sagemaker-notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Development Environment and Permissions\n",
    "\n",
    "### Installation \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "69c59d90",
   "metadata": {},
   "outputs": [],
   "source": [
    "%pip install sagemaker --upgrade\n",
    "import sagemaker\n",
    "\n",
    "assert sagemaker.__version__ >= \"2.75.0\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce0ef431",
   "metadata": {},
   "source": [
    "Install `git` and `git-lfs`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "96d8dfea",
   "metadata": {},
   "outputs": [],
   "source": [
    "# For notebook instances (Amazon Linux)\n",
    "!sudo yum update -y \n",
    "!curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.rpm.sh | sudo bash\n",
    "!sudo yum install git-lfs git -y\n",
    "# For other environments (Ubuntu)\n",
    "!sudo apt-get update -y \n",
    "!curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash\n",
    "!sudo apt-get install git-lfs git -y"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e4386d9",
   "metadata": {},
   "source": [
    "### Permissions\n",
    "\n",
    "_If you are going to use Sagemaker in a local environment (not SageMaker Studio or Notebook Instances). You need access to an IAM Role with the required permissions for Sagemaker. You can find [here](https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-roles.html) more about it._"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1c22e8d5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Couldn't call 'get_role' to get Role ARN from role name philippschmid to get Role path.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sagemaker role arn: arn:aws:iam::558105141721:role/sagemaker_execution_role\n",
      "sagemaker bucket: sagemaker-us-east-1-558105141721\n",
      "sagemaker session region: us-east-1\n"
     ]
    }
   ],
   "source": [
    "import sagemaker\n",
    "import boto3\n",
    "sess = sagemaker.Session()\n",
    "# sagemaker session bucket -> used for uploading data, models and logs\n",
    "# sagemaker will automatically create this bucket if it not exists\n",
    "sagemaker_session_bucket=None\n",
    "if sagemaker_session_bucket is None and sess is not None:\n",
    "    # set to default bucket if a bucket name is not given\n",
    "    sagemaker_session_bucket = sess.default_bucket()\n",
    "\n",
    "try:\n",
    "    role = sagemaker.get_execution_role()\n",
    "except ValueError:\n",
    "    iam = boto3.client('iam')\n",
    "    role = iam.get_role(RoleName='sagemaker_execution_role')['Role']['Arn']\n",
    "\n",
    "sess = sagemaker.Session(default_bucket=sagemaker_session_bucket)\n",
    "\n",
    "print(f\"sagemaker role arn: {role}\")\n",
    "print(f\"sagemaker bucket: {sess.default_bucket()}\")\n",
    "print(f\"sagemaker session region: {sess.boto_region_name}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "952983b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "repository = \"jonatasgrosman/wav2vec2-large-xlsr-53-english\"\n",
    "model_id=repository.split(\"/\")[-1]\n",
    "s3_location=f\"s3://{sess.default_bucket()}/custom_inference/{model_id}/model.tar.gz\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b8452981",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Updated git hooks.\n",
      "Git LFS initialized.\n",
      "Cloning into 'wav2vec2-large-xlsr-53-english'...\n",
      "remote: Enumerating objects: 93, done.\u001b[K\n",
      "remote: Counting objects: 100% (93/93), done.\u001b[K\n",
      "remote: Compressing objects: 100% (88/88), done.\u001b[K\n",
      "remote: Total 93 (delta 44), reused 0 (delta 0), pack-reused 0\u001b[K\n",
      "Unpacking objects: 100% (93/93), 745.20 KiB | 1.57 MiB/s, done.\n",
      "Filtering content: 100% (2/2), 2.35 GiB | 19.68 MiB/s, done.\n"
     ]
    }
   ],
   "source": [
    "!git lfs install\n",
    "!git clone https://huggingface.co/$repository"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6146af09",
   "metadata": {},
   "outputs": [],
   "source": [
    "!cp -r code/ $model_id/code/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e65fd56e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Errno 2] No such file or directory: 'wav2vec2-large-xlsr-53-english'\n",
      "/Users/philipp/Projects/huggingface/notebooks/sagemaker/17_custom_inference_script/wav2vec2-large-xlsr-53-english\n",
      "a README.md\n",
      "a code\n",
      "a code/inference.py\n",
      "a config.json\n",
      "a preprocessor_config.json\n",
      "a pytorch_model.bin\n",
      "a special_tokens_map.json\n",
      "a vocab.json\n"
     ]
    }
   ],
   "source": [
    "%cd $model_id\n",
    "!tar zcvf model.tar.gz *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c581bc40",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "upload: ./model.tar.gz to s3://sagemaker-us-east-1-558105141721/custom_inference/wav2vec2-large-xlsr-53-english/model.tar.gz\n"
     ]
    }
   ],
   "source": [
    "!aws s3 cp model.tar.gz $s3_location\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fc75b436",
   "metadata": {},
   "outputs": [],
   "source": [
    "!mkdir code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3965a632",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%writefile code/inference.py\n",
    "\n",
    "from transformers import pipeline\n",
    "\n",
    "\n",
    "def model_fn(model_dir):\n",
    "    pipe = pipeline(\"automatic-speech-recognition\", model=model_dir, chunk_length_s=10)\n",
    "    return pipe\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----!"
