             0   \n",
       "4  DIVMTQSPATLSLSPGERATLSCRASQSVSSYLAWYQQKPGQAPRL...             0   \n",
       "\n",
       "   seqsvh_length  seqsvl_length  \\\n",
       "0            121            112   \n",
       "1            124            108   \n",
       "2            117            113   \n",
       "3            124            113   \n",
       "4            129            107   \n",
       "\n",
       "                                                 seq  \n",
       "0  [EVQLVQSGAEVSQPGESLKISCKGSGYSFTGYWISWVRQMPGKGL...  \n",
       "1  [EVQLVESGPGLVKPSETLSLTCTASGGSISTYYWSWIRQPPGKGL...  \n",
       "2  [EVQLVESGGGLVQPGGSLRLSCAASGFTFSTYAMHWVRQAPGKGL...  \n",
       "3  [EVQLVESGPGLVKPSETLSLTCAVSGDSTSSSSSYWDWIRQPPGK...  \n",
       "4  [QVQLVQSGGGVVQPGRSLRLSCAAPGFTFSSYGMHWVRQAPGKGL...  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['seq'] = df.apply(lambda row: [row['seqsvh'], row['seqsvl']], axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3a3c9eec-abea-4edd-b44c-512da780b59f",
   "metadata": {},
   "outputs": [],
   "source": [
    "seqs = list(df['seq'])\n",
    "seqsvh_list = [[item, ''] for item in seqsvh]\n",
    "seqsvl_list = [['', item] for item in seqsvl]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "62dc0711-9ab9-4569-8853-8fc89a097b26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "non_string_values = [item for item in seqsvl if not isinstance(item, str)]\n",
    "non_string_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "78353972-9b56-4996-83bd-8c251803aed9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The output shape of the seq-codings: (3000, 480)\n",
      "[[-0.20339753  0.24752703  0.08483118 ...  0.20981445 -0.12832902\n",
      "  -0.09889554]\n",
      " [-0.32648866  0.26106842  0.184011   ...  0.1755036  -0.19497609\n",
      "  -0.1738531 ]\n",
      " [-0.30553812  0.08313523  0.10455568 ...  0.22529813 -0.13755226\n",
      "  -0.17712779]\n",
      " ...\n",
      " [-0.24408736  0.1386824  -0.0065362  ...  0.17710072 -0.11731781\n",
      "  -0.16415748]\n",
      " [-0.18310766  0.02512652  0.07194669 ...  0.12494069 -0.1937089\n",
      "  -0.16352598]\n",
      " [-0.18434102  0.04303722 -0.03622567 ...  0.2049059  -0.13858666\n",
      "  -0.19174961]]\n"
     ]
    }
   ],
   "source": [
    "# prepare the sequence in a format that can be read by ablang2 model\n",
    "seqcodings = ablang2_model(seqs, mode='seqcoding')\n",
    "print(\"The output shape of the seq-codings:\", seqcodings.shape)\n",
    "print(seqcodings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "0073532e-cc06-4d53-b3e4-c8b63bd17e50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Perform PCA with the desired number of components (let's say 2 components for visualization)\n",
    "num_components = 2\n",
    "pca = PCA(n_components=num_components)\n",
    "pca_result = pca.fit_transform(seqcodings)\n",
    "\n",
    "# Create a DataFrame to store the PCA results\n",
    "pca_df = pd.DataFrame(data=pca_result, columns=[f'PC{i + 1}' for i in range(num_components)])\n",
    "\n",
    "# # Add the neutralising columns back to the PCA DataFrame\n",
    "pca_df['neutralising'] = neutralising\n",
    "\n",
    "# # Plot the PCA results\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.scatter(pca_df['PC1'][pca_df['neutralising'] == 0], pca_df['PC2'][pca_df['neutralising'] == 0], color='r', label='Not Neutralising')\n",
    "plt.scatter(pca_df['PC1'][pca_df['neutralising'] == 1], pca_df['PC2'][pca_df['neutralising'] == 1], color='g', label='Neutralising')\n",
    "plt.title('Scatter plot of PC1 vs PC2')\n",
    "plt.xlabel('PC1')\n",
    "plt.ylabel('PC2')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "32000b25-f649-434f-ba42-0833d6205fc1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PC1</th>\n",
       "      <th>PC2</th>\n",
       "      <th>neutralising</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.511791</td>\n",
       "      <td>-0.490360</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.507727</td>\n",
       "      <td>-0.365473</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.528354</td>\n",
       "      <td>-0.282361</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.522685</td>\n",
       "      <td>-0.439243</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.521960</td>\n",
       "      <td>-0.242844</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        PC1       PC2  neutralising\n",
       "0 -0.511791 -0.490360             0\n",
       "1 -0.507727 -0.365473             1\n",
       "2 -0.528354 -0.282361             0\n",
       "3 -0.522685 -0.439243             0\n",
       "4 -0.521960 -0.242844             0"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "00fb3356-9155-43f3-bc3d-c80e2f53dd02",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Perform t-SNE\n",
    "tsne = TSNE(n_components=2, random_state=42)\n",
    "tsne_results = tsne.fit_transform(seqcodings)\n",
    "\n",
    "# Plotting\n",
    "plt.figure(figsize=(8, 5))\n",
    "plt.scatter(tsne_results[:, 0], tsne_results[:, 1], c=df['neutralising'], cmap='viridis')\n",
    "plt.colorbar(label='Neutralising')\n",
    "plt.xlabel('t-SNE 1')\n",
    "plt.ylabel('t-SNE 2')\n",
    "plt.title('t-SNE Visualization')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd218e7c-6f14-4fea-82fb-f28a3c73c9b2",
   "metadata": {},
   "source": [
    "# Modeling Strategy \n",
    "For binary classification tasks, initiating the modeling process with a simple logistic regression model before progressing to more complex architectures like neural networks (NN) or fully connected (FC) models is a practical approach.\n",
    "\n",
    "Here’s an expanded view on developing a robust machine learning model, incorporating both simple and complex methodologies:\n",
    "\n",
    "- Initial Simple Model: Starting with logistic regression is beneficial due to its simplicity, interpretability, and efficiency in training. This step allows for a preliminary assessment of the dataset's characteristics and the feasibility of achieving meaningful classification.\n",
    "- Progression to Complex Models: After evaluating the performance of the logistic regression model, transitioning to a neural network or a fully connected model can offer deeper insights and potentially improved accuracy. These complex models are capable of capturing nonlinear relationships and interactions within the data.\n",
    "\n",
    "When developing these models, consider several important factors:\n",
    "\n",
    "Data Imbalance: Acknowledge the presence of data imbalance and its potential effects on model performance.\n",
    "Balancing the Dataset: Employ appropriate techniques to balance the dataset, such as resampling methods, to ensure equitable representation of classes.\n",
    "Balancing the Dataset: Employ appropriate techniques to balance the dataset, such as resampling methods, to ensure equitable representation of classes.\n",
    "Data Division: Strategically divide the data into training, validation, and testing sets. Utilize the training and validation sets for fine-tuning the hyperparameters of the models.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "d37167b9-bee6-418b-b351-d75077949db0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set the random seeds for numpy and tensorflow to ensure reproducible results\n",
    "np.random.seed(42)\n",
    "tf.random.set_seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "ea3d2ed7-117b-4dce-a7db-23871e950326",
   "metadata": {},
   "outputs": [],
   "source": [
    "# This split ensures balanced classes in the train and test sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    seqcodings, neutralising, test_size=0.1, random_state=0, stratify=neutralising\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "427121e6-41bd-4c0b-b74d-958db4b15ac7",
   "metadata": {},
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)\n",
    "X_train = np.array(X_train)\n",
    "X_test = np.array(X_test)\n",
    "y_train = np.array(y_train)\n",
    "y_test = np.array(y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c10b4c90-abfc-4888-8518-8445848b3fff",
   "metadata": {},
   "source": [
    "### logistic Regression model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "15d7c367-ac7d-4a1a-9dc2-b0bc04ca32cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize an empty DataFrame to store simplified metrics\n",
    "metrics_df = pd.DataFrame(columns=['Model','Accuracy', 'f1_Score'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "3b38a05c-2dd3-462d-b677-b837b60ff75e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7866666666666666\n",
      "f1_score: 0.5294117647058824\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.85      0.87      0.86       229\n",
      "           1       0.55      0.51      0.53        71\n",
      "\n",
      "    accuracy                           0.79       300\n",
      "   macro avg       0.70      0.69      0.70       300\n",
      "weighted avg       0.78      0.79      0.78       300\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_41697/247066646.py:13: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  metrics_df = pd.concat([metrics_df, new_row_df], ignore_index=True)\n"
     ]
    }
   ],
   "source": [
    "# Create and train the Logistic Regression model\n",
    "model = LogisticRegression(max_iter=1000, C=1.0)\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "# Predict the labels for the test set\n",
    "y_pred = model.predict(X_test)\n",
    "\n",
    "# # Evaluate the model\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred, average='binary')  # Ensure average='binary' for binary classification\n",
    "new_row = {'Model': 'LogisticRegression', 'Accuracy': accuracy, 'f1_Score': f1}\n",
    "new_row_df = pd.DataFrame([new_row])\n",
    "metrics_df = pd.concat([metrics_df, new_row_df], ignore_index=True)\n",
    "print(f'Accuracy: {accuracy}')\n",
    "print(f'f1_score: {f1}')\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "3b97626f-866b-41ef-9b13-30e9145b4943",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>f1_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.786667</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Model  Accuracy  f1_Score\n",
       "0  LogisticRegression  0.786667  0.529412"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e2ab1db3-afca-42d4-bfc2-ec6c74c69549",
   "metadata": {},
   "source": [
    "### A NN model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "c2b2cabc-f182-4e46-a5b4-c3fdb8d44ed9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# set parameters:\n",
    "n_epochs = 20\n",
    "batch_size = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "19bd1e2f-c53a-478c-b80b-a29bd77f6fc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "ann = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Dense(units=6, activation='relu'),\n",
    "    tf.keras.layers.Dense(units=6, activation='relu'),\n",
    "    tf.keras.layers.Dense(units=1, activation='sigmoid')\n",
    "])\n",
    "# Compiling the ANN\n",
    "ann.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "53b45e60-ad01-430e-bb84-7797169042ca",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 5ms/step - loss: 0.5335 - accuracy: 0.7671 - val_loss: 0.4522 - val_accuracy: 0.7870\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.4300 - accuracy: 0.7981 - val_loss: 0.4324 - val_accuracy: 0.7926\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.4137 - accuracy: 0.8144 - val_loss: 0.4293 - val_accuracy: 0.7907\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.4121 - accuracy: 0.8171 - val_loss: 0.4216 - val_accuracy: 0.7963\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3988 - accuracy: 0.8213 - val_loss: 0.4157 - val_accuracy: 0.7907\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3937 - accuracy: 0.8222 - val_loss: 0.4150 - val_accuracy: 0.7944\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3864 - accuracy: 0.8269 - val_loss: 0.4101 - val_accuracy: 0.7889\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3798 - accuracy: 0.8264 - val_loss: 0.4081 - val_accuracy: 0.7981\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3749 - accuracy: 0.8338 - val_loss: 0.4075 - val_accuracy: 0.7981\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3672 - accuracy: 0.8394 - val_loss: 0.4129 - val_accuracy: 0.7889\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3633 - accuracy: 0.8380 - val_loss: 0.4114 - val_accuracy: 0.7907\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3582 - accuracy: 0.8380 - val_loss: 0.4072 - val_accuracy: 0.7944\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3532 - accuracy: 0.8486 - val_loss: 0.4073 - val_accuracy: 0.7944\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3474 - accuracy: 0.8481 - val_loss: 0.4061 - val_accuracy: 0.7981\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3477 - accuracy: 0.8477 - val_loss: 0.4114 - val_accuracy: 0.8019\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3408 - accuracy: 0.8542 - val_loss: 0.4077 - val_accuracy: 0.8019\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3347 - accuracy: 0.8560 - val_loss: 0.4081 - val_accuracy: 0.8074\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3317 - accuracy: 0.8602 - val_loss: 0.4094 - val_accuracy: 0.8000\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3253 - accuracy: 0.8569 - val_loss: 0.4070 - val_accuracy: 0.8000\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.3237 - accuracy: 0.8634 - val_loss: 0.4128 - val_accuracy: 0.7981\n"
