llama-2-hf-tgi/llama-2-13b-chat-hf/2-sagemaker-realtime-inference-llama-2-13b-chat-hf-tgi-streaming-response.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7e682cf5-0f37-4227-ab76-9eae1482c1f1",
   "metadata": {},
   "source": [
    "# SageMaker Realtime Inference Llama 2 13B chat HF TGI using response streaming"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a973c90-2e61-4829-b15b-86d2525c3418",
   "metadata": {},
   "source": [
    "## Set up development environment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10f4e66f-fd39-4a9e-a8ad-886d34978351",
   "metadata": {},
   "source": [
    "Upgrade `pip` and install the latest version of `sagemaker` and `boto3` packages."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6a05da92-1d79-41d7-8a24-3d9de3c73493",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "!pip install -Uq pip\n",
    "!pip install -Uq boto3 sagemaker"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a24e68c2-ad02-4683-9bd0-4da9179e8309",
   "metadata": {},
   "source": [
    "Restore the `endpoint_name` from the deployment notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8d58d853-b948-49bf-9887-23e955055825",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "%store -r \\\n",
    "endpoint_name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "07a86527-22ad-41b5-9826-fd7aaba8c77e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "endpoint_name"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b9e1a9f-4c82-4d6d-9780-b7cafc2450c6",
   "metadata": {},
   "source": [
    "## Setup Code for Realtime Inference for response streaming"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ad0ad99-1d5f-481f-83ce-b1b95d91f937",
   "metadata": {},
   "source": [
    "Amazon SageMaker's new `InvokeEndpointWithResponseStream` API allows developers to stream responses back from SageMaker models, which can help to improve customer satisfaction by reducing the perceived latency. This is especially important for applications built with generative AI models, where immediate processing is more important than waiting for the entire response.\n",
    "\n",
    "The `LineIterator` helper class will parse the response stream received from the inference request made via `InvokeEndpointWithResponseStream` API."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b4204e00-cb88-48c7-b9f9-6c5922bc1ae7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import boto3\n",
    "from utils.LineIterator import LineIterator\n",
    "import json\n",
    "sagemaker_runtime = boto3.client('sagemaker-runtime')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "79c682c2-0a5a-4320-b2a8-23f06af5fef3",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def get_realtime_response_stream(sagemaker_runtime, endpoint_name, payload):\n",
    "    response_stream = sagemaker_runtime.invoke_endpoint_with_response_stream(\n",
    "        EndpointName=endpoint_name,\n",
    "        Body=json.dumps(payload), \n",
    "        ContentType=\"application/json\",\n",
    "        CustomAttributes='accept_eula=true'\n",
    "    )\n",
    "    return response_stream\n",
    "\n",
    "def print_response_stream(response_stream):\n",
    "    event_stream = response_stream['Body']\n",
    "    start_json = b'{'\n",
    "    stop_token = '</s>'\n",
    "    for line in LineIterator(event_stream):\n",
    "        if line != b'' and start_json in line:\n",
    "            data = json.loads(line[line.find(start_json):].decode('utf-8'))\n",
    "            if data['token']['text'] != stop_token:\n",
    "                print(data['token']['text'],end='')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a80f8bd-9642-44c6-9064-1e1ec02fe57f",
   "metadata": {},
   "source": [
    "## Prepare Prompt and instructions"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c1eefb5-1619-40ba-b7b4-ed173366fb52",
   "metadata": {},
   "source": [
    "To prompt Llama 2 Chat, you need to have following prompt format\n",
    "\n",
    "```\n",
    "<s>[INST] <<SYS>>\n",
    "{{ system_prompt }}\n",
    "<</SYS>>\n",
    "\n",
    "{{ user_message }} [/INST]\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d542e7e0-9be5-46ff-9af1-12849dda30fb",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def build_llama2_prompt(instructions):\n",
    "    stop_token = \"</s>\"\n",
    "    start_token = \"<s>\"\n",
    "    startPrompt = f\"{start_token}[INST] \"\n",
