Add gemini-ultra

Cleaned up `makefile`

.github/workflows/ci.yml

@@ -66,7 +66,7 @@ jobs:
         shell: bash
         run: |
           echo "Running tests, installing dependencies with poetry..."
-          poetry install --with test,lint,typing
+          poetry install --with test,lint,typing,docs
 
       - name: Run tests
         run: make test
@@ -82,3 +82,69 @@ jobs:
           # grep will exit non-zero if the target message isn't found,
           # and `set -e` above will cause the step to fail.
           echo "$STATUS" | grep 'nothing to commit, working tree clean'
+  test_docs:
+    timeout-minutes: 5
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        working-directory: ${{ env.WORKDIR }}
+    strategy:
+      matrix:
+        python-version:
+          - "3.11"
+    name: Documentation Build for Python ${{ matrix.python-version }}
+    steps:
+      - uses: actions/checkout@v3
+
+      - name: Set up Python ${{ matrix.python-version }} + Poetry ${{ env.POETRY_VERSION }}
+        uses: "./.github/actions/poetry_setup"
+        with:
+          python-version: ${{ matrix.python-version }}
+          poetry-version: ${{ env.POETRY_VERSION }}
+          working-directory: .
+          cache-key: benchmarks-all
+
+      - name: Install dependencies
+        shell: bash
+        run: |
+          echo "Running tests, installing dependencies with poetry..."
+          poetry install --with test,lint,typing,docs
+
+      - name: Test Sphinx Docs
+        shell: bash
+        run: |
+          echo "Attempting to build docs..."
+          make docs_build
+  test_datasets:
+    timeout-minutes: 5
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        working-directory: ${{ env.WORKDIR }}
+    strategy:
+      matrix:
+        python-version:
+          - "3.11"
+    name: Validate Public Datasets
+    steps:
+      - uses: actions/checkout@v3
+
+      - name: Set up Python ${{ matrix.python-version }} + Poetry ${{ env.POETRY_VERSION }}
+        uses: "./.github/actions/poetry_setup"
+        with:
+          python-version: ${{ matrix.python-version }}
+          poetry-version: ${{ env.POETRY_VERSION }}
+          working-directory: .
+          cache-key: benchmarks-all
+
+      - name: Install dependencies
+        shell: bash
+        run: |
+          echo "Running tests, installing dependencies with poetry..."
+          poetry install --with test,lint,typing,docs
+
+      - name: Request datasets
+        shell: bash
+        run: |
+          echo "Attempting to build docs..."
+          poetry run python -m scripts.check_datasets

.github/workflows/doc_publish.yaml

@@ -0,0 +1,44 @@
+name: Publish Docs
+on: [workflow_dispatch]
+permissions:
+    contents: write
+
+env:
+  POETRY_VERSION: "1.6.1"
+
+jobs:
+  docs:
+    strategy:
+      matrix:
+        python-version:
+          - "3.11"
+    runs-on: ubuntu-latest
+    name: Documentation Publish
+    steps:
+      - uses: actions/checkout@v3
+
+      - name: Set up Python ${{ matrix.python-version }} + Poetry ${{ env.POETRY_VERSION }}
+        uses: "./.github/actions/poetry_setup"
+        with:
+          python-version: ${{ matrix.python-version }}
+          poetry-version: ${{ env.POETRY_VERSION }}
+          working-directory: .
+          cache-key: benchmarks-all
+
+      - name: Install dependencies
+        shell: bash
+        run: |
+          echo "Running tests, installing dependencies with poetry..."
+          poetry install --with test,lint,typing,docs
+
+      - name: Sphinx build
+        shell: bash
+        run: |
+          make docs_build
+      - name: Publish Docs
+        uses: peaceiris/actions-gh-pages@v3
+        with:
+          publish_branch: gh-pages
+          github_token: ${{ secrets.GITHUB_TOKEN }}
+          publish_dir: ./docs/build
+          force_orphan: true

.gitignore

@@ -159,4 +159,4 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
-
+.DS_Store

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Langchain AI
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

Makefile

@@ -3,22 +3,7 @@
 # Default target executed when no arguments are given to make.
 all: help
 
-######################
-# TESTING AND COVERAGE
-######################
-
-# Define a variable for the test file path.
-TEST_FILE ?= tests/unit_tests/
-
-test:
-	poetry run pytest --disable-socket --allow-unix-socket $(TEST_FILE)
-
-test_watch:
-	poetry run ptw . -- $(TEST_FILE)
-
-######################
-# LINTING AND FORMATTING
-######################
+# LINTING AND FORMATTING:
 
 # Define a variable for Python and notebook files.
 lint format: PYTHON_FILES=.
@@ -38,19 +23,45 @@ spell_check:
 spell_fix:
 	poetry run codespell --toml pyproject.toml -w
 