     ]
    }
   ],
   "source": [
    "# Training the ANN on the Training set\n",
    "history = ann.fit(X_train, y_train, batch_size=batch_size, epochs=n_epochs, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "2cf0283f-520d-4369-9122-d3e2c6f4103b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Extracting loss and validation loss from the history object\n",
    "training_loss = history.history['loss']\n",
    "validation_loss = history.history['val_loss']\n",
    "epochs = range(1, len(training_loss) + 1)\n",
    "\n",
    "# Plotting training and validation loss\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(epochs, training_loss, 'bo-', label='Training Loss')\n",
    "plt.plot(epochs, validation_loss, 'ro-', label='Validation Loss')\n",
    "plt.title('Training vs. Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc6becc2-3432-4849-a73e-065789c5e2ff",
   "metadata": {},
   "source": [
    "### How should we interpert this plot?\n",
    "\n",
    "1. Training Loss vs. Validation Loss:\n",
    "- The training loss measures how well the model is fitting the training data. As the training progresses (more epochs), the model learns and the training loss typically decreases.\n",
    "- The validation loss measures how well the model performs on a separate set of data not seen during training, giving an indication of how well the model generalizes.\n",
    "\n",
    "2. Interpretation of the Loss Trends:\n",
    "- Initially, both training and validation losses decrease, which is good—it means the model is learning and generalizing well.\n",
    "- However, after a certain number of epochs, the validation loss stops decreasing and starts to fluctuate or even increase slightly while the training loss continues to decrease. This is a classic sign of overfitting.\n",
    "\n",
    "3. Overfitting:\n",
    "- Overfitting occurs when the model learns the training data too well, including noise and details that do not generalize to new data. It's like memorizing the answers instead of understanding the subject.\n",
    "- In the plot, overfitting is suggested because the model's loss on the validation set is no longer decreasing alongside the training loss. The model is improving on the training data but not on the validation data.\n",
    "\n",
    "### Improving the Model:\n",
    "- Early Stopping: Implement early stopping to terminate the training process when the validation loss stops decreasing, preventing overfitting by not allowing the model to train excessively.\n",
    "- Regularization: Introduce regularization methods like L1 or L2 regularization to constrain the magnitude of the model weights, which can discourage overfitting by imposing complexity penalties on the model's loss function.\n",
    "- Dropout: Add dropout layers to the neural network architecture to introduce regularization by randomly setting a fraction of the input units to 0 at each update during training, which can help prevent overfitting.\n",
    "- Data Augmentation: Employ data augmentation techniques to synthetically expand the training dataset by applying random, yet realistic, transformations to the training data. This can lead to a model that generalizes better.\n",
    "- Reduce Model Complexity: Simplify the model's architecture by reducing the number of layers or the number of neurons in each layer if the model is too complicated and overfitting the training data.\n",
    "- Adjust Learning Rate and Batch Size: Fine-tune the learning rate and batch size to control the training dynamics. A learning rate that's too high can cause the model to converge too quickly to suboptimal solutions, whereas one that's too low can slow down the training process. Similarly, an appropriate batch size can balance between computational efficiency and the regularizing effect of stochasticity in the gradient descent.\n",
    "- Perform Hyperparameter Tuning: Run systematic experiments to find the best model parameters, including the learning rate, batch size, and other model-specific hyperparameters. This can be done using techniques like grid search or random search, or more sophisticated methods like Bayesian optimization. By tuning hyperparameters, you can potentially improve the model's performance and robustness.\n",
    "    \n",
    "For beginners, it's important to monitor both training and validation loss and look for signs of overfitting. Interventions like those mentioned above can help ensure that your model learns to generalize well, rather than just memorizing the training dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "520e99c5-820b-40af-a307-6cb5177a31eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 778us/step\n",
      "Accuracy: 0.82\n",
      "f1_score: 0.6197183098591549\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.88      0.88      0.88       229\n",
      "           1       0.62      0.62      0.62        71\n",
      "\n",
      "    accuracy                           0.82       300\n",
      "   macro avg       0.75      0.75      0.75       300\n",
      "weighted avg       0.82      0.82      0.82       300\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>f1_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.786667</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>nn_shallow</td>\n",
       "      <td>0.820000</td>\n",
       "      <td>0.619718</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Model  Accuracy  f1_Score\n",
       "0  LogisticRegression  0.786667  0.529412\n",
       "1          nn_shallow  0.820000  0.619718"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predict the labels for the test set\n",
    "predictions_prob = ann.predict(X_test)\n",
    "\n",
    "# Assuming `predictions` are the output probabilities from your model\n",
    "y_pred = (predictions_prob > 0.5).astype(int)\n",
    "\n",
    "# Evaluate the model\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred, average='binary')  # Ensure average='binary' for binary classification\n",
    "new_row = {'Model': 'nn_shallow', 'Accuracy': accuracy, 'f1_Score': f1}\n",
    "new_row_df = pd.DataFrame([new_row])\n",
    "metrics_df = pd.concat([metrics_df, new_row_df], ignore_index=True)\n",
    "print(f'Accuracy: {accuracy}')\n",
    "print(f'f1_score: {f1}')\n",
    "print(classification_report(y_test, y_pred))\n",
    "metrics_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01e5df4e-45d2-477f-a178-a052f6c503f7",
   "metadata": {},
   "source": [
    "The original model is a simple neural network with two hidden layers, each consisting of 6 units. This design may be too simplistic, especially if the input dimensionality is high (e.g., 480 features). With only 6 units in each layer, the model may not have enough capacity to learn complex patterns in the data, which could lead to underfitting.\n",
    "\n",
    "Increased Layer Units:\n",
    "- The proposed model begins with a significantly larger number of units (240) in the first hidden layer. Starting with more units can help the model capture a wide variety of patterns and relationships in the data, given the high dimensionality of the inputs.\n",
    "-mGradually decreasing the number of units in subsequent layers (from 240 to 120, then 60, and finally 30) before reaching the output layer allows the model to refine and distill these patterns into more abstract representations. This approach is akin to funneling the information in a controlled manner, which can be particularly effective in learning from a large number of features.\n",
    "\n",
    "Regularization:\n",
    "- Regularization is applied to each layer in the form of L2 regularization with a factor of 0.01. L2 regularization penalizes the weights' magnitudes by adding a term to the loss function proportional to the sum of the squared weights. This encourages the model to keep the weights small, which can help prevent overfitting by discouraging the model from relying too heavily on any single feature or combination of features.\n",
    "- Regularization can be crucial when increasing the number of units in the layers as it helps to maintain the generalization ability of the model despite the added complexity.\n",
    "\n",
    "Learning rate:\n",
    "- Adjusting the learning rate to 0.001 for better performance. A smaller learning rate can make the optimization process more gradual and precise, potentially leading to better convergence on a minimum of the loss function.\n",
    "\n",
    "\n",
    "While the proposed model has the potential to learn complex patterns better, it is also more prone to overfitting due to its increased capacity. Therefore, it is essential to monitor the model's performance on both training and validation datasets and consider implementing additional strategies such as dropout, early stopping, or further hyperparameter tuning to find the optimal model configuration."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "15afbb37-c06c-44e6-ae82-068e35db9a5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the model architecture\n",
    "ann = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Dense(units=240, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=120, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=60, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=30, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=1, activation='sigmoid')\n",
    "])\n",
    "# Compiling the ANN\n",
    "ann.compile(optimizer = keras.optimizers.Adam(learning_rate=0.001), loss = 'binary_crossentropy', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "2ca22b1e-9fd4-4e2a-a1da-dd86f585b5cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 6ms/step - loss: 4.7357 - accuracy: 0.7852 - val_loss: 3.1627 - val_accuracy: 0.7926\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 4ms/step - loss: 2.2711 - accuracy: 0.8171 - val_loss: 1.5972 - val_accuracy: 0.8056\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.2488 - accuracy: 0.8134 - val_loss: 0.9872 - val_accuracy: 0.7963\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.8312 - accuracy: 0.8199 - val_loss: 0.7322 - val_accuracy: 0.8000\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6470 - accuracy: 0.8176 - val_loss: 0.6155 - val_accuracy: 0.8000\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5694 - accuracy: 0.8259 - val_loss: 0.5651 - val_accuracy: 0.8111\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5289 - accuracy: 0.8241 - val_loss: 0.5233 - val_accuracy: 0.8000\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5012 - accuracy: 0.8139 - val_loss: 0.5015 - val_accuracy: 0.8037\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4981 - accuracy: 0.8074 - val_loss: 0.5004 - val_accuracy: 0.7889\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4744 - accuracy: 0.8227 - val_loss: 0.4989 - val_accuracy: 0.8000\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4723 - accuracy: 0.8292 - val_loss: 0.4839 - val_accuracy: 0.8000\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4610 - accuracy: 0.8218 - val_loss: 0.4754 - val_accuracy: 0.8000\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4537 - accuracy: 0.8278 - val_loss: 0.4804 - val_accuracy: 0.7926\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4476 - accuracy: 0.8245 - val_loss: 0.4781 - val_accuracy: 0.7759\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4510 - accuracy: 0.8204 - val_loss: 0.4810 - val_accuracy: 0.7870\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4398 - accuracy: 0.8208 - val_loss: 0.4673 - val_accuracy: 0.8056\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4335 - accuracy: 0.8407 - val_loss: 0.4673 - val_accuracy: 0.8130\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4355 - accuracy: 0.8366 - val_loss: 0.4826 - val_accuracy: 0.7981\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4324 - accuracy: 0.8356 - val_loss: 0.4815 - val_accuracy: 0.7796\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4369 - accuracy: 0.8315 - val_loss: 0.4878 - val_accuracy: 0.8037\n"