    "    endPrompt = \" [/INST]\"\n",
    "    conversation = []\n",
    "    for index, instruction in enumerate(instructions):\n",
    "        if instruction[\"role\"] == \"system\" and index == 0:\n",
    "            conversation.append(f\"<<SYS>>\\n{instruction['content']}\\n<</SYS>>\\n\\n\")\n",
    "        elif instruction[\"role\"] == \"user\":\n",
    "            conversation.append(instruction[\"content\"].strip())\n",
    "        else:\n",
    "            conversation.append(f\"{endPrompt} {instruction['content'].strip()} {stop_token}{startPrompt}\")\n",
    "\n",
    "    return startPrompt + \"\".join(conversation) + endPrompt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "87318429-3537-4079-bfb7-5ffd4c7f0edb",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def get_instructions(user_content):\n",
    "    \n",
    "    '''\n",
    "    Note: We are creating a fresh user content everytime by initializing instructions for every user_content.\n",
    "    This is to avoid past user_content when you are inferencing multiple times with new ask everytime.\n",
    "    ''' \n",
    "    \n",
    "    system_content = '''\n",
    "    You are a friendly and knowledgeable email marketing agent, Mr.MightyMark, working at AnyCompany. \n",
    "    Your goal is to send email to subscribers to help them understand the value of the new product and generate excitement for the launch.\n",
    "\n",
    "    Here are some tips on how to achieve your goal:\n",
    "\n",
    "    Be personal. Address each subscriber by name and use a friendly and conversational tone.\n",
    "    Be informative. Explain the key features and benefits of the new product in a clear and concise way.\n",
    "    Be persuasive. Highlight how the new product can solve the subscriber's problems or improve their lives.\n",
    "    Be engaging. Use emojis to make your emails more visually appealing and interesting to read.\n",
    "\n",
    "    By following these tips, you can use email marketing to help your company launch a successful software product.\n",
    "    '''\n",
    "\n",
    "    instructions = [\n",
    "        { \"role\": \"system\",\"content\": f\"{system_content} \"},\n",
    "    ]\n",
    "    \n",
    "    instructions.append({\"role\": \"user\", \"content\": f\"{user_content}\"})\n",
    "    \n",
    "    return instructions\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b57349c-5f95-49fc-8c87-1e8308c50dc9",
   "metadata": {},
   "source": [
    "## Inference the Llama 2 13B chat HF TGI SageMaker endpoint for Streaming Response"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36ce7ba9-6a48-404a-ad83-7bee465d8adb",
   "metadata": {},
   "source": [
    "### Inference example 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f8e7f6cf-80cb-480a-b050-632ef23386e1",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "user_ask_1 = f'''\n",
    "AnyCompany recently announced new service launch named AnyCloud Internet Service.\n",
    "Write a short email about the product launch with Call to action to Alice Smith, whose email is alice.smith@example.com\n",
    "Mention the Coupon Code: EARLYB1RD to get 20% for 1st 3 months.\n",
    "'''\n",
    "instructions = get_instructions(user_ask_1)\n",
    "prompt = build_llama2_prompt(instructions)\n",
    "print(prompt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "02751b2b-3969-4824-b66e-daa0bab509d5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "inference_params = {\n",
    "        \"do_sample\": True,\n",
    "        \"top_p\": 0.6,\n",
    "        \"temperature\": 0.9,\n",
    "        \"top_k\": 50,\n",
    "        \"max_new_tokens\": 512,\n",
    "        \"repetition_penalty\": 1.03,\n",
    "        \"stop\": [\"</s>\"],\n",
    "        \"return_full_text\": False\n",
    "    }\n",
    "\n",
    "payload = {\n",
    "    \"inputs\":  prompt,\n",
    "    \"parameters\": inference_params,\n",
    "    \"stream\": True ## <-- to have response stream.\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d8b79b1-5cde-44f0-b129-48b9b1a605d8",
   "metadata": {},
   "source": [
    "As we are interested in streaming response, the request payload must provide a key value pair with **\"stream\": True**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7aea65dc-ff94-4fec-90ef-2f379185eb2b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "resp = get_realtime_response_stream(sagemaker_runtime, endpoint_name, payload)\n",