-######################
-# HELP
-######################
 
+# TESTING AND COVERAGE:
+
+# Define a variable for the test file path.
+TEST_FILE ?= tests/unit_tests/
+
+test:
+	poetry run pytest --disable-socket --allow-unix-socket $(TEST_FILE)
+
+test_watch:
+	poetry run ptw . -- $(TEST_FILE)
+
+
+# DOCUMENTATION:
+
+docs_clean:
+	rm -rf ./docs/build
+
+docs_build:
+	# Copy README.md to docs/index.md
+	cp README.md ./docs/source/index.md
+	# Append to the table of contents the contents of the file
+	cat ./docs/source/toc.segment >> ./docs/source/index.md
+	poetry run sphinx-build "./docs/source" "./docs/build"
+
+
+# HELP:
 help:
-	@echo '===================='
-	@echo '-- LINTING --'
-	@echo 'format                       - run code formatters'
-	@echo 'lint                         - run linters'
-	@echo 'spell_check                 	- run codespell on the project'
-	@echo 'spell_fix                		- run codespell on the project and fix the errors'
-	@echo '-- TESTS --'
-	@echo 'coverage                     - run unit tests and generate coverage report'
-	@echo 'test                         - run unit tests'
-	@echo 'test TEST_FILE=<test_file>   - run all tests in file'
-	@echo '-- DOCUMENTATION tasks are from the top-level Makefile --'
+	@echo ''
+	@echo 'LINTING:'
+	@echo '  format             - run code formatters'
+	@echo '  lint               - run linters'
+	@echo '  spell_check        - run codespell'
+	@echo '  spell_fix          - run codespell and fix the errors'
+	@echo 'TESTS:'
+	@echo '  test               - run unit tests'
+	@echo '  test TEST_FILE=<test_file>   - run tests in <test_file>'
+	@echo '  coverage           - run unit tests and generate coverage report'
+	@echo 'DOCUMENTATION:'
+	@echo '  docs_clean         - delete the docs/build directory'
+	@echo '  docs_build         - build the documentation'
+	@echo ''

README.md

@@ -1,8 +1,19 @@
-# 🦜💪 LangChain Benchmarks
+# 🦜💯 LangChain Benchmarks
 
-This repository shows how we benchmark some of our more popular chains and agents.
-The benchmarks are organized by end-to-end use cases.
-They utilize [LangSmith](https://smith.langchain.com/) heavily.
+[![Release Notes](https://img.shields.io/github/release/langchain-ai/langchain-benchmarks)](https://github.com/langchain-ai/langchain-benchmarks/releases)
+[![CI](https://github.com/langchain-ai/langchain-benchmarks/actions/workflows/ci.yml/badge.svg)](https://github.com/langchain-ai/langchain-benchmarks/actions/workflows/ci.yml)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![Twitter](https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI)](https://twitter.com/langchainai)
+[![](https://dcbadge.vercel.app/api/server/6adMQxSpJS?compact=true&style=flat)](https://discord.gg/6adMQxSpJS)
+[![Open Issues](https://img.shields.io/github/issues-raw/langchain-ai/langchain-benchmarks)](https://github.com/langchain-ai/langchain-benchmarks/issues)
+
+
+[📖 Documentation](https://langchain-ai.github.io/langchain-benchmarks/index.html)
+
+A package to help benchmark various LLM related tasks.
+
+The benchmarks are organized by end-to-end use cases, and
+utilize [LangSmith](https://smith.langchain.com/) heavily.
 
 We have several goals in open sourcing this:
 
@@ -11,8 +22,39 @@ We have several goals in open sourcing this:
 - Showing how we evaluate each task
 - Encouraging others to benchmark their solutions on these tasks (we are always looking for better ways of doing things!)
 
-We currently include the following tasks:
-- [CSV Question Answering](csv-qa)
-- [Extraction](extraction)
-- [Q&A over the LangChain docs](langchain-docs-benchmarking)
-- [Meta-evaluation of 'correctness' evaluators](meta-evals)
+## Installation
+
+To install the packages, run the following command:
+
+```bash
+pip install -U langchain-benchmarks
+```
+
+All the benchmarks come with an associated benchmark dataset stored in [LangSmith](https://smith.langchain.com). To take advantage of the eval and debugging experience, [sign up](https://smith.langchain.com), and set your API key in your environment:
+
+```bash
+export LANGCHAIN_API_KEY=ls-...
+```
+
+## Repo Structure
+
+The package is located within [langchain_benchmarks](./langchain_benchmarks/). Check out the [docs](https://langchain-ai.github.io/langchain-benchmarks/index.html) for information on how to get starte.
+
+The other directories are legacy and may be moved in the future.
+
+
+## Archived
+
+Below are archived benchmarks that require cloning this repo to run.
+
+- [CSV Question Answering](https://github.com/langchain-ai/langchain-benchmarks/tree/main/csv-qa)
+- [Extraction](https://github.com/langchain-ai/langchain-benchmarks/tree/main/extraction)
+- [Q&A over the LangChain docs](https://github.com/langchain-ai/langchain-benchmarks/tree/main/langchain-docs-benchmarking)
+- [Meta-evaluation of 'correctness' evaluators](https://github.com/langchain-ai/langchain-benchmarks/tree/main/meta-evals)
+
+
+## Related
+
+- For cookbooks on other ways to test, debug, monitor, and improve your LLM applications, check out the [LangSmith docs](https://docs.smith.langchain.com/)
+- For information on building with LangChain, check out the [python documentation](https://python.langchain.com/docs/get_started/introduction) or [JS documentation](https://js.langchain.com/docs/get_started/introduction)
+

