-
Notifications
You must be signed in to change notification settings - Fork 4.7k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Add entity metadata extractor (#7163)
- Loading branch information
1 parent
0b91476
commit d6cf818
Showing
7 changed files
with
461 additions
and
1 deletion.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
343 changes: 343 additions & 0 deletions
343
docs/examples/metadata_extraction/EntityExtractionClimate.ipynb
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,343 @@ | ||
| { | ||
| "cells": [ | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "# Entity Metadata Extraction\n", | ||
| "\n", | ||
| "In this demo, we use the new `EntityExtractor` to extract entities from each node, stored in metadata. The default model is `tomaarsen/span-marker-xlm-roberta-base-multinerd`, which is downloaded an run locally from [HuggingFace](https://huggingface.co/tomaarsen/span-marker-xlm-roberta-base-multinerd).\n", | ||
| "\n", | ||
| "For more information on metadata extraction in LlamaIndex, see our [documentation](https://gpt-index.readthedocs.io/en/stable/core_modules/data_modules/documents_and_nodes/usage_metadata_extractor.html)." | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 1, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# Needed to run the entity extractor\n", | ||
| "# !pip install span_marker\n", | ||
| "\n", | ||
| "import os\n", | ||
| "import openai\n", | ||
| "\n", | ||
| "os.environ[\"OPENAI_API_KEY\"] = \"YOUR_API_KEY\"\n", | ||
| "openai.api_key = os.getenv(\"OPENAI_API_KEY\")" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "## Setup the Extractor and Parser" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 2, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stderr", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "/Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", | ||
| " from .autonotebook import tqdm as notebook_tqdm\n", | ||
| "/Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages/bitsandbytes/cextension.py:34: UserWarning: The installed version of bitsandbytes was compiled without GPU support. 8-bit optimizers, 8-bit multiplication, and GPU quantization are unavailable.\n", | ||
| " warn(\"The installed version of bitsandbytes was compiled without GPU support. \"\n" | ||
| ] | ||
| }, | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "'NoneType' object has no attribute 'cadam32bit_grad_fp32'\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "from llama_index.node_parser.extractors.metadata_extractors import (\n", | ||
| " MetadataExtractor,\n", | ||
| " EntityExtractor,\n", | ||
| ")\n", | ||
| "from llama_index.node_parser.simple import SimpleNodeParser\n", | ||
| "\n", | ||
| "entity_extractor = EntityExtractor(\n", | ||
| " prediction_threshold=0.5,\n", | ||
| " label_entities=False, # include the entity label in the metadata (can be erroneous)\n", | ||
| " device=\"cpu\", # set to \"cuda\" if you have a GPU\n", | ||
| ")\n", | ||
| "\n", | ||
| "metadata_extractor = MetadataExtractor(extractors=[entity_extractor])\n", | ||
| "\n", | ||
| "node_parser = SimpleNodeParser.from_defaults(metadata_extractor=metadata_extractor)" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "## Load the data\n", | ||
| "\n", | ||
| "Here, we will download the 2023 IPPC Climate Report - Chapter 3 on Oceans and Coastal Ecosystems (172 Pages)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 3, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| " % Total % Received % Xferd Average Speed Time Time Time Current\n", | ||
| " Dload Upload Total Spent Left Speed\n", | ||
| "100 20.7M 100 20.7M 0 0 22.1M 0 --:--:-- --:--:-- --:--:-- 22.1M\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "!curl https://www.ipcc.ch/report/ar6/wg2/downloads/report/IPCC_AR6_WGII_Chapter03.pdf --output IPCC_AR6_WGII_Chapter03.pdf" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "Next, load the documents." | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 4, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "from llama_index import SimpleDirectoryReader\n", | ||
| "\n", | ||
| "documents = SimpleDirectoryReader(\n", | ||
| " input_files=[\"./IPCC_AR6_WGII_Chapter03.pdf\"]\n", | ||
| ").load_data()" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "## Extracting Metadata\n", | ||
| "\n", | ||
| "Now, this is a pretty long document. Since we are not running on CPU, for now, we will only run on a subset of documents. Feel free to run it on all documents on your own though!" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 5, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n", | ||
| "To disable this warning, you can either:\n", | ||
| "\t- Avoid using `tokenizers` before the fork if possible\n", | ||
| "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "import random\n", | ||
| "\n", | ||
| "random.seed(42)\n", | ||