     ]
    }
   ],
   "source": [
    "from sagemaker.huggingface.model import HuggingFaceModel\n",
    "from sagemaker.serializers import DataSerializer\n",
    "from sagemaker.serverless import ServerlessInferenceConfig\n",
    "\n",
    "# Hub Model configuration. <https://huggingface.co/models>\n",
    "hub = {\n",
    "    'HF_MODEL_ID':'distilbert-base-uncased-finetuned-sst-2-english',\n",
    "    'HF_TASK':'text-classification'\n",
    "}\n",
    "\n",
    "# create Hugging Face Model Class\n",
    "huggingface_model = HuggingFaceModel(\n",
    "   model_data=s3_location,  # path to your trained sagemaker model\n",
    "   role=role,                    # iam role with permissions to create an Endpoint\n",
    "   transformers_version=\"4.17\",  # transformers version used\n",
    "   pytorch_version=\"1.10\",        # pytorch version used\n",
    "   py_version='py38',            # python version used\n",
    ")\n",
    "\n",
    "# Specify MemorySizeInMB and MaxConcurrency in the serverless config object\n",
    "serverless_config = ServerlessInferenceConfig(\n",
    "    memory_size_in_mb=4096, max_concurrency=10,\n",
    ")\n",
    "\n",
    "# create a serializer for the data\n",
    "audio_serializer = DataSerializer(content_type='audio/x-audio') # using x-audio to support multiple audio formats\n",
    "\n",
    "# deploy the endpoint endpoint\n",
    "predictor = huggingface_model.deploy(\n",
    "serverless_inference_config=serverless_config,\n",
    "  serializer=audio_serializer, # serializer for our audio data.\n",
    "\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6b3812f",
   "metadata": {},
   "source": [
    "## Request Inference Endpoint using the `HuggingfacePredictor`\n",
    "\n",
    "The `.deploy()` returns an `HuggingFacePredictor` object which can be used to request inference."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b72ff640",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2022-05-20 19:06:10--  https://cdn-media.huggingface.co/speech_samples/sample1.flac\n",
      "Auflösen des Hostnamens cdn-media.huggingface.co (cdn-media.huggingface.co)… 13.227.153.89, 13.227.153.26, 13.227.153.12, ...\n",
      "Verbindungsaufbau zu cdn-media.huggingface.co (cdn-media.huggingface.co)|13.227.153.89|:443 … verbunden.\n",
      "HTTP-Anforderung gesendet, auf Antwort wird gewartet … 200 OK\n",
      "Länge: 282378 (276K) [audio/flac]\n",
      "Wird in »sample1.flac« gespeichert.\n",
      "\n",
      "sample1.flac        100%[===================>] 275,76K  --.-KB/s    in 0,05s   \n",
      "\n",
      "2022-05-20 19:06:10 (5,60 MB/s) - »sample1.flac« gespeichert [282378/282378]\n",
      "\n",
      "{'text': \"going along slashy country roads and speaking to damp audiences in drafty school rooms day-after-day for a fortnighthe'll have to put in an appearance at some place of worship on sunday morning and he can come\"}\n"
     ]
    }
   ],
   "source": [
    "!wget https://cdn-media.huggingface.co/speech_samples/sample1.flac\n",
    "audio_path = \"sample1.flac\"\n",
    "\n",
    "res = predictor.predict(data=audio_path)\n",
    "print(res)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb10007d",
   "metadata": {},
   "source": [
    "### Delete model and endpoint\n",
    "\n",
    "To clean up, we can delete the model and endpoint."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "1e6fb7b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictor.delete_model()\n",
    "predictor.delete_endpoint()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bb862af7",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "c281c456f1b8161c8906f4af2c08ed2c40c50136979eaae69688b01f70e9f4a9"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}