Hugging Face - Token Classification for Sequence Labelling Tasks using Inference API

Hugging Face/Hugging_Face_Token_Classification_for_Sequence_Labelling_Tasks_using_Inference_API.ipynb

@@ -0,0 +1,267 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "b58bfcc4-c6b6-4938-9a34-86f97c6a80a8",
+   "metadata": {},
+   "source": [
+    "<img width=\"8%\" alt=\"Hugging Face.png\" src=\"https://raw.githubusercontent.com/jupyter-naas/awesome-notebooks/master/.github/assets/logos/Hugging%20Face.png\" style=\"border-radius: 15%\">"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "db404ea5-a162-41ad-a89a-5b7d45856526",
+   "metadata": {},
+   "source": [
+    "# Hugging Face - Token Classification for Sequence Labelling Tasks using Inference API\n",
+    "<a href=\"https://app.naas.ai/user-redirect/naas/downloader?url=https://raw.githubusercontent.com/jupyter-naas/awesome-notebooks/master/Hugging%20Face/Hugging_Face_Token_Classification_for_Sequence_Labelling_Tasks_using_Inference_API.ipynb\" target=\"_parent\"><img src=\"https://naasai-public.s3.eu-west-3.amazonaws.com/Open_in_Naas_Lab.svg\"/></a><br><br><a href=\"https://bit.ly/3JyWIk6\">Give Feedback</a> | <a href=\"https://github.com/jupyter-naas/awesome-notebooks/issues/new?assignees=&labels=bug&template=bug_report.md&title=Hugging+Face+-+Token+Classification+for+Sequence+Labelling+Tasks+using+Inference+API:+Error+short+description\">Bug report</a>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1e8ef932-a062-4abb-b762-317e32648054",
+   "metadata": {},
+   "source": [
+    "**Tags:** #huggingface #ml #token_classification #prompt #inference_api #ai #text"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a99ea7c5-5a04-4340-b312-388fbe368ffd",
+   "metadata": {},
+   "source": [
+    "**Author:** [Kevin Joseph Scaria](https://www.linkedin.com/in/kevin-scaria/)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3d629165-0640-46ac-aa14-2ce67eea8718",
+   "metadata": {},
+   "source": [
+    "**Last update:** 2023-11-17 (Created: 2023-11-17)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2c5f0e91-b512-40a3-b417-9e28466ae410",
+   "metadata": {},
+   "source": [
+    "**Description:** This notebook demonstrates how to utilize the <a href=\"https://huggingface.co/docs/inference-endpoints/index\">inference endpoints</a> (additional information can be found here: <a href=\"https://huggingface.co/docs/api-inference/index#-hosted-inference-api\">link</a>) of hugging face models for token classification based sequence labelling tasks."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d65c3967-2a28-4dc3-9158-fbfc4684dbaa",
+   "metadata": {},
+   "source": [
+    "## Input"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ba7aa84f-f166-4b45-98c3-8e5fc9bbe144",
+   "metadata": {},
+   "source": [
+    "### Installations and Imports"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "80b10f53-fec0-472c-98b5-56b8722e5a80",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2023-11-18T04:25:00.763574Z",
+     "iopub.status.busy": "2023-11-18T04:25:00.763302Z",
+     "iopub.status.idle": "2023-11-18T04:25:10.601595Z",
+     "shell.execute_reply": "2023-11-18T04:25:10.600857Z",
+     "shell.execute_reply.started": "2023-11-18T04:25:00.763506Z"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "def import_and_install(package, name=None):\n",
+    "    try:\n",
+    "        if name is None:\n",
+    "            globals()[package] = __import__(package)\n",
+    "        else:\n",
+    "            globals()[name] = __import__(package)\n",
+    "    except ImportError:\n",
+    "        !pip install -q {package}\n",
+    "        globals()[package] = __import__(package)\n",
+    "\n",
+    "# Usage\n",
+    "import_and_install('datasets')\n",
+    "import_and_install('numpy', 'np')\n",
+    "import_and_install('requests')\n",
+    "import_and_install('json')\n",
+    "import_and_install('naas')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "28bcbd55-f82c-4da3-8ee9-bee13686a20f",
+   "metadata": {},
+   "source": [
+    "### Inference API\n",
+    "\n",
+    "To run inference on huggingface models using their API, we need to extract the API token.\n",
+    "\n",
+    "#### Steps to get API token\n",
+    "- Create an account on [Hugging Face](https://huggingface.co)\n",
+    "- Log in, and click on profile icon (top right corner)\n",