    "print_response_stream(resp)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "062bfe00-01fd-466b-a46c-a0a668c912f4",
   "metadata": {},
   "source": [
    "### Inference example 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "72b5fd71-f4c7-4d50-a69c-4756cdd42885",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "user_ask_2 = f'''\n",
    "AnyCompany recently announced new service launch named AnyCloud Streaming Service.\n",
    "Write a short email about the product launch with Call to action to Alice Smith, whose email is alice.smith@example.com\n",
    "Mention the Coupon Code: STREAM2DREAM to get 15% for 1st 6 months.\n",
    "'''\n",
    "\n",
    "instructions = get_instructions(user_ask_2)\n",
    "prompt = build_llama2_prompt(instructions)\n",
    "\n",
    "inference_params = {\n",
    "        \"do_sample\": True,\n",
    "        \"top_p\": 0.6,\n",
    "        \"temperature\": 0.9,\n",
    "        \"top_k\": 50,\n",
    "        \"max_new_tokens\": 512,\n",
    "        \"repetition_penalty\": 1.03,\n",
    "        \"stop\": [\"</s>\"],\n",
    "        \"return_full_text\": False\n",
    "    }\n",
    "\n",
    "payload = {\n",
    "    \"inputs\":  prompt,\n",
    "    \"parameters\": inference_params,\n",
    "    \"stream\": True ## <-- to have response stream.\n",
    "}\n",
    "\n",
    "resp = get_realtime_response_stream(sagemaker_runtime, endpoint_name, payload)\n",
    "print_response_stream(resp)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52c9ec5c-4435-4d0b-9cde-440ceec1e34f",
   "metadata": {},
   "source": [
    "## Clean Up\n",
    "\n",
    "Finally, let's programatically clean up SageMaker Endpoint, EndpointConfig and Model\n",
    "\n",
    "Change the value of the variable *cleanup* to `True` to perform clean up activity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7a47f51-80da-47cc-802f-1f6ba24f75db",
   "metadata": {},
   "outputs": [],
   "source": [
    "cleanup = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f5841828-99c2-4944-b245-fd8cb806255a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import boto3\n",
    "sm_client = boto3.client('sagemaker')\n",
    "endpoint = sm_client.describe_endpoint(EndpointName=endpoint_name)\n",
    "endpoint_config_name = endpoint['EndpointConfigName']\n",
    "endpoint_config = sm_client.describe_endpoint_config(EndpointConfigName=endpoint_config_name)\n",
    "model_name = endpoint_config['ProductionVariants'][0]['ModelName']\n",
    "\n",
    "print(f\"\"\"\n",
    "About to delete the following sagemaker resources:\n",
    "Endpoint: {endpoint_name}\n",
    "Endpoint Config: {endpoint_config_name}\n",
    "Model: {model_name}\n",
    "\"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ff18a26-5b22-4a65-a63f-5e437baf0ab7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "if cleanup:\n",
    "    # delete endpoint\n",
    "    sm_client.delete_endpoint(EndpointName=endpoint_name)\n",
    "    # delete endpoint config\n",
    "    sm_client.delete_endpoint_config(EndpointConfigName=endpoint_config_name)\n",
    "    # delete model\n",
    "    sm_client.delete_model(ModelName=model_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0119832d-cd96-4f73-924b-b794513e3d4d",
   "metadata": {},
   "source": [
    "## References:\n",
    "1. [Sagemaker Real-time Inference now supports response streaming](https://aws.amazon.com/about-aws/whats-new/2023/09/sagemaker-real-time-inference-response-streaming/)\n",
    "2. [Elevating the generative AI experience: Introducing streaming support in Amazon SageMaker hosting](https://aws.amazon.com/blogs/machine-learning/elevating-the-generative-ai-experience-introducing-streaming-support-in-amazon-sagemaker-hosting/)"
   ]
  }
 ],
 "metadata": {
  "availableInstances": [
   {
    "_defaultOrder": 0,
    "_isFastLaunch": true,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 4,
    "name": "ml.t3.medium",
    "vcpuNum": 2
   },
   {
    "_defaultOrder": 1,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 8,
    "name": "ml.t3.large",
    "vcpuNum": 2
   },
   {
    "_defaultOrder": 2,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.t3.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 3,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.t3.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 4,