| "# comment out to run on all documents\n", | ||
| "# 100 documents takes about 5 minutes on CPU\n", | ||
| "documents = random.sample(documents, 100)\n", | ||
| "\n", | ||
| "nodes = node_parser.get_nodes_from_documents(documents)" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "### Examine the outputs" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 6, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "{'page_label': '387', 'file_name': 'IPCC_AR6_WGII_Chapter03.pdf'}\n", | ||
| "{'page_label': '410', 'file_name': 'IPCC_AR6_WGII_Chapter03.pdf', 'entities': {'Parmesan', 'Boyd', 'Riebesell', 'Gattuso'}}\n", | ||
| "{'page_label': '391', 'file_name': 'IPCC_AR6_WGII_Chapter03.pdf', 'entities': {'Gulev', 'Fox-Kemper'}}\n", | ||
| "{'page_label': '430', 'file_name': 'IPCC_AR6_WGII_Chapter03.pdf', 'entities': {'Kessouri', 'van der Sleen', 'Brodeur', 'Siedlecki', 'Fiechter', 'Ramajo', 'Carozza'}}\n", | ||
| "{'page_label': '388', 'file_name': 'IPCC_AR6_WGII_Chapter03.pdf'}\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "samples = random.sample(nodes, 5)\n", | ||
| "for node in samples:\n", | ||
| " print(node.metadata)" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "## Try a Query!" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 8, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "from llama_index import ServiceContext, VectorStoreIndex\n", | ||
| "from llama_index.llms import OpenAI\n", | ||
| "\n", | ||
| "service_context = ServiceContext.from_defaults(\n", | ||
| " llm=OpenAI(model=\"gpt-3.5-turbo\", temperature=0.2)\n", | ||
| ")\n", | ||
| "\n", | ||
| "index = VectorStoreIndex(nodes, service_context=service_context)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 9, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "According to the provided context information, Fox-Kemper is mentioned in relation to the observed and projected trends of ocean warming and marine heatwaves. It is stated that Fox-Kemper et al. (2021) reported that ocean warming has increased on average by 0.88°C from 1850-1900 to 2011-2020. Additionally, it is mentioned that Fox-Kemper et al. (2021) projected that ocean warming will continue throughout the 21st century, with the rate of global ocean warming becoming scenario-dependent from the mid-21st century. Fox-Kemper is also cited as a source for the information on the increasing frequency, intensity, and duration of marine heatwaves over the 20th and early 21st centuries, as well as the projected increase in frequency of marine heatwaves in the future.\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "query_engine = index.as_query_engine()\n", | ||
| "response = query_engine.query(\"What is said by Fox-Kemper?\")\n", | ||
| "print(response)" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "### Contrast without metadata\n", | ||
| "\n", | ||
| "Here, we re-construct the index, but without metadata" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 10, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "{'page_label': '542', 'file_name': 'IPCC_AR6_WGII_Chapter03.pdf'}\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "for node in nodes:\n", | ||
| " node.metadata.pop(\"entities\", None)\n", | ||
| "\n", | ||
| "print(nodes[0].metadata)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 11, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "from llama_index import ServiceContext, VectorStoreIndex\n", | ||
| "from llama_index.llms import OpenAI\n", | ||
| "\n", | ||
| "service_context = ServiceContext.from_defaults(\n", | ||
| " llm=OpenAI(model=\"gpt-3.5-turbo\", temperature=0.2)\n", | ||
| ")\n", | ||
| "\n", | ||
| "index = VectorStoreIndex(nodes, service_context=service_context)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 12, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "According to the provided context information, Fox-Kemper is mentioned in relation to the decline of the AMOC (Atlantic Meridional Overturning Circulation) over the 21st century. The statement mentions that there is high confidence in the decline of the AMOC, but low confidence for quantitative projections.\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "query_engine = index.as_query_engine()\n", | ||
| "response = query_engine.query(\"What is said by Fox-Kemper?\")\n", | ||
| "print(response)" | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "As we can see, our metadata-enriched index is able to fetch more relevant information." | ||
| ] | ||
| }, | ||
| { | ||
| "attachments": {}, | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [] | ||
| } | ||
| ], | ||
| "metadata": { | ||
| "kernelspec": { | ||
| "display_name": "llama-index", | ||
| "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" | ||
| }, | ||
| "orig_nbformat": 4 | ||
| }, | ||
| "nbformat": 4, | ||
| "nbformat_minor": 2 | ||
| } |
Oops, something went wrong.