     ]
    }
   ],
   "source": [
    "# Training the ANN on the Training set\n",
    "history = ann.fit(X_train, y_train, batch_size=batch_size, epochs=n_epochs, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "bbc55ad5-e279-42d1-b07f-eb2be21ce864",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Extracting loss and validation loss from the history object\n",
    "training_loss = history.history['loss']\n",
    "validation_loss = history.history['val_loss']\n",
    "epochs = range(1, len(training_loss) + 1)\n",
    "\n",
    "# Plotting training and validation loss\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(epochs, training_loss, 'bo-', label='Training Loss')\n",
    "plt.plot(epochs, validation_loss, 'ro-', label='Validation Loss')\n",
    "plt.title('Training vs. Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "ddfe486a-eb8c-47cc-aa83-3cb63a61c8bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 1ms/step\n",
      "Accuracy: 0.8\n",
      "f1_score: 0.5384615384615384\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.85      0.90      0.87       229\n",
      "           1       0.59      0.49      0.54        71\n",
      "\n",
      "    accuracy                           0.80       300\n",
      "   macro avg       0.72      0.69      0.71       300\n",
      "weighted avg       0.79      0.80      0.79       300\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>f1_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.786667</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>nn_shallow</td>\n",
       "      <td>0.820000</td>\n",
       "      <td>0.619718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>nn_deep</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.538462</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Model  Accuracy  f1_Score\n",
       "0  LogisticRegression  0.786667  0.529412\n",
       "1          nn_shallow  0.820000  0.619718\n",
       "2             nn_deep  0.800000  0.538462"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predict the labels for the test set\n",
    "predictions_prob = ann.predict(X_test)\n",
    "\n",
    "# Assuming `predictions` are the output probabilities from your model\n",
    "y_pred = (predictions_prob > 0.5).astype(int)\n",
    "\n",
    "# Evaluate the model\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred, average='binary')  # Ensure average='binary' for binary classification\n",
    "new_row = {'Model': 'nn_deep', 'Accuracy': accuracy, 'f1_Score': f1}\n",
    "new_row_df = pd.DataFrame([new_row])\n",
    "metrics_df = pd.concat([metrics_df, new_row_df], ignore_index=True)\n",
    "print(f'Accuracy: {accuracy}')\n",
    "print(f'f1_score: {f1}')\n",
    "print(classification_report(y_test, y_pred))\n",
    "metrics_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7201898e-63c1-479e-adaf-720e0e178337",
   "metadata": {},
   "source": [
    "### balancing data using weight adjustment "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "bd6ba3b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# let's improve the model by adjusting the weight of low sample data.\n",
    "# Calculate class weights\n",
    "weights = class_weight.compute_class_weight(\n",
    "    'balanced',\n",
    "    classes=np.unique(y_train),\n",
    "    y=y_train\n",
    ")\n",
    "class_weights = dict(enumerate(weights))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1adf7429-fd1d-4c52-9126-3fe8274a0e28",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 0.6543868153174988, 1: 2.119309262166405}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "7b392966-f7b9-4123-92bb-55d057345565",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the model architecture\n",
    "ann = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Dense(units=240, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=120, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=60, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=30, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "    tf.keras.layers.Dense(units=1, activation='sigmoid')\n",
    "])\n",
    "# Compiling the ANN\n",
    "ann.compile(optimizer = keras.optimizers.Adam(learning_rate=0.001), loss = 'binary_crossentropy', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "7d6a3c08-9d6b-46a2-98ec-d754416c3acf",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 6ms/step - loss: 4.6669 - accuracy: 0.7190 - val_loss: 3.0690 - val_accuracy: 0.7463\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 2.1629 - accuracy: 0.7792 - val_loss: 1.6256 - val_accuracy: 0.7259\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.2393 - accuracy: 0.7537 - val_loss: 0.9807 - val_accuracy: 0.7963\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.8861 - accuracy: 0.7769 - val_loss: 0.7757 - val_accuracy: 0.7815\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.7229 - accuracy: 0.7884 - val_loss: 0.7156 - val_accuracy: 0.7815\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6537 - accuracy: 0.7875 - val_loss: 0.7354 - val_accuracy: 0.7333\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6234 - accuracy: 0.7940 - val_loss: 0.6242 - val_accuracy: 0.7833\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5949 - accuracy: 0.7875 - val_loss: 0.5814 - val_accuracy: 0.8000\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6009 - accuracy: 0.7745 - val_loss: 0.6287 - val_accuracy: 0.7556\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5710 - accuracy: 0.7870 - val_loss: 0.6055 - val_accuracy: 0.7315\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5594 - accuracy: 0.7764 - val_loss: 0.5762 - val_accuracy: 0.7833\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5520 - accuracy: 0.7870 - val_loss: 0.5831 - val_accuracy: 0.7500\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5500 - accuracy: 0.7935 - val_loss: 0.5792 - val_accuracy: 0.7407\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5412 - accuracy: 0.7847 - val_loss: 0.6322 - val_accuracy: 0.7333\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5361 - accuracy: 0.7949 - val_loss: 0.5950 - val_accuracy: 0.7574\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5328 - accuracy: 0.7968 - val_loss: 0.6223 - val_accuracy: 0.7778\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5301 - accuracy: 0.7926 - val_loss: 0.6025 - val_accuracy: 0.7426\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5216 - accuracy: 0.8009 - val_loss: 0.5786 - val_accuracy: 0.7370\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5170 - accuracy: 0.8051 - val_loss: 0.5840 - val_accuracy: 0.7759\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5106 - accuracy: 0.7977 - val_loss: 0.5477 - val_accuracy: 0.7741\n"
     ]
    }
   ],
   "source": [
    "# Training the ANN on the Training set with class weight adjustment\n",
    "history = ann.fit(X_train, y_train, batch_size=batch_size, epochs=n_epochs, validation_split=0.2, class_weight=class_weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "ffd32a64-b0e1-4492-a5a8-5e878b70b115",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0EAAAIjCAYAAADFthA8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAABxHElEQVR4nO3deZxN9ePH8fedfcZsjG0wjH2NJEoSoVAJQyTfok0LRav82lTfvtq+pVXxLdpo0ZBKiTJIKhJJkmWYse/GOsad8/vjdK+5Zrtz95n7ej4e53HPPfecz/3c47pz3/ezHIthGIYAAAAAIEiE+LsCAAAAAOBLhCAAAAAAQYUQBAAAACCoEIIAAAAABBVCEAAAAICgQggCAAAAEFQIQQAAAACCCiEIAAAAQFAhBAEAAAAIKoQgAPCx4cOHKzU11aVjx48fL4vF4tkKBZGMjAxZLBZlZGTYtzn777FlyxZZLBZNmzbNo3VKTU3V8OHDPVomAKBkhCAA+IfFYnFqKfgFGt7TunVr1a1bV4ZhFLtPp06dVKNGDZ0+fdqHNSu7H3/8UePHj9ehQ4f8XRW7adOmyWKxaMWKFf6uCgD4XJi/KwAAgeL99993uP/ee+9p/vz5hbY3b97creeZMmWK8vPzXTr2kUce0UMPPeTW85cXQ4cO1UMPPaQlS5bokksuKfT4li1btGzZMo0aNUphYa7/OXPn38NZP/74o5544gkNHz5ciYmJDo+tX79eISH8JgkAvkQIAoB//Otf/3K4/9NPP2n+/PmFtp/t+PHjiomJcfp5wsPDXaqfJIWFhbn1hb88ue666zRu3DhNnz69yBA0Y8YMGYahoUOHuvU87vx7eEJkZKRfnx8AghE/PQFAGXTt2lWtWrXSr7/+qksuuUQxMTH6v//7P0nS559/riuvvFK1atVSZGSkGjZsqKeeekpWq9WhjLPHoNjGmrzwwguaPHmyGjZsqMjISLVv317Lly93OLaoMUEWi0WjRo3S7Nmz1apVK0VGRqply5b65ptvCtU/IyND559/vqKiotSwYUO99dZbTo0zGjVqlGJjY3X8+PFCjw0ZMkQ1a9a0v84VK1aoZ8+eqlq1qqKjo1W/fn3ddNNNJZZflJSUFF1yySWaOXOm8vLyCj0+ffp0NWzYUBdccIG2bt2qO++8U02bNlV0dLSSkpJ0zTXXaMuWLaU+T1Fjgg4dOqThw4crISFBiYmJGjZsWJFd2X7//XcNHz5cDRo0UFRUlGrWrKmbbrpJ+/fvt+8zfvx4PfDAA5Kk+vXr27tV2upW1JigzZs365prrlGVKlUUExOjCy+8UF999ZXDPrbxTZ988omefvpp1alTR1FRUerevbs2btxY6ut21m+//abevXsrPj5esbGx6t69u3766SeHffLy8vTEE0+ocePGioqKUlJSki6++GLNnz/fvs+uXbt04403qk6dOoqMjFRycrL69u3r1L8RAHhacPycCAAetH//fvXu3VvXXnut/vWvf6lGjRqSzDEWsbGxuvfeexUbG6vvv/9ejz32mHJycvT888+XWu706dN15MgR3XbbbbJYLHruueeUlpamzZs3l9pa8cMPPyg9PV133nmn4uLi9Morr2jAgAHKyspSUlKSJPPLbK9evZScnKwnnnhCVqtVTz75pKpVq1Zq3QYPHqzXX39dX331la655hr79uPHj+uLL77Q8OHDFRoaqj179ujyyy9XtWrV9NBDDykxMVFbtmxRenp6qc9RlKFDh2rEiBGaN2+errrqKvv2NWvW6I8//tBjjz0mSVq+fLl+/PFHXXvttapTp462bNmiSZMmqWvXrvrzzz/L1FJnGIb69u2rH374QbfffruaN2+uWbNmadiwYYX2nT9/vjZv3qwbb7xRNWvW1Nq1azV58mStXbtWP/30kywWi9LS0vT3339rxowZeumll1S1alVJKva87969WxdddJGOHz+uu+++W0lJSXr33Xd19dVXa+bMmerfv7/D/s8884xCQkJ0//336/Dhw3ruuec0dOhQ/fzzz06/5uKsXbtWnTt3Vnx8vB588EGFh4frrbfeUteuXbVo0SJdcMEFksygN2HCBN1yyy3q0KGDcnJytGLFCq1cuVKXXXaZJGnAgAFau3at7rrrLqWmpmrPnj2aP3++srKyXJ4oBABcZgAAijRy5Ejj7I/JLl26GJKMN998s9D+x48fL7TttttuM2JiYoyTJ0/atw0bNsyoV6+e/X5mZqYhyUhKSjIOHDhg3/75558bkowvvvjCvu3xxx8vVCdJRkREhLFx40b7ttWrVxuSjFdffdW+rU+fPkZMTIyxfft2+7YNGzYYYWFhhco8W35+vlG7dm1jwIABDts/+eQTQ5KxePFiwzAMY9asWYYkY/ny5SWW56wDBw4YkZGRxpAhQxy2P/TQQ4YkY/369YZhFH3uly1bZkgy3nvvPfu2hQsXGpKMhQsX2red/e8xe/ZsQ5Lx3HPP2bedPn3a6Ny5syHJmDp1qn17Uc87Y8YMh3NiGIbx/PPPG5KMzMzMQvvXq1fPGDZsmP3+mDFjDEnGkiVL7NuOHDli1K9f30hNTTWsVqvDa2nevLmRm5tr3/fll182JBlr1qwp9FwFTZ06tdR/q379+hkRERHGpk2b7Nt27NhhxMXFGZdccol9W5s2bYwrr7yy2HIOHjxoSDKef/75EusEAL5CdzgAKKPIyEjdeOONhbZHR0fb148cOaJ9+/apc+fOOn78uP76669Syx08eLAqV65sv9+5c2dJZteo0vTo0UMNGza032/durXi4+Ptx1qtVi1YsED9+vVTrVq17Ps1atRIvXv3LrV8i8Wia665RnPnztXRo0ft2z/++GPVrl1bF198sSTZB/1/+eWXRXZhK6vKlSvriiuu0Jw5c3Ts2DFJZkvNRx99pPPPP19NmjSR5Hju8/LytH//fjVq1EiJiYlauXJlmZ5z7ty5CgsL0x133GHfFhoaqrvuuqvQvgWf9+TJk9q3b58uvPBCSSrz8xZ8/g4dOtjPqSTFxsZqxIgR2rJli/7880+H/W+88UZFRETY75flfVMSq9Wqb7/9Vv369VODBg3s25OTk3Xdddfphx9+UE5OjiTz333t2rXasGFDkWVFR0crIiJCGRkZOnjwoFv1AgBPIAQBQBnVrl3b4Uunzdq1a9W/f38lJCQoPj5e1apVs0+qcPjw4VLLrVu3rsN9WyBy5kvj2cfajrcdu2fPHp04cUKNGjUqtF9R24oyePBgnThxQnPmzJEkHT16VHPnztU111xjH1PUpUsXDRgwQE888YSqVq2qvn37aurUqcrNzXXqOYoydOhQHTt2TJ9//rkkc6a1LVu2OEyIcOLECT322GNKSUlRZGSkqlatqmrVqunQoUNOnfuCtm7dquTkZMXGxjpsb9q0aaF9Dxw4oNGjR6tGjRqKjo5WtWrVVL9+fUnO/ZsX9/xFPZdtVsKtW7c6bHfnfVOSvXv36vjx48XWJT8/X9nZ2ZKkJ598UocOHVKTJk10zjnn6IEHHtDvv/9u3z8yMlLPPvusvv76a9WoUUOXXHKJnnvuOe3atcutOgKAqwhBAFBGBX/9tzl06JC6dOmi1atX68knn9QXX3yh+fPn69lnn5Ukp6ZgDg0NLXK7UcJ1cjxxrLMuvPBCpaam6pNPPpEkffHFFzpx4oQGDx5s38disWjmzJn2qau3b9+um266Se3atXNoQSqLq666SgkJCZo+fbokc+xUaGiorr32Wvs+d911l55++mkNGjRIn3zyib799lvNnz9fSUlJXp3+etCgQZoyZYpuv/12paen69tvv7VPSOHtabdtfPFvX5pLLrlEmzZt0jvvvKNWrVrpf//7n8477zz973//s+8zZswY/f3335owYYKioqL06KOPqnnz5vrtt998Vk8AsCEEAYAHZGRkaP/+/Zo2bZpGjx6tq666Sj169HDo3uZP1atXV1RUVJGzhpVlJrFBgwbpm2++UU5Ojj7++GOlpqbau38VdOGFF+rpp5/WihUr9OGHH2rt2rX66KOPXKp7ZGSkBg4cqG+//Va7d+/Wp59+qm7duqlmzZr2fWbOnKlhw4bpv//9rwYOHKjLLrtMF198sUsXJ61Xr5527txZKLStX7/e4f7Bgwf13Xff6aGHHtITTzyh/v3767LLLnPoOmZT2ux7Zz//2c8lyd6lsl69ek6X5Y5q1aopJiam2LqEhIQoJSXFvq1KlSq68cYbNWPGDGVnZ6t169YaP368w3ENGzbUfffdp2+//VZ//PGHTp06pf/+97/efikAUAghCAA8wPZrfMFf30+dOqU33njDX1VyEBoaqh49emj27NnasWOHffvGjRv19ddfO13O4MGDlZubq3fffVfffPONBg0a5PD4wYMHC7VAnHvuuZLk0CVu06ZN2rRpk9PPO3ToUOXl5em2227T3r17C10bKDQ0tNDzvvrqq4WmJ3fGFVdcodOnT2vSpEn2bVarVa+++mqh55QKt7hMnDixUJmVKlWSJKdC2RVXXKFffvlFy5Yts287duyYJk+erNTUVLVo0cLZl+KW0NBQXX755fr8888dprHevXu3pk+frosvvljx8fGS5DAluGSOYWrUqJH93/z48eM6efKkwz4NGzZUXFycW10lAcBVTJENAB5w0UUXqXLlyho2bJjuvvtuWSwWvf/++z7tklSa8ePH69tvv1WnTp10xx13yGq16rXXXlOrVq20atUqp8o477zz1KhRIz388MPKzc116AonSe+++67eeOMN9e/fXw0bNtSRI0c0ZcoUxcfH64orrrDv1717d0ly+hoxXbp0UZ06dfT5558rOjpaaWlpDo9fddVVev/995WQkKAWLVpo2bJlWrBggX168LLo06ePOnXqpIceekhbtmxRixYtlJ6eXmiMT3x8vH1sS15enmrXrq1vv/1WmZmZhcps166dJOnhhx/Wtddeq/DwcPXp08cejgp66KGHNGPGDPXu3Vt33323qlSponfffVeZmZn67LPPFBLi2d8v33nnnSKvKTV69Gj9+9//1vz583XxxRfrzjvvVFhYmN566y3l5ubqueees+/bokULde3aVe3atVOVKlW0YsUKzZw5U6NGjZIk/f333+revbsGDRqkFi1aKCwsTLNmzdLu3bsdujUCgK8QggDAA5KSkvTll1/qvvvu0yOPPKLKlSvrX//6l7p3766ePXv6u3qSzC/iX3/9te6//349+uijSklJ0ZNPPql169Y5NXudzeDBg/X000+rUaNGOu+88xwe69Kli3755Rd99NFH2r17txISEtShQwd9+OGH9gkDXBESEqIhQ4bo+eefV58+fRQXF+fw+Msvv6zQ0FB9+OGHOnnypDp16qQFCxa4dO5DQkI0Z84cjRkzRh988IEsFouuvvpq/fe//1Xbtm0d9p0+fbruuusuvf766zIMQ5dffrm+/vprhxn4JKl9+/Z66qmn9Oabb+qbb75Rfn6+MjMziwxBNWrU0I8//qixY8fq1Vdf1cmTJ9W6dWt98cUXuvLKK8v8ekpTsMWroOHDh6tly5ZasmSJxo0bpwkTJig/P18XXHCBPvjgA/s1giTp7rvv1pw5c/Ttt98qNzdX9erV07///W/7RWJTUlI0ZMgQfffdd3r//fcVFhamZs2a6ZNPPtGAAQM8/poAoDQWI5B+pgQA+Fy/fv1KnN4YAICKhjFBABBETpw44XB/w4YNmjt3rrp27eqfCgEA4Ae0BAFAEElOTtbw4cPVoEEDbd26VZMmTVJubq5+++03NW7c2N/VAwDAJxgTBABBpFevXpoxY4Z27dqlyMhIdezYUf/5z38IQACAoEJLEAAAAICgwpggAAAAAEGFEAQAAAAgqJTrMUH5+fnasWOH4uLiZLFY/F0dAAAAAH5iGIaOHDmiWrVqlXph6XIdgnbs2KGUlBR/VwMAAABAgMjOzladOnVK3KdchyDbFcOzs7MVHx/v59oAAAAA8JecnBylpKTYM0JJynUIsnWBi4+PJwQBAAAAcGqYDBMjAAAAAAgqhCAAAAAAQYUQBAAAACColOsxQQAAAAg8VqtVeXl5/q4GKpjQ0FCFhYV55NI4hCAAAAB4zNGjR7Vt2zYZhuHvqqACiomJUXJysiIiItwqhxAEAAAAj7Bardq2bZtiYmJUrVo1LmYPjzEMQ6dOndLevXuVmZmpxo0bl3pB1JIQggAAAOAReXl5MgxD1apVU3R0tL+rgwomOjpa4eHh2rp1q06dOqWoqCiXy2JiBAAAAHgULUDwFndafxzK8UgpAAAAAFBOEIIAAAAABBVCEAAAAAKK1SplZEgzZpi3Vqu/a1R2qampmjhxotP7Z2RkyGKx6NChQ16rE84gBAEAACBgpKdLqanSpZdK111n3qammtu9wWKxlLiMHz/epXKXL1+uESNGOL3/RRddpJ07dyohIcGl53MWYcvE7HAAAAAICOnp0sCB0tmXGNq+3dw+c6aUlubZ59y5c6d9/eOPP9Zjjz2m9evX27fFxsba1w3DkNVqVVhY6V+hq1WrVqZ6REREqGbNmmU6Bq6jJcgDKkKTLQAAgKcZhnTsmHNLTo50992FA5CtHEkaPdrcz5nynL1Wa82aNe1LQkKCLBaL/f5ff/2luLg4ff3112rXrp0iIyP1ww8/aNOmTerbt69q1Kih2NhYtW/fXgsWLHAo9+zucBaLRf/73//Uv39/xcTEqHHjxpozZ4798bNbaKZNm6bExETNmzdPzZs3V2xsrHr16uUQ2k6fPq27775biYmJSkpK0tixYzVs2DD169fPuRdfhIMHD+qGG25Q5cqVFRMTo969e2vDhg32x7du3ao+ffqocuXKqlSpklq2bKm5c+fajx06dKh9ivTGjRtr6tSpLtfFmwhBbvJ1ky0AAEB5cfy4FBvr3JKQYLb4FMcwpG3bzP2cKe/4cc+9joceekjPPPOM1q1bp9atW+vo0aO64oor9N133+m3335Tr1691KdPH2VlZZVYzhNPPKFBgwbp999/1xVXXKGhQ4fqwIEDxe5//PhxvfDCC3r//fe1ePFiZWVl6f7777c//uyzz+rDDz/U1KlTtXTpUuXk5Gj27Nluvdbhw4drxYoVmjNnjpYtWybDMHTFFVcoLy9PkjRy5Ejl5uZq8eLFWrNmjZ599ll7a9mjjz6qP//8U19//bXWrVunSZMmqWrVqm7Vx1voDucGfzTZAgAAwLeefPJJXXbZZfb7VapUUZs2bez3n3rqKc2aNUtz5szRqFGjii1n+PDhGjJkiCTpP//5j1555RX98ssv6tWrV5H75+Xl6c0331TDhg0lSaNGjdKTTz5pf/zVV1/VuHHj1L9/f0nSa6+9Zm+VccWGDRs0Z84cLV26VBdddJEk6cMPP1RKSopmz56ta665RllZWRowYIDOOeccSVKDBg3sx2dlZalt27Y6//zzJZmtYYGKliAXWa1mk2xJTbZjxtA1DgAABK+YGOnoUecWZ7+7z53rXHkxMZ57HbYv9TZHjx7V/fffr+bNmysxMVGxsbFat25dqS1BrVu3tq9XqlRJ8fHx2rNnT7H7x8TE2AOQJCUnJ9v3P3z4sHbv3q0OHTrYHw8NDVW7du3K9NoKWrduncLCwnTBBRfYtyUlJalp06Zat26dJOnuu+/Wv//9b3Xq1EmPP/64fv/9d/u+d9xxhz766COde+65evDBB/Xjjz+6XBdvIwS5aMkSs0m2OIYhZWeb+wEAAAQji0WqVMm55fLLpTp1zGOKKyslxdzPmfKKK8cVlSpVcrh///33a9asWfrPf/6jJUuWaNWqVTrnnHN06tSpEssJDw8/6zVZlJ+fX6b9DWcHO3nJLbfcos2bN+v666/XmjVrdP755+vVV1+VJPXu3Vtbt27VPffcox07dqh79+4O3fcCCSHIRQXGpHlkPwAAgGAWGiq9/LK5fnaAsd2fONHcz9+WLl2q4cOHq3///jrnnHNUs2ZNbdmyxad1SEhIUI0aNbR8+XL7NqvVqpUrV7pcZvPmzXX69Gn9/PPP9m379+/X+vXr1aJFC/u2lJQU3X777UpPT9d9992nKVOm2B+rVq2ahg0bpg8++EATJ07U5MmTXa6PNzEmyEXJyZ7dDwAAINilpZljqkePduxxU6eOGYACZax148aNlZ6erj59+shisejRRx8tsUXHW+666y5NmDBBjRo1UrNmzfTqq6/q4MGDsjjRDLZmzRrFxcXZ71ssFrVp00Z9+/bVrbfeqrfeektxcXF66KGHVLt2bfXt21eSNGbMGPXu3VtNmjTRwYMHtXDhQjVv3lyS9Nhjj6ldu3Zq2bKlcnNz9eWXX9ofCzSEIBd17mz+h9y+vehxQRaL+Xjnzr6vGwAAQHmVlib17WsOKdi50/xBuXPnwGgBsnnxxRd100036aKLLlLVqlU1duxY5eTk+LweY8eO1a5du3TDDTcoNDRUI0aMUM+ePRXqxMm65JJLHO6Hhobq9OnTmjp1qkaPHq2rrrpKp06d0iWXXKK5c+fau+ZZrVaNHDlS27ZtU3x8vHr16qWXXnpJknmto3HjxmnLli2Kjo5W586d9dFHH3n+hXuAxfB3x0I35OTkKCEhQYcPH1Z8fLzPn982O5zkGIRs4ZvZ4QAAQDA5efKkMjMzVb9+fUVFRfm7OkEnPz9fzZs316BBg/TUU0/5uzpeUdJ7rCzZgDFBbrA12dau7bi9Zk0CEAAAALxr69atmjJliv7++2+tWbNGd9xxhzIzM3Xdddf5u2oBjxDkprQ0acsWaeFCs/ubJP3vfwQgAAAAeFdISIimTZum9u3bq1OnTlqzZo0WLFgQsONwAgljgjwgNFTq2lU6/3xzEN/mzf6uEQAAACq6lJQULV261N/VKJdoCfKgJk3M27//9m89AAAAABSPEORBthC0YYN/6wEAAACgeIQgD2rc2LylJQgAAAAIXIQgD7K1BG3ZIp065deqAAAAACgGIciDatSQYmOl/HwmRwAAAAACFSHIgywWJkcAAAAAAh0hyMNs44KYHAEAAMBFVquUkSHNmGHeWq3+rlGpunbtqjFjxtjvp6amauLEiSUeY7FYNHv2bLef21PlBBNCkIfREgQAAOCG9HQpNVW69FLpuuvM29RUc7sX9OnTR7169SrysSVLlshisej3338vc7nLly/XiBEj3K2eg/Hjx+vcc88ttH3nzp3q3bu3R5/rbNOmTVNiYqJXn8OXCEEeRksQAACAi9LTpYEDzavPF7R9u7ndC0Ho5ptv1vz587Xt7OeUNHXqVJ1//vlq3bp1mcutVq2aYmJiPFHFUtWsWVORkZE+ea6KghDkYbQEAQAA/MMwpGPHnFtycqS77zaPKaocSRo92tzPmfKKKqcIV111lapVq6Zp06Y5bD969Kg+/fRT3Xzzzdq/f7+GDBmi2rVrKyYmRuecc45mzJhRYrlnd4fbsGGDLrnkEkVFRalFixaaP39+oWPGjh2rJk2aKCYmRg0aNNCjjz6qvLw8SWZLzBNPPKHVq1fLYrHIYrHY63x2d7g1a9aoW7duio6OVlJSkkaMGKGjR4/aHx8+fLj69eunF154QcnJyUpKStLIkSPtz+WKrKws9e3bV7GxsYqPj9egQYO0e/du++OrV6/WpZdeqri4OMXHx6tdu3ZasWKFJGnr1q3q06ePKleurEqVKqlly5aaO3euy3VxRphXSw9Ctpag7dvN/3+VKvm3PgAAAH5z/Lg5da4nGIbZQpSQ4Nz+R4869UUsLCxMN9xwg6ZNm6aHH35YFotFkvTpp5/KarVqyJAhOnr0qNq1a6exY8cqPj5eX331la6//no1bNhQHTp0KPU58vPzlZaWpho1aujnn3/W4cOHHcYP2cTFxWnatGmqVauW1qxZo1tvvVVxcXF68MEHNXjwYP3xxx/65ptvtGDBAklSQhHn4tixY+rZs6c6duyo5cuXa8+ePbrllls0atQoh6C3cOFCJScna+HChdq4caMGDx6sc889V7feemupr6eo12cLQIsWLdLp06c1cuRIDR48WBkZGZKkoUOHqm3btpo0aZJCQ0O1atUqhYeHS5JGjhypU6dOafHixapUqZL+/PNPxXrqfVMMQpCHVakiJSVJ+/dLGzdKbdr4u0YAAAAoyU033aTnn39eixYtUteuXSWZXeEGDBighIQEJSQk6P7777fvf9ddd2nevHn65JNPnApBCxYs0F9//aV58+apVq1akqT//Oc/hcbxPPLII/b11NRU3X///froo4/04IMPKjo6WrGxsQoLC1PNmjWLfa7p06fr5MmTeu+991TpnxD42muvqU+fPnr22WdVo0YNSVLlypX12muvKTQ0VM2aNdOVV16p7777zqUQ9N1332nNmjXKzMxUSkqKJOm9995Ty5YttXz5crVv315ZWVl64IEH1KxZM0lSY1vLgcxWpAEDBuicc86RJDVo0KDMdSgrusN5AeOCAAAAJMXEmC0yzizOdn+aO9e58sowHqdZs2a66KKL9M4770iSNm7cqCVLlujmm2+WJFmtVj311FM655xzVKVKFcXGxmrevHnKyspyqvx169YpJSXFHoAkqWPHjoX2+/jjj9WpUyfVrFlTsbGxeuSRR5x+joLP1aZNG3sAkqROnTopPz9f69evt29r2bKlQkND7feTk5O1Z8+eMj1XwedMSUmxByBJatGihRITE7Vu3TpJ0r333qtbbrlFPXr00DPPPKNNmzbZ97377rv173//W506ddLjjz/u0kQUZUUI8gLGBQEAAMi8iGKlSs4tl18u1aljHlNcWSkp5n7OlFdcOcW4+eab9dlnn+nIkSOaOnWqGjZsqC5dukiSnn/+eb388ssaO3asFi5cqFWrVqlnz546deqUu2fIbtmyZRo6dKiuuOIKffnll/rtt9/08MMPe/Q5CrJ1RbOxWCzKz8/3ynNJ5sx2a9eu1ZVXXqnvv/9eLVq00KxZsyRJt9xyizZv3qzrr79ea9as0fnnn69XX33Va3WRCEFeQUsQAABAGYWGSi+/bK6fHWBs9ydONPfzgkGDBikkJETTp0/Xe++9p5tuusk+Pmjp0qXq27ev/vWvf6lNmzZq0KCB/i7Dr93NmzdXdna2du7cad/2008/Oezz448/ql69enr44Yd1/vnnq3Hjxtq6davDPhEREbKWcs2k5s2ba/Xq1Tp27Jh929KlSxUSEqKmTZs6XeeysL2+7Oxs+7Y///xThw4dUosWLezbmjRponvuuUfffvut0tLSNHXqVPtjKSkpuv3225Wenq777rtPU6ZM8UpdbQhBXkBLEAAAgAvS0qSZM6XatR2316ljbk9L89pTx8bGavDgwRo3bpx27typ4cOH2x9r3Lix5s+frx9//FHr1q3Tbbfd5jDzWWl69OihJk2aaNiwYVq9erWWLFmihx9+2GGfxo0bKysrSx999JE2bdqkV155xd5SYpOamqrMzEytWrVK+/btU25ubqHnGjp0qKKiojRs2DD98ccfWrhwoe666y5df/319vFArrJarVq1apXDsm7dOvXo0UPnnHOOhg4dqpUrV+qXX37RDTfcoC5duuj888/XiRMnNGrUKGVkZGjr1q1aunSpli9frubNm0uSxowZo3nz5ikzM1MrV67UwoUL7Y95CyHIC2gJAgAAcFFamrRli7RwoTR9unmbmenVAGRz88036+DBg+rZs6fD+J1HHnlE5513nnr27KmuXbuqZs2a6tevn9PlhoSEaNasWTpx4oQ6dOigW265RU8//bTDPldffbXuuecejRo1Sueee65+/PFHPfroow77DBgwQL169dKll16qatWqFTlNd0xMjObNm6cDBw6offv2GjhwoLp3767XXnutbCejCEePHlXbtm0dlj59+shisejzzz9X5cqVdckll6hHjx5q0KCBPv74Y0lSaGio9u/frxtuuEFNmjTRoEGD1Lt3bz3xxBOSzHA1cuRINW/eXL169VKTJk30xhtvuF3fklgMw8lJ1ANQTk6OEhISdPjwYcXHx/u7OnZHj0pxceb6wYNSBbq4LgAAQLFOnjypzMxM1a9fX1FRUf6uDiqgkt5jZckGtAR5QWyslJxsrtMaBAAAAAQWQpCXMC4IAAAACEyEIC9hXBAAAAAQmAhBXkJLEAAAABCYCEFeQksQAAAIVuV43i0EOE+9twhBXlKwJYjPAQAAEAxC/7mQ6alTp/xcE1RUx48flySFh4e7VU6YJyqDwho0MC9unJMj7d0rVa/u7xoBAAB4V1hYmGJiYrR3716Fh4crJITf2+EZhmHo+PHj2rNnjxITE+2B21WEIC+JipLq1TOv9fX334QgAABQ8VksFiUnJyszM1Nbt271d3VQASUmJqpmzZpul0MI8qLGjc0QtGGDdPHF/q4NAACA90VERKhx48Z0iYPHhYeHu90CZEMI8qImTaT585khDgAABJeQkBBFRUX5uxpAseio6UXMEAcAAAAEHkKQF3GtIAAAACDwEIK8yNYStHGjlJ/v37oAAAAAMBGCvCg1VQoLk06ckLZv93dtAAAAAEiEIK8KCzOvFyQxLggAAAAIFIQgL2NcEAAAABBYCEFeZgtBtAQBAAAAgYEQ5GW2yRFoCQIAAAACAyHIy+gOBwAAAAQWQpCX2VqCNm+WTp/2b10AAAAAEIK8rnZtKTraDEBbtvi7NgAAAAAIQV4WEiI1amSuMzkCAAAA4H+EIB9gXBAAAAAQOAhBPmAbF0RLEAAAAOB/hCAfoCUIAAAACByEIB+gJQgAAAAIHIQgH7C1BG3dKp086d+6AAAAAMGOEOQD1apJ8fGSYZjXCwIAAADgP4QgH7BYGBcEAAAABApCkI8wLggAAAAIDIQgH6ElCAAAAAgMhCAfoSUIAAAACAyEIB+hJQgAAAAIDIQgH7G1BO3cKR096t+6AAAAAMGMEOQjiYnmVNkSXeIAAAAAfyIE+RDjggAAAAD/IwT5EOOCAAAAAP8jBPkQLUEAAACA/wVMCHrmmWdksVg0ZswYf1fFa2gJAgAAAPwvIELQ8uXL9dZbb6l169b+ropX0RIEAAAA+J/fQ9DRo0c1dOhQTZkyRZUrV/Z3dbyqUSPzdv9+6cAB/9YFAAAACFZ+D0EjR47UlVdeqR49epS6b25urnJychyW8qRSJal2bXOd1iAAAADAP/wagj766COtXLlSEyZMcGr/CRMmKCEhwb6kpKR4uYaex7ggAAAAwL/8FoKys7M1evRoffjhh4qKinLqmHHjxunw4cP2JTs728u19DzGBQEAAAD+FeavJ/7111+1Z88enXfeefZtVqtVixcv1muvvabc3FyFhoY6HBMZGanIyEhfV9WjaAkCAAAA/MtvIah79+5as2aNw7Ybb7xRzZo109ixYwsFoIqCliAAAADAv/wWguLi4tSqVSuHbZUqVVJSUlKh7RVJwZYgw5AsFv/WBwAAAAg2fp8dLtg0aCCFhEhHj0q7d/u7NgAAAEDw8VtLUFEyMjL8XQWvi4iQUlOlzZvN1qCaNf1dIwAAACC40BLkB7ZxQUyOAAAAAPgeIcgPbOOCmBwBAAAA8D1CkB8wTTYAAADgP4QgP2CabAAAAMB/CEF+YGsJ2rhRys/3b10AAACAYEMI8oO6dc1Z4nJzpexsf9cGAAAACC6EID8IDZUaNjTXGRcEAAAA+BYhyE8YFwQAAAD4ByHIT5ghDgAAAPAPQpCf0BIEAAAA+AchyE9oCQIAAAD8gxDkJ7aWoMxMKS/Pv3UBAAAAggkhyE9q1ZJiYiSr1QxCAAAAAHyDEOQnFgvjggAAAAB/IAT5EeOCAAAAAN8jBPkRLUEAAACA7xGC/IiWIAAAAMD3CEF+REsQAAAA4HuEID+ytQRlZUknTvi3LgAAAECwIAT5UVKSlJhorm/a5NeqAAAAAEGDEORHFgvjggAAAABfIwT5GeOCAAAAAN8iBPkZLUEAAACAbxGC/IyWIAAAAMC3CEF+RksQAAAA4FuEID+ztQTt3i3l5Pi3LgAAAEAwIAT5WXy8VKOGuU6XOAAAAMD7CEEBgHFBAAAAgO8QggIA44IAAAAA3yEEBQBbSxAhCAAAAPA+QlAAsLUE0R0OAAAA8D5CUAAo2BJkGP6tCwAAAFDREYICQKNG5u2hQ9L+/X6tCgAAAFDhEYICQHS0lJJirjMuCAAAAPAuQlCAYFwQAAAA4BuEoADBNNkAAACAbxCCAgQXTAUAAAB8gxAUIGgJAgAAAHyDEBQgCrYEMU02AAAA4D2EoABRv74UGiodPy7t2OHv2gAAAAAVFyEoQISHm0FIYlwQAAAA4E2EoADCuCAAAADA+whBAYQZ4gAAAADvIwQFEFqCAAAAAO8jBAUQWoIAAAAA7yMEBRBbS9CmTZLV6t+6AAAAABUVISiApKRIkZHSqVNSVpa/awMAAABUTISgABISIjVqZK4zLggAAADwDkJQgGFcEAAAAOBdhKAAwwxxAAAAgHcRggIMLUEAAACAdxGCAgwtQQAAAIB3EYICjK0laMsWc5Y4AAAAAJ5FCAowNWtKsbFSfr60ebO/awMAAABUPISgAGOxMC4IAAAA8CZCUABiXBAAAADgPYSgAERLEAAAAOA9hKAAREsQAAAA4D2EoABESxAAAADgPYSgAGRrCdq2TTp2zL91AQAAACoaQlAAqlLFXCRp40b/1gUAAACoaAhBAcrWGkSXOAAAAMCzCEEByjYuiMkRAAAAAM8iBAUoWoIAAAAA7yAEBShaggAAAADvIAQFKFqCAAAAAO8gBAWoRo3M2717pUOH/FoVAAAAoEIhBAWouDgpOdlcpzUIAAAA8BxCUACzdYljXBAAAADgOYSgAGabHIGWIAAAAMBzCEEBjJYgAAAAwPMIQQGMliAAAADA8whBAaxgS5Bh+LcuAAAAQEVBCApgDRpIFouUk2NOlQ0AAADAfYSgABYVJdWrZ64zLggAAADwDEJQgGNcEAAAAOBZhKAAxwxxAAAAgGcRggIcLUEAAACAZxGCAhwtQQAAAIBnEYICnK0laONGKT/fv3UBAAAAKgJCUIBLTZXCwqQTJ6Tt2/1dGwAAAKD8IwQFuLAw83pBEuOCAAAAAE8gBHmC1SplZEgzZpi3VqtHi2dcEAAAAOA5Yf6uQLmXni6NHi1t23ZmW5060ssvS2lpHnkKZogDAAAAPIeWIHekp0sDBzoGIMkcvDNwoPm4B9ASBAAAAHgOIchVVqvZAmQYhR+zbRszxiNd42gJAgAAADyHEOSqJUsKtwAVZBhSdra5n5tsLUGbNkmnT7tdHAAAABDUCEGu2rnTs/uVoHZtKSrKDEBbt7pdHAAAABDUCEGuSk727H4lCAk50yWOcUEAAACAewhBrurc2ZwFzmIp+nGLRUpJMffzAEIQAAAA4Bl+DUGTJk1S69atFR8fr/j4eHXs2FFff/21P6vkvNBQcxpsqfggNHGiuZ8H2MYFMTkCAAAA4B6/hqA6deromWee0a+//qoVK1aoW7du6tu3r9auXevPajkvLU2aOdMctFNQXJy53UPXCZJoCQIAAAA8xWIYRc3x7D9VqlTR888/r5tvvrnUfXNycpSQkKDDhw8rPj7eB7UrhtVqzgL32WfSa69J554r/fabR5/ihx/MnnWpqVJmpkeLBgAAAMq9smSDMB/VqVRWq1Wffvqpjh07po4dOxa5T25urnJzc+33c3JyfFW9koWGSl27Sk2bmiFo9WrpwAGpShWPPYWtJWjrVunkSXO2OAAAAABl5/eJEdasWaPY2FhFRkbq9ttv16xZs9SiRYsi950wYYISEhLsS0pKio9rW4rkZKlZM/MaQYsXe7To6tWl+Hiz6M2bPVo0AAAAEFT8HoKaNm2qVatW6eeff9Ydd9yhYcOG6c8//yxy33Hjxunw4cP2JTs728e1dcKll5q3Cxd6tFiLhXFBAAAAgCf4PQRFRESoUaNGateunSZMmKA2bdroZdusa2eJjIy0zyRnWwKOl0KQxAxxAAAAgCf4PQSdLT8/32HcT7nTtat5u2aNtHevR4umJQgAAABwn19D0Lhx47R48WJt2bJFa9as0bhx45SRkaGhQ4f6s1ruqVZNatXKXM/I8GjRtAQBAAAA7vNrCNqzZ49uuOEGNW3aVN27d9fy5cs1b948XXbZZf6slvu81CWOliAAAADAfX6dIvvtt9/259N7T7du0quvei0E7dwpHT0qxcZ6tHgAAAAgKATcmKAKoUsXczq3v/4yE4uHVK5s9raT6BIHAAAAuIoQ5A2VK0vnnmuue3hckK01iBAEAAAAuIYQ5C1eGhdkmxyBcUEAAACAawhB3uLlyRFoCQIAAABcQwjyls6dpZAQaeNGads2jxVLSxAAAADgHkKQtyQkSO3amesebA2iJQgAAABwDyHIm7zQJa5RI/N2/37pwAGPFQsAAAAEDUKQN3khBFWqJNWuba7TGgQAAACUHSHImy6+WAoLk7ZskTIzPVYs44IAAAAA1xGCvCk2Vmrf3lxnXBAAAAAQEAhB3uaFLnG0BAEAAACuIwR5W7du5u3ChZJheKRIWoIAAAAA1xGCvO2ii6SICGn7dvOaQR5QsCXIQ7kKAAAACBqEIG+LjpYuvNBc91CXuAYNzOuwHj0q7d7tkSIBAACAoEEI8gUPjwuKiJBSU811xgUBAAAAZUMI8oWCIYhxQQAAAIBfEYJ84cILpagos+/aX395pEhmiAMAAABcQwjyhchIc4IEyWNd4mgJAgAAAFxDCPIVD48LoiUIAAAAcA0hyFdsISgjQ8rPd7s4W0vQxo0eKQ4AAAAIGoQgX2nfXoqJkfbtk9audbu4evWk8HApN1fKzvZA/QAAAIAgQQjylYgI6eKLzfXvv3e7uNBQqWFDc50ucQAAAIDzCEG+5KVxQUyOAAAAADjPpRCUnZ2tbdu22e//8ssvGjNmjCZPnuyxilVI3bqZt4sWSVar28XZxgXREgQAAAA4z6UQdN1112nhP60Zu3bt0mWXXaZffvlFDz/8sJ588kmPVrBCOe88KS5OOnRIWr3a7eJoCQIAAADKzqUQ9Mcff6hDhw6SpE8++UStWrXSjz/+qA8//FDTpk3zZP0qlrAw6ZJLzHUPdImjJQgAAAAoO5dCUF5eniIjIyVJCxYs0NVXXy1JatasmXbu3Om52lVEHhwXZGsJysyU8vLcLg4AAAAICi6FoJYtW+rNN9/UkiVLNH/+fPXq1UuStGPHDiUlJXm0ghWOLQQtXiydPu1WUbVqmbNuW61mEAIAAABQOpdC0LPPPqu33npLXbt21ZAhQ9SmTRtJ0pw5c+zd5FCMNm2kxETpyBFp5Uq3irJYznSJY1wQAAAA4JwwVw7q2rWr9u3bp5ycHFWuXNm+fcSIEYqJifFY5Sqk0FCpSxfp88/NLnFuhsbGjc05Fv7+W7rySg/VEQAAAKjAXGoJOnHihHJzc+0BaOvWrZo4caLWr1+v6tWre7SCFZIXxgXREgQAAAA4x6UQ1LdvX7333nuSpEOHDumCCy7Qf//7X/Xr10+TJk3yaAUrJFsI+uEHt2c0YIY4AAAAoGxcCkErV65U586dJUkzZ85UjRo1tHXrVr333nt65ZVXPFrBCqlVKykpSTp2TFq+3K2iaAkCAAAAysalEHT8+HHFxcVJkr799lulpaUpJCREF154obZu3erRClZIISFS167m+vffu1WULQRlZUknTrhXLQAAACAYuBSCGjVqpNmzZys7O1vz5s3T5ZdfLknas2eP4uPjPVrBCstD44KSkszJ5iRp0yb3qgQAAAAEA5dC0GOPPab7779fqamp6tChgzp27CjJbBVq27atRytYYdlC0I8/Srm5LhdjsZxpDWJcEAAAAFA6l0LQwIEDlZWVpRUrVmjevHn27d27d9dLL73kscpVaM2bSzVqSCdPSj/95FZRXCsIAAAAcJ5LIUiSatasqbZt22rHjh3atm2bJKlDhw5q1qyZxypXoVksHusSR0sQAAAA4DyXQlB+fr6efPJJJSQkqF69eqpXr54SExP11FNPKT8/39N1rLg8FIJoCQIAAACcF+bKQQ8//LDefvttPfPMM+rUqZMk6YcfftD48eN18uRJPf300x6tZIVlC0E//WRO7RYd7VIxtAQBAAAAzrMYhmGU9aBatWrpzTff1NVXX+2w/fPPP9edd96p7du3e6yCJcnJyVFCQoIOHz5cPmelMwwpJUXavl1asEDq3t2lYnJypIQEc/3wYak8ngoAAADAHWXJBi51hztw4ECRY3+aNWumAwcOuFJkcPLQuKD4eHOOBYkucQAAAEBpXApBbdq00WuvvVZo+2uvvabWrVu7XamgwrggAAAAwKdcGhP03HPP6corr9SCBQvs1whatmyZsrOzNXfuXI9WsMKzhaBffpGOHpViY10qpkkT6YcfGBcEAAAAlMallqAuXbro77//Vv/+/XXo0CEdOnRIaWlpWrt2rd5//31P17Fiq19fqldPOn1aWrrU5WJoCQIAAACc41JLkGROjnD2LHCrV6/W22+/rcmTJ7tdsaBy6aXStGlml7iePV0qghniAAAAAOe4fLFUeJCtS9z337tchK0l6O+/zUnnAAAAABSNEBQIbCHo11/NOa5d0KiReXvokLR/v2eqBQAAAFREhKBAkJJippj8fGnJEpeKiI42i5EYFwQAAACUpExjgtLS0kp8/NChQ+7UJbhdeqm0caM5Luiqq1wqokkTKTvb7BL3z6R9AAAAAM5SphCUkJBQ6uM33HCDWxUKWpdeKk2Z4tb1gho3lr77jskRAAAAgJKUKQRNnTrVW/VA167m7apV0oEDUpUqZS7CNkMc3eEAAACA4jEmKFAkJ0vNmplTuy1e7FIRBWeIAwAAAFA0QlAgsc0S52KXuIItQUyTDQAAABSNEBRI3AxB9etLoaHS8ePSjh0erBcAAABQgRCCAoltXNCaNdLevWU+PDzcDEIS44IAAACA4hCCAkm1alKrVub6okUuFcG4IAAAAKBkhKBA48FxQQAAAAAKIwQFGlsI+v57lw6nJQgAAAAoGSEo0HTpIlks0l9/STt3lvlwWoIAAACAkhGCAk2VKtK555rrGRllPtzWErRpk2S1eqxWAAAAQIVBCApEbowLSkmRIiOlU6ekrCwP1wsAAACoAAhBgciNEBQaKjVsaK4zLggAAAAojBAUiDp3lkJCpI0bpW3bynw444IAAACA4hGCAlFCgtSunbnuQmsQM8QBAAAAxSMEBSo3usTREgQAAAAUjxAUqNwIQbQEAQAAAMUjBAWqiy+WwsKkLVvMpQxsLUFbtpizxAEAAAA4gxAUqGJjpfbtzfUytgbVrGkenp8vbd7shboBAAAA5RghKJC52CXOYjnTJY5xQQAAAIAjQlAgs4Wg77+XDKNMh9q6xDEuCAAAAHBECApkF10khYdL27eb1wwqA1qCAAAAgKIRggJZTIzUsaO5XsYucbQEAQAAAEUjBAU6F8cFNWhg3q5eLWVkSFarZ6sFAAAAlFeEoEBXMAQ5OS4oPV0aNMhcP3DALCI11dwOAAAABDtCUKC78EIpKkravVv6669Sd09PlwYOlHbscNy+fbu5nSAEAACAYEcICnSRkeYECVKpXeKsVmn06KIbjGzbxoyhaxwAAACCGyGoPHByXNCSJdK2bcU/bhhSdra5HwAAABCsCEHlgS0EZWRI+fnF7rZzp3PFObsfAAAAUBERgsqD9u3N6bL37ZPWri12t+Rk54pzdj8AAACgIiIElQcREdLFF5vrJXSJ69xZqlNHsliKftxikVJSzP0AAACAYEUIKi+cGBcUGiq9/LK5XlwQmjjR3A8AAAAIVoSg8qLguKASpndLS5NmzpRq1y782NCh5uMAAABAMCMElRft2klxcdKhQ9Lq1SXumpYmbdliNhpNny49+KC5fdmyEudVAAAAAIICIai8CAuTLrnEXC9lqmzJ7PLWtas0ZIj02GNSfLy0aZNThwIAAAAVGiGoPHHyekFnq1TJ7AonSVOmeLhOAAAAQDlDCCpPbCFo8WLp9OkyHTpihHmbni7t3evhegEAAADlCCGoPGnTRkpMlI4ckVauLNOh555rXm4oL096912v1A4AAAAoFwhB5UloqNSli7nuwuCeW281b6dMkQzDg/UCAAAAyhFCUHnj4rggSbr2Wik2Vvr7b7NHHQAAABCM/BqCJkyYoPbt2ysuLk7Vq1dXv379tH79en9WKfDZQtAPP5h928ogLk667jpzffJkD9cLAAAAKCf8GoIWLVqkkSNH6qefftL8+fOVl5enyy+/XMeOHfNntQJbq1ZSUpJ07Ji0fHmZD7dNkDBzprR/v4frBgAAAJQDfg1B33zzjYYPH66WLVuqTZs2mjZtmrKysvTrr7/6s1qBLSTEvACQ5FKXuHbtpLZtpVOnpPff92zVAAAAgPIgoMYEHT58WJJUpUqVIh/Pzc1VTk6OwxKU3BgXJJ1pDZo8mQkSAAAAEHwCJgTl5+drzJgx6tSpk1q1alXkPhMmTFBCQoJ9SUlJ8XEtA4QtBC1dKuXmlvnw666TYmKkdevMIgAAAIBgEjAhaOTIkfrjjz/00UcfFbvPuHHjdPjwYfuSnZ3twxoGkObNpRo1pJMnpZ9+KvPh8fHmTHESEyQAAAAg+ARECBo1apS+/PJLLVy4UHXq1Cl2v8jISMXHxzssQcli8ViXuE8/lQ4e9FC9AAAAgHLAryHIMAyNGjVKs2bN0vfff6/69ev7szrli5shqEMHqXVrszHpgw88WC8AAAAgwPk1BI0cOVIffPCBpk+frri4OO3atUu7du3SiRMn/Fmt8sEWgn76SXLhfFksTJAAAACA4GQxDP99/bVYLEVunzp1qoYPH17q8Tk5OUpISNDhw4eDr2ucYUgpKdL27dKCBVL37mUu4tAhqVYtM0MtWyZdeKHnqwkAAAD4Qlmygd+7wxW1OBOAgp4HxgUlJkqDBpnrTJAAAACAYBEQEyPARW6GIOlMl7iPPpL+uUwTAAAAUKERgsozWwj65Rfp6FGXiujYUWrZ0uwS9+GHHqwbAAAAEKAIQeVZ/fpSvXrS6dMuX/XUYpFuvdVcZ4IEAAAABANCUHnngS5x118vRUZKq1dLK1Z4qF4AAABAgCIElXceCEFVqkjXXGOuM0ECAAAAKjpCUHlnC0ErVrg1s4GtS9yMGdKRIx6oFwAAABCgCEHlXUqK1LChlJ8vLVnicjGdO0tNm0rHjplBCAAAAKioCEEVQbdu5q0bXeIsljPTZdMlDgAAABUZIagi8MC4IEm64QYpIkL69Vdp5UoP1AsAAAAIQISgiqBrV/N21SrpwAGXi6laVUpLM9enTHG7VgAAAEBAIgRVBMnJUrNm5kV+Fi92qyhbl7gPP3T5+qsAAABAQCMEVRQe6hLXtavUqJE5Q9zHH7tfLQAAACDQEIIqCg+FIIvlzHTZdIkDAABARUQIqihs44LWrJH27nWrqOHDpfBw6eefpdWr3a4ZAAAAEFAIQRVFtWpSq1bm+qJFbhVVvbrUr5+5TmsQAAAAKhpCUEXioS5x0pkJEj74QDp+3O3iAAAAgIBBCKpIPBiCunWT6teXDh+WPv3U7eIAAACAgEEIqki6dDFnNli3Ttq1y62iQkLOTJAwebIH6gYAAAAECEJQRVKlitSmjbnugdagG2+UwsKkH3+U1q51uzgAAAAgIBCCKppu3cxbD4SgmjWlPn3MdSZIAAAAQEVBCKpoPDguSDozQcJ770knTnikSAAAAMCvCEEVTefO5oCejRulbdvcLu6yy6R69aSDB6XPPvNA/QAAAAA/IwRVNAkJUrt25roHWoNCQ6VbbjHXmSABAAAAFQEhqCLycJe4G280G5eWLJH++ssjRQIAAAB+QwiqiDwcgmrXlq66ylxnggQAAACUd4Sgiujii825rbdsMRcPsE2Q8O67Um6uR4oEAAAA/IIQVBHFxkrnn2+uT5ggZWRIVqtbRfbqJdWpI+3fL82a5X4VAQAAAH8hBFVE6elnrm46ebLZPS411dzuotBQ6eabzxQJAAAAlFeEoIomPV0aOFA6csRx+/bt5nY3gtBNN5kTJCxcKP39t5v1BAAAAPyEEFSRWK3S6NGSYRR+zLZtzBiXu8bVrSv17m2u/+9/rlURAAAA8DdCUEWyZEnJF0g1DCk729zPRbfeat5OmyadOuVyMQAAAIDfEIIqkp07PbtfEa68UkpOlvbulT7/3OViAAAAAL8hBFUkycme3a8IYWFMkAAAAIDyjRBUkXTubM5jbbEU/bjFIqWkmPu54eabzaIWLJA2b3arKAAAAMDnCEEVSWio9PLL5npRQcgwpIkTzf3ckJoqXX65uc4ECQAAAChvCEEVTVqaNHOmVLt20Y/Xq+eRpxkxwrx95x0pL88jRQIAAAA+QQiqiNLSpC1bzAv6TJ9u3g4ZYj52zz1FT6FdRn36SDVqSLt3S1984XZxAAAAgM8Qgiqq0FCpa1cz/HTtKj37rBQdbU6P/dlnbhcfHi7deKO5PmWK28UBAAAAPkMIChYpKdIDD5jrDzwgnTzpdpG33GLezptnNjwBAAAA5QEhKJg8+KBUq5aZWGwTKLihYUOpRw+zd93bb7tfPQAAAMAXCEHBpFIl6ZlnzPWnn5Z27XK7yFtvNW/feUc6fdrt4gAAAACvIwQFm6FDpfbtpSNHpEcfdbu4fv2katWkHTukuXPdrx4AAADgbYSgYBMSIr30krn+9tvSqlVuFRcRIQ0fbq5PnuxWUQAAAIBPEIKCUadO0uDB5mAeD0yZbZsg4euvpexsD9QPAAAA8CJCULB69lkpMlLKyJA+/9ytopo0MWfhzs83xwYBAAAAgYwQFKzq1ZPuu89cv/9+KTfXreJGjDBv//c/yWp1s24AAACAFxGCgtlDD0k1a0qbNkmvveZWUf37S0lJ0rZt0jffeKh+AAAAgBcQgoJZXJz0n/+Y608+Ke3d63JRUVHSDTeY61OmeKBuAAAAgJcQgoLdsGFS27ZSTo702GNuFWW7ZtCXX0rbt3ugbgAAAIAXEIKCXcEpsydPlv74w+WimjeXOnc2xwRNneqh+gEAAAAeRgiC1KWLNGCAOb2bm1NmF5wgIT/fQ/UDAAAAPIgQBNNzz5lXPl2wQPrqK5eLGTBASkyUtm6V5s/3XPUAAAAATyEEwdSggTRmjLl+331SXp5LxURHn5kgYfJkz1QNAAAA8CRCEM54+GGpenXp77+lN95wuRjbBAlz5ki7dnmobgAAAICHEIJwRny89NRT5vr48dL+/S4V06qV1LGjdPq0NG2ax2oHAAAAeAQhCI5uvllq3Vo6dEh64gmXi7FNkDBlChMkAAAAILAQguAoNPTMlNlvvCGtW+dSMYMGSQkJ0ubN0vffe7B+AAAAgJsIQSisWzepb1/zgj/33edSETEx0r/+Za4zQQIAAAACCSEIRXv+eSk8XPr6a+mbb1wqwjZBwuzZ0p49nqsaAAAA4A5CEIrWuLF0113m+r33mrMclFGbNlKHDuZs2+++6+H6AQAAAC4iBKF4jz4qJSWZ44LeesulImwTJEyeLBmGB+sGAAAAuIgQhOIlJkpPPmmuP/64dPBgmYsYPFiKjZU2bpQyMjxaOwAAAMAlhCCUbMQIqUUL85pBtmsIlUFsrDR0qLk+ZYqH6wYAAAC4gBCEkoWFnZky+9VXpb//LnMRti5xn30m7dvnwboBAAAALiAEoXSXXy5dcYU5OcIDD5T58PPOk9q1k06dkt57zwv1AwAAAMqAEATn/Pe/5oVU58yRFiwo8+G26bInT5YWLpRmzDDHCFmtnq0mAAAAUBpCEJzTrJk0cqS5fs89ZZ4ye8gQKTJSWr/evBbrdddJl14qpaZK6emery4AAABQHEIQnPf441LlytIff0hvv12mQxcskHJzC2/fvl0aOJAgBAAAAN8hBMF5VapI48eb648+Kh0+7NRhVqs0enTRj9muHTRmDF3jAAAA4BuEIJTNHXdITZtKe/dKTz/t1CFLlkjbthX/uGFI2dnmfgAAAIC3EYJQNuHh5iQJkjRxorRpU6mH7NzpXNHO7gcAAAC4gxCEsrviCnPa7Lw86cEHS909Odm5Yp3dDwAAAHAHIQhlZ7FIL74ohYSYMxpkZJS4e+fOUp065mHFSUkx9wMAAAC8jRAE17RsKd1+u7l+zz0lzmoQGiq9/LK5XlwQat685JAEAAAAeAohCK574gkpIUFatUp6990Sd01Lk2bOlGrXdtxepYp5++230k03MUMcAAAAvI8QBNdVrSo99pi5/n//Jx05UuLuaWnSli3SwoXS9Onm7Z495npoqJmjhg41hxoBAAAA3mIxDNuVWsqfnJwcJSQk6PDhw4qPj/d3dYLTqVNm17iNG6Vx46T//MelYmbNkgYPNgNQ377Sxx9LkZEerisAAAAqrLJkA1qC4J6ICOmFF8z1F180m3pc0L+/9PnnUlSUedu3r3T8uOeqCQAAANgQguC+q6+WunWTcnOlsWNdLqZ3b+mrr6SYGGnePOnKK6WjRz1YTwAAAECEIHiCxSK99JI5ZfYnn0g//OByUd26mZMkxMWZM29ffrl0+LDnqgoAAAAQguAZrVtLN99srt9zj5Sf73JRnTpJ330nVa4sLVsmde8u7d/voXoCAAAg6BGC4DlPPWU24axYIX3wgVtFtW9vzh5Xtar066/SpZdKu3d7qJ4AAAAIaoQgeE6NGtIjj5jr48ZJx465VVybNtKiRVJysrRmjdSli7R9uwfqCQAAgKBGCIJnjR4t1a8v7dghPfec28W1aCEtXizVrSutXy9dconLE9ABAAAAkghB8LTISOn55831556TsrLcLrJRIzMINWggbd5sBqENG9wuFgAAAEGKEATPS0szk8rJk2a3OA+oV88MQs2aSdnZZvF//umRogEAABBkCEHwPNuU2RaLNH269NNPHim2dm1zjFDr1tKuXeYYoVWrPFI0AAAAggghCN5x3nnS8OHm+pgxkmF4pNjq1c1Z484/X9q3z5w17pdfPFI0AAAAggQhCN7z9NNSpUrSzz9LM2Z4rNgqVaQFC6SLLpIOHZJ69HDr+qwAAAAIMoQgeE9ysvR//2eujx0rHT/usaITEqR588yWoCNHpJ49zQusAgAAAKXxawhavHix+vTpo1q1aslisWj27Nn+rA684Z57zPmtt22T/vtfjxYdGyt99ZXUq5eZr668Upo716NPAQAAgArIryHo2LFjatOmjV5//XV/VgPeFB195npBzzxjTpmdkWF2j8vIkKxWt4ufPVvq21fKzZX69ZPS092sMwAAACo0i2F4aMS6mywWi2bNmqV+/fo5fUxOTo4SEhJ0+PBhxcfHe69ycI9hSBdfLP34oxQT49gtrk4d6eWXzWm13ZCXJ11/vfTxx1JoqPTee9J117lZbwAAAJQbZckG5WpMUG5urnJychwWlAMWi9lUIxUeF7R9uzRwoNvNN+Hh0ocfmhPSWa3Sv/4lvfOOW0UCAACggipXIWjChAlKSEiwLykpKf6uEpxhtUqvvlr0Y7aGyDFj3O4aFxoqvf22dMcdZrE33yzR0xIAAABnK1chaNy4cTp8+LB9yc7O9neV4IwlS8yJEYpjGFJ2trmfm0JCzOBzzz3m/VGjpBdecLtYAAAAVCBh/q5AWURGRioyMtLf1UBZ7dzp2f1KYbGYE9HFxJiXKnrgAbMX3qOPmo8BAAAguJWrliCUU8nJnt3PCRaL9O9/m4skPf64ecmiwJgGBAAAAP7k1xB09OhRrVq1SqtWrZIkZWZmatWqVcrKyvJnteBpnTubs8CV1gzzxhvSli0efeqHH5ZefNFcf+YZc+gRQQgAACC4+TUErVixQm3btlXbtm0lSffee6/atm2rxx57zJ/VgqeFhprTYEuFg1DB+59+KjVtKj34oHTokMee/p57pEmTzPVXXpFuu03Kz/dY8QAAAChn/BqCunbtKsMwCi3Tpk3zZ7XgDWlp0syZUu3ajtvr1JE++0z67Tepe3fp1Cnp+eelRo2k114zLwDkAbffLk2bZk6cMGWKOZX26dMeKRoAAADlTMBcLNUVXCy1HLJazVngdu40xwB17my2FElmP7W5c82ZDNatM7c1bSo995zUp49HZjX4+GNp6FCzGgMHmtcWCg0tvkoAAAAoH8qSDQhBCDynT5vNNY8/Lu3da2679FJzyrd/uk664/PPpUGDzEan8883w8/27Wcer1PH7L2Xlub2UwEAAMBHypINmB0OgScszLzi6YYN0kMPSZGR0sKFUrt2Zj+2kq455IS+fc0gFB4urVjhGIAk8/7AgVJ6ultPAwAAgABFCELgSkiQJkyQ1q+XrrvO7C737rtSkybmRX+OHnW56MsukxITi37M1jY6ZozZbQ4AAAAVCyEIga9ePXPwzs8/SxdfLJ04YV4AqFEj6X//cympLFlypqddUQxDys429wMAAEDFQghC+dGhg7R4sTmbXMOG0u7d0q23muOEvv22TEXt3OnZ/QAAAFB+EIJQvlgs5owFf/4pvfSSVLmytGaN1LOn1Lu39McfThWTnOzc0z3zjDl+iG5xAAAAFQchCOVTRIQ5aGfjRvNqqOHh0jffSG3aSCNGSLt2lXh4587mLHClzbr9++9Sv35Ss2bS669Lx4557BUAAADATwhBKN+qVJFefNFsGUpLk/Lzzem1GzeWnn5aOn68yMNCQ81psKXCQchiMZfJk83J6RITzaw1apSUkiKNG1d4RjkAAACUH4QgVAyNGpljhRYvltq3N2eOe+QR82Kr779vhqOzpKVJM2dKtWs7bq9Tx9x+663m5HTZ2dJrr5lPcfCg2UUuNVX617+klSt98/IAAADgOVwsFRVPfr708cdmM05WlrmtXTvzYqtduhTa3Wo1Z4HbudMcK9S5s9lSVNR+X35pNjwtXnxme5cu0r33SlddJYXwswIAAIBfcLFUBLeQEGnIEOmvv8ymnLg46ddfpa5dzQE+f//tsHuorOqqDA3RDHVVhkJV9CwIoaHmhVYXLTIvsjp0qHld10WLzO1Nm0pvvMG4IQAAgEBHSxAqvj17pPHjzUE+VquZXO64Q3rsMbNJZ/Roadu2M/vXqWMOGEpLK7XobdvMrnJvvSUdOmRuq1xZuu02cwzR2V3tAAAA4B1lyQaEIASPdeukBx6QvvrKvB8TU/TECbaZEmbOdCoISeYQpGnTpIkTpU2bzG1hYdK115qT1513ntu1BwAAQAnoDgcUpXlzc1DPggVS69bFzhwn2+8CY8Y4fYGg2Fiz5Wf9emn2bOmSS6TTp6UPPjCHI3XtKs2ZU+T8DAAAAPAxQhCCT/fu5uwGJTEMc1q4996TTp1yuuiC44aWL5euu85x3FCzZuV43JDVKmVkSDNmmLdcQRYAAJRTdIdDcJoxw0wozggNNa871KKF49K0qRQVVerhtim2J092HDd0++3SyJFFjxtydsY6n0lPd2vsFAAAgLcxJggoTUaGdOmlpe8XHS2dOFH0YyEhUoMGhcNRs2ZSpUqFdi9p3NC990pt25rb0tOle+62qv72JUrWTu1UsjJrd9ZLr4T6J2+kp0sDB57pJmjjwtgpAAAAbyEEAaWxWs0rnm7fXvjLvWR+wa9TR9q8Wdq1S/rzz8LLwYPFl5+aao5BKhiOmjeXEhJktUpffGH2yFuy5MwhXbtKF14o/f1MuiZqtFJ0ptUlW3U0Ri9r6Gdpvs0btvNUsAWoINt5ysz0c1MVAAAIdoQgwBm2Fg7JMQg508JhGNLu3WYYWrfOMRzt2VP8c9au7RCM1lla6MVvWmjq51VktUr9la6ZGijJcBiwly+zTrcnzdSk3WllyxtWq9kMdeSIlJNTeClp+44dZ5qtSrJwoZniAAAA/IQQBDirqLEuKSlmnzVXm1z27SscjP780wwUxbBWq6G/1Fx19y5XrI79E3kc5Uvar6ra+9CLalH3aOkBxrZ+9Khrr6Ms7rxTeuopqUoV7z8XAABAEQhBQFn4ahaCQ4eKDkdZWZ5/rqKEhUkJCVJ8/JklLs7x/tnbMjOlBx90rvzQUOnii6U+faSrrzYnkwAAAPARQhBQnhw9Kv31l7Y/9bZqz3mz1N3XqJX+VhMdUZxyFK/ciHjF1opXldR41Wgcp5QW8UppGa+IqmcFm8jIM139nFXa2CnJLL9uXemPPxy3N21qhqGrr5Y6dmTMEAAA8CpCEFAOWb/LUGiP0mesS79rob4+0VW//WbmjtzcwvuEhZnDjtq2lc4998ySmOhCxf4ZO2VIshT4uDAsFrPbnm3s1JYt5owPc+aYs++dPn2mjKQk6corzUB0+eVmKAMAAPAgQhBQHlmtOl4jVVH7tytEhf9b5suik0l1FLP7zExseXnS+vXSqlXSb7+duS1u4rrUVDMMFQxHKSmlNxD99GC66r44WrWsZ8ZObQ9NUfa9E3Xhc0WMnTp8WJo3zwxEc+c6VigiwpyevE8fc6lbt+QnBwAAcAIhCCiv0tNlDDBbXQoGoXyZrS6Wz0q/Jo9hmBdoXbXKMRxt2VL0/lWqOLYWtW1r9mQLD7dXSQMHShbDqs46c+2iH9RZ+ZbQ0i8TdPq0tHSpGYjmzJE2bnR8vE2bM93mzjvPvP4SAJRFwF1hGoA/EIKA8iw9Xcbo0bIUmLHOqJMiy8sT3boo6aFD0urVZ0LRqlXS2rWOvdZsIiOlVq3MfJKebh5blDJfJsgwzKarOXPMrnM//ijl5595PDn5TAtR9+7mxWpLYD1l1Zo3luj4pp2KaZisc+7srNAIvvgUwhdEVGRFzfJZp4708stcyLkofB6gAiMEAeWdj/5I5eaaE9Sd3Wp05EjZynnqKbOHW7Vq5pKY6OQcDHv3mt3lvvhC+uYb6dixM49FR5vjh/r0ka66SqpRw+HQorro7Qito6x7Xy66i16w4gsiKjJbU/XZX2Wcud5bMOLzABUcIQiAy/Lzza5zv/0mffCBNHt22csIC5OqVj0TimxL9eqFt1WrZnbJCzl10pxQwTa5QsE/0haLdMEF9um3f3p3vTq8cI2Ku6jsLw/MJAhJgf0FkV+j4S7b7JUFPysKKnNTdQXH5wGCACEIgEdkZJgtPKVp3txsVdq7t+ytSJI5DCgpqUAwqmrovNDVunDPHLXcNEfVs3912P+0QhUqazEXlbVoZ2gd1Tye6ZeucQHztzyQvyDyazQ8wdkPqE8/lfr1M3+d8bVA+UDg8wBBghAEwCNKu0xQUX83T56U9u0zA9HZy549hbcVN96ooFrarqv0pfroC12ubxWhvFKP+eKad5U8aqCq1IlR1armrNxlvUxSWfntb/mxY+Y/QsFl+XJzQorSxMebTXSJiVLlyuZtwaWkbVFRZa9rIP8aHYgC5Uu0Px09ak6osmGD47JmjZST41wZISFml9patQovycln1qtV89zkLP78cp+XJx04IO3fby6LFkmPPlr6cf/5j9SpU+ELaEdGeqeefB6UDZ8HpSIEAfAY298oyfHvlKf+RuXlOYamooJSweWq/VM1TTc5Xf5eVVWW6mqbpa72RqfoUHxdHatSV7k16iq/Tl1F1K2ppGohqlpV9iUpybytVMn54OT2LHolOXVKysoqHHRsy969LhbspshI58KSbYmLM0/Crl1Fl+fv7kuB9gUjUH8h98Z5OnGi6KCzYYP5PO4ICXGcgKUkYWFSzZpFh6WCS5UqJX84eOrLvWGYP3LYwsy+fWfWS1qcDYfOiohwDEa2cHT2ttK2x8aeCZmB3Dol8XlQThGCAHhUUZ+9KSnSxIm+/+z99b8Zand/6V1gTihK0TpZ6n6nFK7tqq0s1S207AqvqxNVUxRVPd4hJJ0dmCpXNmf4vnBnul7WaKXozInKVh2N0ctanpJW6G+5YZh/Z0+fypc1e4fyN50JNiFbMxWSlamw7EyF7d4uSylf4k5VStSx6vV1tFp9HUmqLyMvTy0XvFLq68976x2Ft2hsNsnZloMHHe8Xtc2bfzpuvFFq3978Vd42uMx2sr3VpSnQvmAE6i/k7pyn3Fxp06aig05xX4RtkpKkxo0dlwYNpP79zS+pJTVVb9xotors2FHysmeP8+/riAjHFqSCS40a0rBh0u7dRR9rsZjv6TffNP8vlRZoTp1yrk5FPU/lymf+36xbV/oxTZuatzk55lJwshpPsYWk0FDzx53SvPKK1K2b+VoqVzZboCtss34J9QnEz4MARAgC4HGB8qOY9ZRVu2NSVdNa/EVld4bWUc1jmxV64qiUlaWTG7J04q8sndqYpfytWQrdka2oPVmKPbRdIYa11Oc8pIRCASlbKfb1HaqlPvpCMzVQxU3WcKPeUWZUS6UamaprNZd6RqbqK1P1tFWRKvmLznFFa4tSZR5ReDmsRIf9Q2TVFqWqtoo/T9tUR/WVqYTKoUVOWFFwKTjRRVREvtlFyZmwVHDb9u3mlzp3VK5cOBydfVtwPTa29C9M/3zBMAzDYZyZYTGvz+XzLxiB+gu5M1/ErrrKrFdRQScrq+SQkZhYOOjYlsqVS66T5H5TdV6eGYRKC0v79jlXnidFRJz5IcDZpXLlM+8PV/o22447evRMKLItR44U3lbS9pycoq/H4Oq5sAUiW4vz2evFPRYXV3p3x0ALHIH6eRCgCEEAKrSfHkxXh+fNLz5nX1RWKsPscKdPm6kuK8tcsrOlrCwZWVnK32JuCz18sNRirLLIUEixkzUYUpHbHaqiUGWprjJVX1tUX1tD6ys7tL6yw+prR2R9HQivobBwi8LCzAvZhoU5Lmdv27NHarA6/Z9gVvR5GqiZmqWy/zGPjS09KBVc7DnE2YHsV1xhftHZu9f8wrlvn/lLvit/riIjSw5MVapId90lY+/eYv7tLLLUqW2OP8nPN78o5+WZ7x3beknbyrKvbdvmzdJnn5X+2vr0MZtkQ0PNJSTkzPrZ90t6zJl9JWnEiJIDQGio+W9UUqtlXFzxQScpybVf+H3dVJ2ba3bp3LHD/Pw4OyStW1d6y5ZktmQ1aeJcoClL39zieLtvc0kMwzxvBUPR4sXSPfeUfmytWuaxhw6ZgcAdISGO3XXPDkvx8dILL5Q8WLVaNendd83XdPr0mcVqdbzv7jbb9p07zV8gSzNlitkloWpV3150PFB+If0HIQhAhVfUdYK2h6Yo+96Jnp0e+8gRMxz9E5AKLdnZ5hdXJ5ysVEVG02ay1q2v/Hr1pdT6Uv36sjSor9B6dRQWFaawMM/8/bLljf4q3EUvSykao4mapTSlp5vfwwqOuypuYot9+1z7MTcy8p9AVMWqOb+nqlYJrVM7Qupo3VeZik0IVVSUzixhpxV98qCijuxVRM4+hewvUNHibk+W3h0SXhATU3zQqV7dK12ZAurCyc6G/YULpa5dvV0bR166GLdLyto6ZRhnWqBtrctFrRf3WLB8HtjGtdm6ayYnF71erZr7YSXQug2KEAQgSATEF5/8fOVPeksho+4sfdcPpitk6BAfVMrx+8XZkzUsUWcZltAy96AwDPP7RHEhqaggdeKEYxn95ZnWqYgIOYakf5bo6H/WIw0lhB9XdcteJRn7lJS/V5Wt+5SYt1cJefsUl7tPsSf3Kmn3OlXes965E2A7D7amN9sSFiZLwfsFthfa5sz2HTukGTNKr8jw4VLduuY/ttVqtsDY1j19f88eczxPaV5/XbrjDu+P2Sgg4L6H/fOfz9i2XZYiwr4hiywp/um+lJ4u3XO3VfW3n/k8yKzdWS+9Euqfc2XriirJUuDrqFe6op48WXpY+vVXcya90tSrZ4aIsDDz3/Dspvmzt7mzz+bNZqtmaRITnZtu1SY01PxRoqSglJxsjnEraixmoHUb/AchCAB8KUB/+fVn7xebY8fOBKNPP5Wef7701qnatc3Wo5MnzywnTnhnLoYuylCGSv+3u1zfaKG66bTCVFTnxpAQM79ERBTONmdvK22fMItVj09LVS2jhBYzSx29O968FpbFYv6bhoTI7fXiHq+2NkPdnir9PP391kJZO3dVRIT5miIjZV+PiDC/d3kyHwXo9zDPddn1oEA+Vz5p1XdGIH6Wl6XFLD/fnJDD1k1z586i1/fscX7GRIvFDEtnB6M33ig+dPlxnBIhCAB8KcB/+Q2Umf0Kfr8IUeHWqXyZ56ao7xe27vcFg1HBgFTU9uKWgvtv3mDVzF9Ln0CiRVSm8i2h9qE73uapFjNPKctEG7Z/x6JYLCo2IBVcinus4PawMOmtt0q+QHOVKubkYrYAVvAH99Juy7JvwXBn+87aflvRYf8eTSxytkhvCtSx9V69tIArAvWzPD1dxgCzxezszwOLJMtnZTxRttbd0sLSrl3ujcPyQ5dPQhAA+Jovu3WUUaCMW3V1gipvysiQXrm09MBx98I0+99yWyDLyzNnLy5qLoSitju776+/mq1mpbWYde1qjq3Pzz8zH4FhFL/uzuO7d0vN/yr9PH2fkKbQUPM15eY6PVyuQrDNJ2GxnJnVuqSwX7v2mcnKbMc6u16WfUNCzH+/r74q/TVcd51Uv/6Z4wq2EJ69uLvdMKSRI0ueMLJGDWn+fLOba1GhODzc8z0vA7UV78MB6ZpYTKge+lmad/685OebfZvPDkeLF5v/MKWZPl0a4psu4DaEIADwh0BqdglQgdBFryBbMOuwrfgvGL7+1d6dFjNv16m0YHZ2nQzjTNA7e8nNLXq7s/usXi3NnVt63Vu0MIdvFJyEq7jb0vZxd3IyeJ6tO2lRAam4lsbiFlvrYo8jxb/PF1VJ04svmuXbJlC0tQYWd9+ZfYq7L5mT12zbVvTngSvjO90WiN0G/0EIAgB/CZRmlwAWaFnRFsxCDKsuDoAuOYHYYuaNiTbc5Y/vYbaWsZIC1A8/OPfj9yuvSG3anJl/Ij/fcb2oba6ub9wovfNO6XUaMMBsobIde/Zie/1lfayox3ftktY7MSeJbYp9Wyj2lZJ+gAhENWqYM3wXvGRCcbfO7FPSMWEWq64alarkEsYt7gyto5rHM30+WREhCAAQ0AItKwZqMJMCo8UsEOsUiGExUOsViHVyJcQahvlaimoZLK7F0ZnFdqyzrYvnnGPOFVBwMsWCrYWu3C9vLY3OjFss2I3YVwhBAACUEcGs/NUp0IJZINcr0OpUUYKZJ9laywoGpUWLzOsil+b1182WxYLXXXbmtiz72m63bJFWrCi9e6wfhgQRggAAqAgCLZgFYp0CLZjZBGK9Aq1OBLPyWadAHLdoQwgCAABBI9CCWSDXK9DqRDArf3UKxGBmQwgCAABAuUAwK391CrRgZkMIAgAAAFwUaMEsEOsUaMFMIgQBAAAA8LJAC2ZlyQZhPqoTAAAAgAokNNT3kx94Soi/KwAAAAAAvkQIAgAAABBUCEEAAAAAggohCAAAAEBQIQQBAAAACCqEIAAAAABBhRAEAAAAIKgQggAAAAAEFUIQAAAAgKBCCAIAAAAQVAhBAAAAAIIKIQgAAABAUCEEAQAAAAgqYf6ugDsMw5Ak5eTk+LkmAAAAAPzJlglsGaEk5ToEHTlyRJKUkpLi55oAAAAACARHjhxRQkJCiftYDGeiUoDKz8/Xjh07FBcXJ4vF4u/qVGg5OTlKSUlRdna24uPj/V2doMA59y3Ot+9xzn2Pc+57nHPf4nz7XiCdc8MwdOTIEdWqVUshISWP+inXLUEhISGqU6eOv6sRVOLj4/3+Bg82nHPf4nz7Hufc9zjnvsc59y3Ot+8FyjkvrQXIhokRAAAAAAQVQhAAAACAoEIIglMiIyP1+OOPKzIy0t9VCRqcc9/ifPse59z3OOe+xzn3Lc6375XXc16uJ0YAAAAAgLKiJQgAAABAUCEEAQAAAAgqhCAAAAAAQYUQBAAAACCoEIKgCRMmqH379oqLi1P16tXVr18/rV+/vsRjpk2bJovF4rBERUX5qMbl3/jx4wudv2bNmpV4zKeffqpmzZopKipK55xzjubOneuj2lYMqamphc65xWLRyJEji9yf93jZLF68WH369FGtWrVksVg0e/Zsh8cNw9Bjjz2m5ORkRUdHq0ePHtqwYUOp5b7++utKTU1VVFSULrjgAv3yyy9eegXlT0nnPC8vT2PHjtU555yjSpUqqVatWrrhhhu0Y8eOEst05bMpmJT2Ph8+fHih89erV69Sy+V9XrTSzndRn+kWi0XPP/98sWXyHi+ZM98JT548qZEjRyopKUmxsbEaMGCAdu/eXWK5rv4N8CZCELRo0SKNHDlSP/30k+bPn6+8vDxdfvnlOnbsWInHxcfHa+fOnfZl69atPqpxxdCyZUuH8/fDDz8Uu++PP/6oIUOG6Oabb9Zvv/2mfv36qV+/fvrjjz98WOPybfny5Q7ne/78+ZKka665pthjeI8779ixY2rTpo1ef/31Ih9/7rnn9Morr+jNN9/Uzz//rEqVKqlnz546efJksWV+/PHHuvfee/X4449r5cqVatOmjXr27Kk9e/Z462WUKyWd8+PHj2vlypV69NFHtXLlSqWnp2v9+vW6+uqrSy23LJ9Nwaa097kk9erVy+H8zZgxo8QyeZ8Xr7TzXfA879y5U++8844sFosGDBhQYrm8x4vnzHfCe+65R1988YU+/fRTLVq0SDt27FBaWlqJ5bryN8DrDOAse/bsMSQZixYtKnafqVOnGgkJCb6rVAXz+OOPG23atHF6/0GDBhlXXnmlw7YLLrjAuO222zxcs+AxevRoo2HDhkZ+fn6Rj/Med50kY9asWfb7+fn5Rs2aNY3nn3/evu3QoUNGZGSkMWPGjGLL6dChgzFy5Ej7favVatSqVcuYMGGCV+pdnp19zovyyy+/GJKMrVu3FrtPWT+bgllR53zYsGFG3759y1QO73PnOPMe79u3r9GtW7cS9+E9XjZnfyc8dOiQER4ebnz66af2fdatW2dIMpYtW1ZkGa7+DfA2WoJQyOHDhyVJVapUKXG/o0ePql69ekpJSVHfvn21du1aX1SvwtiwYYNq1aqlBg0aaOjQocrKyip232XLlqlHjx4O23r27Klly5Z5u5oV0qlTp/TBBx/opptuksViKXY/3uOekZmZqV27djm8hxMSEnTBBRcU+x4+deqUfv31V4djQkJC1KNHD973Ljp8+LAsFosSExNL3K8sn00oLCMjQ9WrV1fTpk11xx13aP/+/cXuy/vcc3bv3q2vvvpKN998c6n78h533tnfCX/99Vfl5eU5vGebNWumunXrFvuedeVvgC8QguAgPz9fY8aMUadOndSqVati92vatKneeecdff755/rggw+Un5+viy66SNu2bfNhbcuvCy64QNOmTdM333yjSZMmKTMzU507d9aRI0eK3H/Xrl2qUaOGw7YaNWpo165dvqhuhTN79mwdOnRIw4cPL3Yf3uOeY3ufluU9vG/fPlmtVt73HnLy5EmNHTtWQ4YMUXx8fLH7lfWzCY569eql9957T999952effZZLVq0SL1795bVai1yf97nnvPuu+8qLi6u1G5ZvMedV9R3wl27dikiIqLQjyklvWdd+RvgC2F+e2YEpJEjR+qPP/4otX9sx44d1bFjR/v9iy66SM2bN9dbb72lp556ytvVLPd69+5tX2/durUuuOAC1atXT5988olTv2LBPW+//bZ69+6tWrVqFbsP73FUFHl5eRo0aJAMw9CkSZNK3JfPJvdce+219vVzzjlHrVu3VsOGDZWRkaHu3bv7sWYV3zvvvKOhQ4eWOoEN73HnOfudsLyiJQh2o0aN0pdffqmFCxeqTp06ZTo2PDxcbdu21caNG71Uu4otMTFRTZo0Kfb81axZs9DMK7t371bNmjV9Ub0KZevWrVqwYIFuueWWMh3He9x1tvdpWd7DVatWVWhoKO97N9kC0NatWzV//vwSW4GKUtpnE0rWoEEDVa1atdjzx/vcM5YsWaL169eX+XNd4j1enOK+E9asWVOnTp3SoUOHHPYv6T3ryt8AXyAEQYZhaNSoUZo1a5a+//571a9fv8xlWK1WrVmzRsnJyV6oYcV39OhRbdq0qdjz17FjR3333XcO2+bPn+/QUgHnTJ06VdWrV9eVV15ZpuN4j7uufv36qlmzpsN7OCcnRz///HOx7+GIiAi1a9fO4Zj8/Hx99913vO+dZAtAGzZs0IIFC5SUlFTmMkr7bELJtm3bpv379xd7/nife8bbb7+tdu3aqU2bNmU+lve4o9K+E7Zr107h4eEO79n169crKyur2PesK38DfMJvUzIgYNxxxx1GQkKCkZGRYezcudO+HD9+3L7P9ddfbzz00EP2+0888YQxb948Y9OmTcavv/5qXHvttUZUVJSxdu1af7yEcue+++4zMjIyjMzMTGPp0qVGjx49jKpVqxp79uwxDKPw+V66dKkRFhZmvPDCC8a6deuMxx9/3AgPDzfWrFnjr5dQLlmtVqNu3brG2LFjCz3Ge9w9R44cMX777Tfjt99+MyQZL774ovHbb7/ZZyJ75plnjMTEROPzzz83fv/9d6Nv375G/fr1jRMnTtjL6Natm/Hqq6/a73/00UdGZGSkMW3aNOPPP/80RowYYSQmJhq7du3y+esLRCWd81OnThlXX321UadOHWPVqlUOn+25ubn2Ms4+56V9NgW7ks75kSNHjPvvv99YtmyZkZmZaSxYsMA477zzjMaNGxsnT560l8H73Hmlfa4YhmEcPnzYiImJMSZNmlRkGbzHy8aZ74S33367UbduXeP77783VqxYYXTs2NHo2LGjQzlNmzY10tPT7fed+Rvga4QgGJKKXKZOnWrfp0uXLsawYcPs98eMGWPUrVvXiIiIMGrUqGFcccUVxsqVK31f+XJq8ODBRnJyshEREWHUrl3bGDx4sLFx40b742efb8MwjE8++cRo0qSJERERYbRs2dL46quvfFzr8m/evHmGJGP9+vWFHuM97p6FCxcW+TliO6f5+fnGo48+atSoUcOIjIw0unfvXujfoV69esbjjz/usO3VV1+1/zt06NDB+Omnn3z0igJfSec8MzOz2M/2hQsX2ss4+5yX9tkU7Eo658ePHzcuv/xyo1q1akZ4eLhRr14949Zbby0UZnifO6+0zxXDMIy33nrLiI6ONg4dOlRkGbzHy8aZ74QnTpww7rzzTqNy5cpGTEyM0b9/f2Pnzp2Fyil4jDN/A3zNYhiG4Z02JgAAAAAIPIwJAgAAABBUCEEAAAAAggohCAAAAEBQIQQBAAAACCqEIAAAAABBhRAEAAAAIKgQggAAAAAEFUIQAAAAgKBCCAIABA2LxaLZs2f7uxoAAD8jBAEAfGL48OGyWCyFll69evm7agCAIBPm7woAAIJHr169NHXqVIdtkZGRfqoNACBY0RIEAPCZyMhI1axZ02GpXLmyJLOr2qRJk9S7d29FR0erQYMGmjlzpsPxa9asUbdu3RQdHa2kpCSNGDFCR48eddjnnXfeUcuWLRUZGank5GSNGjXK4fF9+/apf//+iomJUePGjTVnzhz7YwcPHtTQoUNVrVo1RUdHq3HjxoVCGwCg/CMEAQACxqOPPqoBAwZo9erVGjp0qK699lqtW7dOknTs2DH17NlTlStX1vLly/Xpp59qwYIFDiFn0qRJGjlypEaMGKE1a9Zozpw5atSokcNzPPHEExo0aJB+//13XXHFFRo6dKgOHDhgf/4///xTX3/9tdatW6dJkyapatWqvjsBAACfsBiGYfi7EgCAim/48OH64IMPFBUV5bD9//7v//R///d/slgsuv322zVp0iT7YxdeeKHOO+88vfHGG5oyZYrGjh2r7OxsVapUSZI0d+5c9enTRzt27FCNGjVUu3Zt3Xjjjfr3v/9dZB0sFoseeeQRPfXUU5LMYBUbG6uvv/5avXr10tVXX62qVavqnXfe8dJZAAAEAsYEAQB85tJLL3UIOZJUpUoV+3rHjh0dHuvYsaNWrVolSVq3bp3atGljD0CS1KlTJ+Xn52v9+vWyWCzasWOHunfvXmIdWrdubV+vVKmS4uPjtWfPHknSHXfcoQEDBmjlypW6/PLL1a9fP1100UUuvVYAQOAiBAEAfKZSpUqFuqd5SnR0tFP7hYeHO9y3WCzKz8+XJPXu3Vtbt27V3LlzNX/+fHXv3l0jR47UCy+84PH6AgD8hzFBAICA8dNPPxW637x5c0lS8+bNtXr1ah07dsz++NKlSxUSEqKmTZsqLi5Oqamp+u6779yqQ7Vq1TRs2DB98MEHmjhxoiZPnuxWeQCAwENLEADAZ3Jzc7Vr1y6HbWFhYfbJBz799FOdf/75uvjii/Xhhx/ql19+0dtvvy1JGjp0qB5//HENGzZM48eP1969e3XXXXfp+uuvV40aNSRJ48eP1+23367q1aurd+/eOnLkiJYuXaq77rrLqfo99thjateunVq2bKnc3Fx9+eWX9hAGAKg4CEEAAJ/55ptvlJyc7LCtadOm+uuvvySZM7d99NFHuvPOO5WcnKwZM2aoRYsWkqSYmBjNmzdPo0ePVvv27RUTE6MBAwboxRdftJc1bNgwnTx5Ui+99JLuv/9+Va1aVQMHDnS6fhERERo3bpy2bNmi6Ohode7cWR999JEHXjkAIJAwOxwAICBYLBbNmjVL/fr183dVAAAVHGOCAAAAAAQVQhAAAACAoMKYIABAQKB3NgDAV2gJAgAAABBUCEEAAAAAggohCAAAAEBQIQQBAAAACCqEIAAAAABBhRAEAAAAIKgQggAAAAAEFUIQAAAAgKDy/4ojEzjMWnhuAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Extracting loss and validation loss from the history object\n",
    "training_loss = history.history['loss']\n",
    "validation_loss = history.history['val_loss']\n",
    "epochs = range(1, len(training_loss) + 1)\n",
    "\n",
    "# Plotting training and validation loss\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(epochs, training_loss, 'bo-', label='Training Loss')\n",
    "plt.plot(epochs, validation_loss, 'ro-', label='Validation Loss')\n",
    "plt.title('Training vs. Validation Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "ba30ad81-336c-44f6-98bb-e97249d86af3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 984us/step\n",
      "Accuracy: 0.7666666666666667\n",
      "f1_score: 0.5882352941176471\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      0.79      0.84       229\n",
      "           1       0.51      0.70      0.59        71\n",
      "\n",
      "    accuracy                           0.77       300\n",
      "   macro avg       0.70      0.75      0.71       300\n",
      "weighted avg       0.80      0.77      0.78       300\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>f1_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.786667</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>nn_shallow</td>\n",
       "      <td>0.820000</td>\n",
       "      <td>0.619718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>nn_deep</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.538462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>nn_deep_weighted</td>\n",
       "      <td>0.766667</td>\n",
       "      <td>0.588235</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Model  Accuracy  f1_Score\n",
       "0  LogisticRegression  0.786667  0.529412\n",
       "1          nn_shallow  0.820000  0.619718\n",
       "2             nn_deep  0.800000  0.538462\n",
       "3    nn_deep_weighted  0.766667  0.588235"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predict the labels for the test set\n",
    "predictions_prob = ann.predict(X_test)\n",
    "\n",
    "# Assuming `predictions` are the output probabilities from your model\n",
    "y_pred = (predictions_prob > 0.5).astype(int)\n",
    "\n",
    "# Evaluate the model\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred, average='binary')  # Ensure average='binary' for binary classification\n",
    "new_row = {'Model': 'nn_deep_weighted', 'Accuracy': accuracy, 'f1_Score': f1}\n",
    "new_row_df = pd.DataFrame([new_row])\n",
    "metrics_df = pd.concat([metrics_df, new_row_df], ignore_index=True)\n",
    "print(f'Accuracy: {accuracy}')\n",
    "print(f'f1_score: {f1}')\n",
    "print(classification_report(y_test, y_pred))\n",
    "metrics_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "44287cec-b566-4637-8bc3-b44329ea124c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training for fold 1 ...\n",