    "_isFastLaunch": true,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 8,
    "name": "ml.m5.large",
    "vcpuNum": 2
   },
   {
    "_defaultOrder": 5,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.m5.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 6,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.m5.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 7,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 64,
    "name": "ml.m5.4xlarge",
    "vcpuNum": 16
   },
   {
    "_defaultOrder": 8,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 128,
    "name": "ml.m5.8xlarge",
    "vcpuNum": 32
   },
   {
    "_defaultOrder": 9,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 192,
    "name": "ml.m5.12xlarge",
    "vcpuNum": 48
   },
   {
    "_defaultOrder": 10,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 256,
    "name": "ml.m5.16xlarge",
    "vcpuNum": 64
   },
   {
    "_defaultOrder": 11,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 384,
    "name": "ml.m5.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 12,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 8,
    "name": "ml.m5d.large",
    "vcpuNum": 2
   },
   {
    "_defaultOrder": 13,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.m5d.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 14,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.m5d.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 15,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 64,
    "name": "ml.m5d.4xlarge",
    "vcpuNum": 16
   },
   {
    "_defaultOrder": 16,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 128,
    "name": "ml.m5d.8xlarge",
    "vcpuNum": 32
   },
   {
    "_defaultOrder": 17,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 192,
    "name": "ml.m5d.12xlarge",
    "vcpuNum": 48
   },
   {
    "_defaultOrder": 18,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 256,
    "name": "ml.m5d.16xlarge",
    "vcpuNum": 64
   },
   {
    "_defaultOrder": 19,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 384,
    "name": "ml.m5d.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 20,
    "_isFastLaunch": false,
    "category": "General purpose",
    "gpuNum": 0,
    "hideHardwareSpecs": true,
    "memoryGiB": 0,
    "name": "ml.geospatial.interactive",
    "supportedImageNames": [
     "sagemaker-geospatial-v1-0"
    ],
    "vcpuNum": 0
   },
   {
    "_defaultOrder": 21,
    "_isFastLaunch": true,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 4,
    "name": "ml.c5.large",
    "vcpuNum": 2
   },
   {
    "_defaultOrder": 22,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 8,
    "name": "ml.c5.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 23,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.c5.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 24,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.c5.4xlarge",
    "vcpuNum": 16
   },
   {
    "_defaultOrder": 25,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 72,
    "name": "ml.c5.9xlarge",
    "vcpuNum": 36
   },
   {
    "_defaultOrder": 26,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 96,
    "name": "ml.c5.12xlarge",
    "vcpuNum": 48
   },
   {
    "_defaultOrder": 27,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 144,
    "name": "ml.c5.18xlarge",
    "vcpuNum": 72
   },
   {
    "_defaultOrder": 28,
    "_isFastLaunch": false,
    "category": "Compute optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 192,
    "name": "ml.c5.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 29,
    "_isFastLaunch": true,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.g4dn.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 30,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.g4dn.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 31,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 64,
    "name": "ml.g4dn.4xlarge",
    "vcpuNum": 16
   },
   {
    "_defaultOrder": 32,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 128,
    "name": "ml.g4dn.8xlarge",
    "vcpuNum": 32
   },
   {
    "_defaultOrder": 33,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 4,
    "hideHardwareSpecs": false,
    "memoryGiB": 192,
    "name": "ml.g4dn.12xlarge",
    "vcpuNum": 48
   },
   {
    "_defaultOrder": 34,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 256,