+    "- Go to settings\n",
+    "- Click on [Access tokens](https://huggingface.co/settings/tokens)\n",
+    "- Now, create a new access token with name: `INFERENCE_API` and role: `read`\n",
+    "- Copy the generated token and paste it below\n",
+    "\n",
+    "\n",
+    "We will use a pretrained model specifically finetuned for token classification tasks on large corpora of data.\n",
+    "\n",
+    "In this demonstration, we will utilize the popular <a href=\"https://huggingface.co/dslim/bert-base-NER\">BERT-base model finetuned for NER tasks</a>."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "82bce589-5972-4ce1-8080-be27f96314ef",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2023-11-18T04:57:11.766689Z",
+     "iopub.status.busy": "2023-11-18T04:57:11.766440Z",
+     "iopub.status.idle": "2023-11-18T04:57:12.148083Z",
+     "shell.execute_reply": "2023-11-18T04:57:12.147446Z",
+     "shell.execute_reply.started": "2023-11-18T04:57:11.766664Z"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "👌 Well done! Your Secret has been sent to production. \n",
+      "\n",
+      "PS: to remove the \"Secret\" feature, just replace .add by .delete\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Settting API token as an environment variable using naas module\n",
+    "\n",
+    "INFERENCE_API = \"ADD_YOUR_API_TOKEN_HERE\"\n",
+    "naas.secret.add(\"INFERENCE_API\", INFERENCE_API)\n",
+    "\n",
+    "API_TOKEN = naas.secret.get(\"INFERENCE_API\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8ef6851a-95f9-4da8-855f-b27ec4a30f5d",
+   "metadata": {},
+   "source": [
+    "## Model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "id": "28a2db1e-bfdd-4b2a-aec5-3459e3e32ad2",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2023-11-18T05:16:58.600520Z",
+     "iopub.status.busy": "2023-11-18T05:16:58.600261Z",
+     "iopub.status.idle": "2023-11-18T05:16:58.610749Z",
+     "shell.execute_reply": "2023-11-18T05:16:58.609871Z",
+     "shell.execute_reply.started": "2023-11-18T05:16:58.600483Z"
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "def get_api_reponse(payload, model, API_KEY):\n",
+    "    API_URL = f\"https://api-inference.huggingface.co/models/{model}\"\n",
+    "    headers = {\"Authorization\": f\"Bearer {API_KEY}\"}\n",
+    "    try:\n",
+    "        response = requests.post(API_URL, headers=headers, json=payload)\n",
+    "        if response.status_code == 200:\n",
+    "            return response.json()\n",
+    "        else:\n",
+    "            print(f\"Error: {response.status_code}, {response.text}\")\n",
+    "    except Exception as e:\n",
+    "        return \"Error: \" + e"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8c3ceb80-fbf9-4de1-97d6-303e04394414",
+   "metadata": {},
+   "source": [
+    "## Output"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1a1e7df8-a8d7-4003-a8ae-bd09c33f90d6",
+   "metadata": {},
+   "source": [
+    "Find the appropriate model id from huggingface [models](https://huggingface.co/models) and replace the model in the `MODEL` variable"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "id": "2fb093bc-0e9a-495e-9ee0-787bc43533cb",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2023-11-18T05:17:26.350844Z",
+     "iopub.status.busy": "2023-11-18T05:17:26.350504Z",
+     "iopub.status.idle": "2023-11-18T05:17:26.691915Z",
+     "shell.execute_reply": "2023-11-18T05:17:26.691178Z",
+     "shell.execute_reply.started": "2023-11-18T05:17:26.350805Z"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[{'entity_group': 'PER', 'score': 0.9979527592658997, 'word': 'Sarah Jessica Parker', 'start': 11, 'end': 31}, {'entity_group': 'ORG', 'score': 0.9990966320037842, 'word': 'Apple Computers', 'start': 46, 'end': 61}]\n"
+     ]
+    }
+   ],
+   "source": [
+    "input_ = {\"inputs\": \"My name is Sarah Jessica Parker and I work at Apple Computers.\", \n",
+    "          \"options\":{\"wait_for_model\":True,\n",
+    "                    \"use_cache\":False}\n",
+    "         }\n",
+    "MODEL = \"dslim/bert-base-NER\"\n",
+    "\n",
+    "response = get_api_reponse(payload=input_, model=MODEL, API_KEY=API_TOKEN)\n",
+    "print(response)"
+   ]
+  }
+ ],
+ "metadata": {
+  "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": "3.9.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}