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 6ms/step - loss: 4.7043 - accuracy: 0.7935 - val_loss: 3.1504 - val_accuracy: 0.7796\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 2.2487 - accuracy: 0.8190 - val_loss: 1.6019 - val_accuracy: 0.7870\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.2278 - accuracy: 0.8167 - val_loss: 1.0016 - val_accuracy: 0.7852\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.8144 - accuracy: 0.8139 - val_loss: 0.7444 - val_accuracy: 0.7889\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6413 - accuracy: 0.8213 - val_loss: 0.6300 - val_accuracy: 0.7944\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5583 - accuracy: 0.8231 - val_loss: 0.5778 - val_accuracy: 0.7870\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5126 - accuracy: 0.8241 - val_loss: 0.5581 - val_accuracy: 0.7704\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4877 - accuracy: 0.8250 - val_loss: 0.5240 - val_accuracy: 0.7944\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4809 - accuracy: 0.8245 - val_loss: 0.5333 - val_accuracy: 0.7981\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4758 - accuracy: 0.8287 - val_loss: 0.5049 - val_accuracy: 0.7981\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4482 - accuracy: 0.8338 - val_loss: 0.5351 - val_accuracy: 0.7704\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4459 - accuracy: 0.8333 - val_loss: 0.5097 - val_accuracy: 0.8074\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4390 - accuracy: 0.8287 - val_loss: 0.5238 - val_accuracy: 0.7944\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4295 - accuracy: 0.8361 - val_loss: 0.5066 - val_accuracy: 0.7870\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4382 - accuracy: 0.8319 - val_loss: 0.5113 - val_accuracy: 0.7852\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4297 - accuracy: 0.8333 - val_loss: 0.4962 - val_accuracy: 0.7907\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4269 - accuracy: 0.8403 - val_loss: 0.5026 - val_accuracy: 0.7870\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4278 - accuracy: 0.8440 - val_loss: 0.5003 - val_accuracy: 0.8037\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4239 - accuracy: 0.8366 - val_loss: 0.5163 - val_accuracy: 0.7833\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4269 - accuracy: 0.8375 - val_loss: 0.4937 - val_accuracy: 0.8000\n",
      "Training for fold 2 ...\n",
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 12ms/step - loss: 4.6746 - accuracy: 0.7741 - val_loss: 2.9798 - val_accuracy: 0.8148\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 2.1384 - accuracy: 0.8014 - val_loss: 1.4566 - val_accuracy: 0.8185\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.1559 - accuracy: 0.8116 - val_loss: 0.8947 - val_accuracy: 0.8185\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.7837 - accuracy: 0.8134 - val_loss: 0.6735 - val_accuracy: 0.8241\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6281 - accuracy: 0.8148 - val_loss: 0.5715 - val_accuracy: 0.8296\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5586 - accuracy: 0.8208 - val_loss: 0.5180 - val_accuracy: 0.8333\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5195 - accuracy: 0.8162 - val_loss: 0.4992 - val_accuracy: 0.8241\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4925 - accuracy: 0.8185 - val_loss: 0.4737 - val_accuracy: 0.8204\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4865 - accuracy: 0.8167 - val_loss: 0.4725 - val_accuracy: 0.8167\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4721 - accuracy: 0.8213 - val_loss: 0.4609 - val_accuracy: 0.8204\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4625 - accuracy: 0.8241 - val_loss: 0.4616 - val_accuracy: 0.8204\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4595 - accuracy: 0.8236 - val_loss: 0.4647 - val_accuracy: 0.8148\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4664 - accuracy: 0.8167 - val_loss: 0.4582 - val_accuracy: 0.8074\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4597 - accuracy: 0.8162 - val_loss: 0.4430 - val_accuracy: 0.8333\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4470 - accuracy: 0.8222 - val_loss: 0.4411 - val_accuracy: 0.8296\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4498 - accuracy: 0.8222 - val_loss: 0.4382 - val_accuracy: 0.8204\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4452 - accuracy: 0.8287 - val_loss: 0.4398 - val_accuracy: 0.8389\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4404 - accuracy: 0.8287 - val_loss: 0.4462 - val_accuracy: 0.8222\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4427 - accuracy: 0.8194 - val_loss: 0.4512 - val_accuracy: 0.8241\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4426 - accuracy: 0.8255 - val_loss: 0.4372 - val_accuracy: 0.8241\n",
      "Training for fold 3 ...\n",
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 6ms/step - loss: 4.6435 - accuracy: 0.7787 - val_loss: 3.0157 - val_accuracy: 0.8074\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 2.1844 - accuracy: 0.8111 - val_loss: 1.5208 - val_accuracy: 0.8037\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.2115 - accuracy: 0.8139 - val_loss: 0.9478 - val_accuracy: 0.8074\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.8195 - accuracy: 0.8162 - val_loss: 0.7209 - val_accuracy: 0.7981\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6541 - accuracy: 0.8157 - val_loss: 0.5881 - val_accuracy: 0.8056\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5739 - accuracy: 0.8176 - val_loss: 0.5424 - val_accuracy: 0.8037\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5303 - accuracy: 0.8227 - val_loss: 0.5140 - val_accuracy: 0.8019\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5023 - accuracy: 0.8213 - val_loss: 0.4922 - val_accuracy: 0.8056\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4859 - accuracy: 0.8208 - val_loss: 0.4701 - val_accuracy: 0.8111\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4834 - accuracy: 0.8227 - val_loss: 0.4738 - val_accuracy: 0.8056\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4663 - accuracy: 0.8199 - val_loss: 0.4489 - val_accuracy: 0.8037\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4581 - accuracy: 0.8282 - val_loss: 0.4462 - val_accuracy: 0.8111\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4600 - accuracy: 0.8236 - val_loss: 0.4508 - val_accuracy: 0.8111\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4452 - accuracy: 0.8264 - val_loss: 0.4636 - val_accuracy: 0.8130\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4564 - accuracy: 0.8227 - val_loss: 0.4385 - val_accuracy: 0.8074\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4472 - accuracy: 0.8259 - val_loss: 0.4512 - val_accuracy: 0.8074\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4507 - accuracy: 0.8162 - val_loss: 0.4418 - val_accuracy: 0.8111\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4397 - accuracy: 0.8301 - val_loss: 0.4437 - val_accuracy: 0.7981\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4362 - accuracy: 0.8282 - val_loss: 0.4329 - val_accuracy: 0.8037\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4294 - accuracy: 0.8361 - val_loss: 0.4343 - val_accuracy: 0.8111\n",
      "Training for fold 4 ...\n",
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 6ms/step - loss: 4.7635 - accuracy: 0.7847 - val_loss: 3.1921 - val_accuracy: 0.7981\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 2.3009 - accuracy: 0.8125 - val_loss: 1.6460 - val_accuracy: 0.8093\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.2725 - accuracy: 0.8231 - val_loss: 1.0409 - val_accuracy: 0.8093\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.8547 - accuracy: 0.8227 - val_loss: 0.7806 - val_accuracy: 0.8019\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6674 - accuracy: 0.8315 - val_loss: 0.6596 - val_accuracy: 0.8111\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5787 - accuracy: 0.8296 - val_loss: 0.6015 - val_accuracy: 0.8056\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5411 - accuracy: 0.8139 - val_loss: 0.5611 - val_accuracy: 0.8037\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5035 - accuracy: 0.8227 - val_loss: 0.5660 - val_accuracy: 0.7889\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4780 - accuracy: 0.8259 - val_loss: 0.5200 - val_accuracy: 0.8093\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4667 - accuracy: 0.8306 - val_loss: 0.5214 - val_accuracy: 0.8074\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4526 - accuracy: 0.8306 - val_loss: 0.5062 - val_accuracy: 0.7981\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4503 - accuracy: 0.8157 - val_loss: 0.5009 - val_accuracy: 0.8074\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4423 - accuracy: 0.8329 - val_loss: 0.5113 - val_accuracy: 0.8037\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4270 - accuracy: 0.8356 - val_loss: 0.4941 - val_accuracy: 0.8130\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4331 - accuracy: 0.8347 - val_loss: 0.4913 - val_accuracy: 0.8148\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4382 - accuracy: 0.8190 - val_loss: 0.4885 - val_accuracy: 0.7963\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4267 - accuracy: 0.8380 - val_loss: 0.5015 - val_accuracy: 0.8000\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4282 - accuracy: 0.8347 - val_loss: 0.5027 - val_accuracy: 0.7944\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4349 - accuracy: 0.8241 - val_loss: 0.5163 - val_accuracy: 0.7907\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4279 - accuracy: 0.8301 - val_loss: 0.5110 - val_accuracy: 0.8056\n",
      "Training for fold 5 ...\n",
      "Epoch 1/20\n",
      "44/44 [==============================] - 1s 6ms/step - loss: 4.4716 - accuracy: 0.7722 - val_loss: 2.7709 - val_accuracy: 0.8370\n",
      "Epoch 2/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.9962 - accuracy: 0.8000 - val_loss: 1.3438 - val_accuracy: 0.8333\n",
      "Epoch 3/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 1.0797 - accuracy: 0.8032 - val_loss: 0.8210 - val_accuracy: 0.8389\n",
      "Epoch 4/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.7511 - accuracy: 0.8079 - val_loss: 0.6358 - val_accuracy: 0.8500\n",
      "Epoch 5/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.6067 - accuracy: 0.8088 - val_loss: 0.5384 - val_accuracy: 0.8481\n",
      "Epoch 6/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5390 - accuracy: 0.8102 - val_loss: 0.4997 - val_accuracy: 0.8481\n",
      "Epoch 7/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.5202 - accuracy: 0.8046 - val_loss: 0.4857 - val_accuracy: 0.8333\n",
      "Epoch 8/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4909 - accuracy: 0.8088 - val_loss: 0.4630 - val_accuracy: 0.8389\n",
      "Epoch 9/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4733 - accuracy: 0.8106 - val_loss: 0.4548 - val_accuracy: 0.8444\n",
      "Epoch 10/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4682 - accuracy: 0.8139 - val_loss: 0.4419 - val_accuracy: 0.8463\n",
      "Epoch 11/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4646 - accuracy: 0.8194 - val_loss: 0.4507 - val_accuracy: 0.8500\n",
      "Epoch 12/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4640 - accuracy: 0.8069 - val_loss: 0.4406 - val_accuracy: 0.8444\n",
      "Epoch 13/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4499 - accuracy: 0.8176 - val_loss: 0.4483 - val_accuracy: 0.8463\n",
      "Epoch 14/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4481 - accuracy: 0.8245 - val_loss: 0.4419 - val_accuracy: 0.8519\n",
      "Epoch 15/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4516 - accuracy: 0.8190 - val_loss: 0.4491 - val_accuracy: 0.8130\n",
      "Epoch 16/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4497 - accuracy: 0.8269 - val_loss: 0.4391 - val_accuracy: 0.8500\n",
      "Epoch 17/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4523 - accuracy: 0.8176 - val_loss: 0.4595 - val_accuracy: 0.8167\n",
      "Epoch 18/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4483 - accuracy: 0.8153 - val_loss: 0.4595 - val_accuracy: 0.8037\n",
      "Epoch 19/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4510 - accuracy: 0.8171 - val_loss: 0.4441 - val_accuracy: 0.8241\n",
      "Epoch 20/20\n",
      "44/44 [==============================] - 0s 3ms/step - loss: 0.4380 - accuracy: 0.8176 - val_loss: 0.4397 - val_accuracy: 0.8426\n"
     ]
    }
   ],
   "source": [
    "# Define the K-fold cross validator\n",
    "num_folds = 5\n",
    "kfold = KFold(n_splits=num_folds, shuffle=True)\n",
    "\n",
    "# Lists to store results of each fold\n",
    "all_y_test = []\n",
    "all_y_pred = []\n",
    "\n",
    "# K-fold Cross Validation model evaluation\n",
    "fold_no = 1\n",
    "X= X_train\n",
    "y = y_train\n",
    "for train, test in kfold.split(X, y):\n",
    "\n",
    "    # Define the model architecture\n",
    "    ann = tf.keras.models.Sequential([\n",
    "        tf.keras.layers.Dense(units=240, activation='relu', kernel_regularizer=regularizers.l2(0.01), input_shape=(X.shape[1],)),\n",
    "        tf.keras.layers.Dense(units=120, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "        tf.keras.layers.Dense(units=60, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "        tf.keras.layers.Dense(units=30, activation='relu', kernel_regularizer=regularizers.l2(0.01)),\n",