    "name": "ml.g4dn.16xlarge",
    "vcpuNum": 64
   },
   {
    "_defaultOrder": 35,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 61,
    "name": "ml.p3.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 36,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 4,
    "hideHardwareSpecs": false,
    "memoryGiB": 244,
    "name": "ml.p3.8xlarge",
    "vcpuNum": 32
   },
   {
    "_defaultOrder": 37,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 8,
    "hideHardwareSpecs": false,
    "memoryGiB": 488,
    "name": "ml.p3.16xlarge",
    "vcpuNum": 64
   },
   {
    "_defaultOrder": 38,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 8,
    "hideHardwareSpecs": false,
    "memoryGiB": 768,
    "name": "ml.p3dn.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 39,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.r5.large",
    "vcpuNum": 2
   },
   {
    "_defaultOrder": 40,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.r5.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 41,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 64,
    "name": "ml.r5.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 42,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 128,
    "name": "ml.r5.4xlarge",
    "vcpuNum": 16
   },
   {
    "_defaultOrder": 43,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 256,
    "name": "ml.r5.8xlarge",
    "vcpuNum": 32
   },
   {
    "_defaultOrder": 44,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 384,
    "name": "ml.r5.12xlarge",
    "vcpuNum": 48
   },
   {
    "_defaultOrder": 45,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 512,
    "name": "ml.r5.16xlarge",
    "vcpuNum": 64
   },
   {
    "_defaultOrder": 46,
    "_isFastLaunch": false,
    "category": "Memory Optimized",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 768,
    "name": "ml.r5.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 47,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 16,
    "name": "ml.g5.xlarge",
    "vcpuNum": 4
   },
   {
    "_defaultOrder": 48,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.g5.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 49,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 64,
    "name": "ml.g5.4xlarge",
    "vcpuNum": 16
   },
   {
    "_defaultOrder": 50,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 128,
    "name": "ml.g5.8xlarge",
    "vcpuNum": 32
   },
   {
    "_defaultOrder": 51,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 1,
    "hideHardwareSpecs": false,
    "memoryGiB": 256,
    "name": "ml.g5.16xlarge",
    "vcpuNum": 64
   },
   {
    "_defaultOrder": 52,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 4,
    "hideHardwareSpecs": false,
    "memoryGiB": 192,
    "name": "ml.g5.12xlarge",
    "vcpuNum": 48
   },
   {
    "_defaultOrder": 53,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 4,
    "hideHardwareSpecs": false,
    "memoryGiB": 384,
    "name": "ml.g5.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 54,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 8,
    "hideHardwareSpecs": false,
    "memoryGiB": 768,
    "name": "ml.g5.48xlarge",
    "vcpuNum": 192
   },
   {
    "_defaultOrder": 55,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 8,
    "hideHardwareSpecs": false,
    "memoryGiB": 1152,
    "name": "ml.p4d.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 56,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 8,
    "hideHardwareSpecs": false,
    "memoryGiB": 1152,
    "name": "ml.p4de.24xlarge",
    "vcpuNum": 96
   },
   {
    "_defaultOrder": 57,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 32,
    "name": "ml.trn1.2xlarge",
    "vcpuNum": 8
   },
   {
    "_defaultOrder": 58,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 512,
    "name": "ml.trn1.32xlarge",
    "vcpuNum": 128
   },
   {
    "_defaultOrder": 59,
    "_isFastLaunch": false,
    "category": "Accelerated computing",
    "gpuNum": 0,
    "hideHardwareSpecs": false,
    "memoryGiB": 512,
    "name": "ml.trn1n.32xlarge",
    "vcpuNum": 128
   }
  ],
  "instance_type": "ml.t3.medium",
  "kernelspec": {
   "display_name": "Python 3 (Data Science 3.0)",
   "language": "python",
   "name": "python3__SAGEMAKER_INTERNAL__arn:aws:sagemaker:us-west-2:236514542706:image/sagemaker-data-science-310-v1"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}