    "        tf.keras.layers.Dense(units=1, activation='sigmoid')\n",
    "    ])\n",
    "\n",
    "    # Compile the model\n",
    "    ann.compile(optimizer=keras.optimizers.Adam(learning_rate=0.001), loss='binary_crossentropy', metrics=['accuracy'])\n",
    "  \n",
    "    # Fit the model\n",
    "    print(f'Training for fold {fold_no} ...')\n",
    "    ann.fit(X[train], y[train], batch_size=batch_size, epochs=n_epochs, verbose=1, validation_data=(X[test], y[test]))\n",
    "    \n",
    "    # Generate predictions\n",
    "    y_pred = model.predict(X[test])\n",
    "    y_pred_binary = (y_pred > 0.5).astype(int)\n",
    "    \n",
    "    # Store the true labels and predictions\n",
    "    all_y_test.extend(y[test])\n",
    "    all_y_pred.extend(y_pred_binary.flatten())  # Flatten to ensure it's a 1D array\n",
    "\n",
    "    # Increase fold number\n",
    "    fold_no += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "fa1fbe9d-db9f-4b72-b961-97916f0227ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overall performance:\n",
      "Accuracy: 0.882962962962963\n",
      "f1_score: 0.7308347529812607\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      0.95      0.93      2063\n",
      "           1       0.80      0.67      0.73       637\n",
      "\n",
      "    accuracy                           0.88      2700\n",
      "   macro avg       0.85      0.81      0.83      2700\n",
      "weighted avg       0.88      0.88      0.88      2700\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>f1_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.786667</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>nn_shallow</td>\n",
       "      <td>0.820000</td>\n",
       "      <td>0.619718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>nn_deep</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.538462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>nn_deep_weighted</td>\n",
       "      <td>0.766667</td>\n",
       "      <td>0.588235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>nn_deep_weighted_5fold</td>\n",
       "      <td>0.882963</td>\n",
       "      <td>0.730835</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Model  Accuracy  f1_Score\n",
       "0      LogisticRegression  0.786667  0.529412\n",
       "1              nn_shallow  0.820000  0.619718\n",
       "2                 nn_deep  0.800000  0.538462\n",
       "3        nn_deep_weighted  0.766667  0.588235\n",
       "4  nn_deep_weighted_5fold  0.882963  0.730835"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Performance assessment\n",
    "# Now, all_y_test and all_y_pred contain all the test labels and predictions across all folds\n",
    "print('Overall performance:')\n",
    "# Evaluate the model\n",
    "accuracy = accuracy_score(all_y_test, all_y_pred)\n",
    "f1 = f1_score(all_y_test, all_y_pred, average='binary')  \n",
    "new_row = {'Model': 'nn_deep_weighted_5fold', 'Accuracy': accuracy, 'f1_Score': f1}\n",
    "new_row_df = pd.DataFrame([new_row])\n",
    "metrics_df = pd.concat([metrics_df, new_row_df], ignore_index=True)\n",
    "print(f'Accuracy: {accuracy}')\n",
    "print(f'f1_score: {f1}')\n",
    "print(classification_report(all_y_test, all_y_pred))\n",
    "metrics_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "0500cc6c-491e-472b-8c1b-c516ce4f46e0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PDB ID: 7M7B, seq: ['QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEINHSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARRWWLRGAFDIWGQGTTVTVSS', 'NFMLTQPHSVSASPGKTVTIPCTGSSGNIASNYVQWYQQRPGSAPTTVIYEDNQRPSGVPDRFSGSIDSSSNSASLTISGLKTEDEADYYCQSYDNNIQVFGGGTKLTVL']\n",
      "totalCnt: 540, matchCnt: 189, Matched ratio: 0.35\n"
     ]
    }
   ],
   "source": [
    "# show 3D structure\n",
    "offset = int(MAXCNT * 0.8)\n",
    "matchCnt = sum([1 for pred, test in zip(y_pred, y_test) if pred == test])\n",
    "totalCnt = len(y_pred)\n",
    "\n",
    "# Find the first non-empty pdbid and its sequence\n",
    "for i, pdbid in enumerate(pdbids[offset:], start=offset):\n",
    "    if pdbid:  # Checks if pdbid is not empty\n",
    "        seq = seqs[i]\n",
    "        print(f\"PDB ID: {pdbid}, seq: {seq}\")\n",
    "\n",
    "        idStr = f'mmdbid={pdbid}'\n",
    "        # Ensure seq is a string for concatenation\n",
    "        seqStr = ''.join(seq) if isinstance(seq, list) else seq\n",
    "        cmdStr = f'select :{seqStr} | name test; color F00; style proteins sphere'\n",
    "        view = icn3dpy.view(q=idStr, command=cmdStr)\n",
    "        view\n",
    "        break  # Exit the loop after the first match\n",
    "\n",
    "print(f\"totalCnt: {totalCnt}, matchCnt: {matchCnt}, Matched ratio: {matchCnt/totalCnt}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "cb8fe2ef-5f1b-4639-a71a-1bf94449be7d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mmdbid=7M7B select :QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEINHSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARRWWLRGAFDIWGQGTTVTVSSNFMLTQPHSVSASPGKTVTIPCTGSSGNIASNYVQWYQQRPGSAPTTVIYEDNQRPSGVPDRFSGSIDSSSNSASLTISGLKTEDEADYYCQSYDNNIQVFGGGTKLTVL | name test; color F00; style proteins sphere\n"
     ]
    }
   ],
   "source": [
    "idStr = 'mmdbid=' + pdbid\n",
    "cmdStr = 'select :' + seqStr + ' | name test; color F00; style proteins sphere'\n",
    "print (idStr, cmdStr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "11b3e5b3-eea3-4a56-a996-ed73e8c4f50a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"icn3dviewer17093052305273967\" style=\"position: relative; width: 640px; min-height: 530px; height: auto\">\n        <p id=\"icn3dwarning17093052305273967\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    //tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('link')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n  });\n};\n\nvar loadCssAsync = function(uri){\n  return new Promise((resolve, reject) => {\nvar tag = document.createElement('link');\ntag.rel = 'stylesheet';\ntag.href = uri;\n//tag.async = true;\ntag.onload = () => {\n  resolve();\n};\nvar firstScriptTag = document.getElementsByTagName('script')[0];\nfirstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n  });\n};\n\nif(typeof js === 'undefined') {\n  js = loadScriptAsync(\"https://www.ncbi.nlm.nih.gov/Structure/icn3d/es5/icn3d_all_full.min.js\");\n  \n  css1 = loadCssAsync(\"https://www.ncbi.nlm.nih.gov/Structure/icn3d/lib/jquery-ui-1.13.2.min.css\");\n  css2 = loadCssAsync(\"https://www.ncbi.nlm.nih.gov/Structure/icn3d/icn3d.css\");\n}\n\nvar viewer17093052305273967 = null;\nvar warn = document.getElementById(\"icn3dwarning17093052305273967\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n\ncss1\n.then(function() { return css2; })\n.then(function() { return js; })\n.then(function() {\ncfg = {divid: \"icn3dviewer17093052305273967\", idname: \"mmdbid\", idvalue: \"7M7B\", mmdbid: \"7M7B\", width: \"640px\", height: \"480px\", mobilemenu: 1, notebook: 1, command: 'select :QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEINHSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARRWWLRGAFDIWGQGTTVTVSSNFMLTQPHSVSASPGKTVTIPCTGSSGNIASNYVQWYQQRPGSAPTTVIYEDNQRPSGVPDRFSGSIDSSSNSASLTISGLKTEDEADYYCQSYDNNIQVFGGGTKLTVL | name test; color F00; style proteins sphere', };\nviewer17093052305273967 = new icn3d.iCn3DUI(cfg);\nviewer17093052305273967.show3DStructure(cfg.pdb);\n});\n</script>",
      "text/html": [
       "<div id=\"icn3dviewer17093052305273967\" style=\"position: relative; width: 640px; min-height: 530px; height: auto\">\n",
       "        <p id=\"icn3dwarning17093052305273967\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    //tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('link')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "  });\n",
       "};\n",
       "\n",
       "var loadCssAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "var tag = document.createElement('link');\n",
       "tag.rel = 'stylesheet';\n",
       "tag.href = uri;\n",
       "//tag.async = true;\n",
       "tag.onload = () => {\n",
       "  resolve();\n",
       "};\n",
       "var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "  });\n",
       "};\n",
       "\n",
       "if(typeof js === 'undefined') {\n",
       "  js = loadScriptAsync(\"https://www.ncbi.nlm.nih.gov/Structure/icn3d/es5/icn3d_all_full.min.js\");\n",
       "  \n",
       "  css1 = loadCssAsync(\"https://www.ncbi.nlm.nih.gov/Structure/icn3d/lib/jquery-ui-1.13.2.min.css\");\n",
       "  css2 = loadCssAsync(\"https://www.ncbi.nlm.nih.gov/Structure/icn3d/icn3d.css\");\n",
       "}\n",
       "\n",
       "var viewer17093052305273967 = null;\n",
       "var warn = document.getElementById(\"icn3dwarning17093052305273967\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "\n",
       "css1\n",
       ".then(function() { return css2; })\n",
       ".then(function() { return js; })\n",
       ".then(function() {\n",
       "cfg = {divid: \"icn3dviewer17093052305273967\", idname: \"mmdbid\", idvalue: \"7M7B\", mmdbid: \"7M7B\", width: \"640px\", height: \"480px\", mobilemenu: 1, notebook: 1, command: 'select :QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEINHSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCARRWWLRGAFDIWGQGTTVTVSSNFMLTQPHSVSASPGKTVTIPCTGSSGNIASNYVQWYQQRPGSAPTTVIYEDNQRPSGVPDRFSGSIDSSSNSASLTISGLKTEDEADYYCQSYDNNIQVFGGGTKLTVL | name test; color F00; style proteins sphere', };\n",
       "viewer17093052305273967 = new icn3d.iCn3DUI(cfg);\n",
       "viewer17093052305273967.show3DStructure(cfg.pdb);\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<icn3dpy.view at 0x7f0d841e1a90>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "view = icn3dpy.view(q=idStr,command=cmdStr)\n",
    "view"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb2526a8-8ca8-4be1-917e-572d2043de0e",
   "metadata": {},
   "source": [
    "# Optional"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "338ffb76-d87f-49d6-8f64-3ab283ad9c2a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# let dig more into the results:\n",
    "from sklearn.metrics import roc_curve\n",
    "y_probs = ann.predict(X_test)\n",
    "# Calculate FPR, TPR, and thresholds for the ROC curve\n",
    "fpr, tpr, roc_thresholds = roc_curve(y_test, y_probs)\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Plotting the ROC curve\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(fpr, tpr, label='ROC Curve')\n",
    "plt.plot([0, 1], [0, 1], 'k--', label='Random Chance')  # random chance line\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5717914-5a2a-4052-bf57-56b8042d4585",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a range of thresholds to evaluate\n",
    "thresholds = np.linspace(0, 1, 101)\n",
    "\n",
    "# Calculate F1 scores for each threshold\n",
    "f1_scores = [f1_score(y_test, y_probs > threshold) for threshold in thresholds]\n",
    "\n",
    "# Plotting F1 scores against thresholds\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(thresholds, f1_scores, label='F1 Score')\n",
    "plt.xlabel('Threshold')\n",
    "plt.ylabel('F1 Score')\n",
    "plt.title('F1 Score vs. Threshold')\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "81cc9a8b-7508-47b8-9aad-44b040ffbd1c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# let find the optimal threshold for balancing false positives and false negatives \n",
    "# to do that we are using calibration plot.\n",
    "# Step 1: we should predict the probablities\n",
    "y_probs = ann.predict(X_test)\n",
    "# Step 2: Calculate Precision, Recall, and Thresholds\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "# Calculate precision, recall, and thresholds\n",
    "precision, recall, thresholds = precision_recall_curve(y_test, y_probs)\n",
    "# Step 3: Create a Calibration Plot\n",
    "\n",
    "from sklearn.calibration import calibration_curve\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Calibration curve\n",
    "prob_true, prob_pred = calibration_curve(y_test, y_probs, n_bins=10)\n",
    "\n",
    "# Plotting the calibration curve\n",
    "plt.plot(prob_pred, prob_true, marker='o', linewidth=1, label='ANN')\n",
    "plt.plot([0, 1], [0, 1], linestyle='--', label='Perfectly calibrated')\n",
    "plt.xlabel('Mean predicted probability')\n",
    "plt.ylabel('Fraction of positives')\n",
    "plt.title('Calibration plot')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}