archived/csv-qa/pandas_agent_instruct.py

@@ -5,7 +5,6 @@ from langchain.agents.agent_toolkits.conversational_retrieval.tool import (
 )
 from langchain.embeddings import OpenAIEmbeddings
 from langchain.llms import OpenAI
-from langchain.prompts import PromptTemplate
 from langchain.smith import RunEvalConfig, run_on_dataset
 from langchain.tools import PythonAstREPLTool
 from langchain.vectorstores import FAISS

archived/csv-qa/pandas_ai.py

@@ -1,6 +1,4 @@
 import pandas as pd
-from langchain.agents.agent_toolkits import create_pandas_dataframe_agent
-from langchain.agents.agent_types import AgentType
 from langchain.chat_models import ChatOpenAI
 from langchain.prompts import ChatPromptTemplate
 from langchain.schema.output_parser import StrOutputParser

archived/langchain-docs-benchmarking/app/server.py

@@ -1,8 +1,7 @@
 from chat_langchain.chain import chain
 from fastapi import FastAPI
-from openai_functions_agent import agent_executor as openai_functions_agent_chain
-
 from langserve import add_routes
+from openai_functions_agent import agent_executor as openai_functions_agent_chain
 
 app = FastAPI()
 

archived/langchain-docs-benchmarking/packages/chat-langchain/chat_langchain/chain.py

@@ -134,7 +134,7 @@ def create_response_chain(
         ]
     )
 
-    response_synthesizer = (prompt | llm | StrOutputParser()).with_config(
+    response_generator = (prompt | llm | StrOutputParser()).with_config(
         run_name="GenerateResponse",
     )
     return (
@@ -147,7 +147,7 @@ def create_response_chain(
             ),
         }
         | _context
-        | response_synthesizer
+        | response_generator
     )
 
 

archived/langchain-docs-benchmarking/prepare_dataset.py

@@ -0,0 +1,57 @@
+"""Copy the public dataset to your own langsmith tenant."""
+from typing import Optional
+
+from langsmith import Client
+
+DATASET_NAME = "LangChain Docs Q&A"
+PUBLIC_DATASET_TOKEN = "452ccafc-18e1-4314-885b-edd735f17b9d"
+
+
+def create_langchain_docs_dataset(
+    dataset_name: str = DATASET_NAME,
+    public_dataset_token: str = PUBLIC_DATASET_TOKEN,
+    client: Optional[Client] = None,
+):
+    shared_client = Client(
+        api_url="https://api.smith.langchain.com", api_key="placeholder"
+    )
+    examples = list(shared_client.list_shared_examples(public_dataset_token))
+    client = client or Client()
+    if client.has_dataset(dataset_name=dataset_name):
+        loaded_examples = list(client.list_examples(dataset_name=dataset_name))
+        if len(loaded_examples) == len(examples):
+            return
+        else:
+            ds = client.read_dataset(dataset_name=dataset_name)
+    else:
+        ds = client.create_dataset(dataset_name=dataset_name)
+    client.create_examples(
+        inputs=[e.inputs for e in examples],
+        outputs=[e.outputs for e in examples],
+        dataset_id=ds.id,
+    )
+    print("Done creating dataset.")
+
+
+if __name__ == "__main__":
+    import argparse
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--target-api-key", type=str, required=False)
+    parser.add_argument("--target-endpoint", type=str, required=False)
+    parser.add_argument("--dataset-name", type=str, default=DATASET_NAME)
+    parser.add_argument(
+        "--public-dataset-token", type=str, default=PUBLIC_DATASET_TOKEN
+    )
+    args = parser.parse_args()
+    client = None
+    if args.target_api_key or args.target_endpoint:
+        client = Client(
+            api_key=args.target_api_key,
+            api_url=args.target_endpoint,
+        )
+    create_langchain_docs_dataset(
+        dataset_name=args.dataset_name,
+        public_dataset_token=args.public_dataset_token,
+        client=client,
+    )

archived/langchain-docs-benchmarking/run_evals.py

@@ -1,6 +1,5 @@
 import argparse
 import importlib.util
-import os
 import sys
 import uuid
 from functools import partial

archived/langchain-docs-benchmarking/run_experiments.py

@@ -120,6 +120,7 @@ if __name__ == "__main__":
         ]
 
     for experiment in selected_experiments:
+        print("Running experiment:", experiment)
         main(
             **experiment,
             dataset_name=args.dataset_name,

docs/Makefile

@@ -0,0 +1,20 @@
+# Minimal makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line, and also
+# from the environment for the first two.
+SPHINXOPTS    ?=
+SPHINXBUILD   ?= sphinx-build
+SOURCEDIR     = source
+BUILDDIR      = build
+
+# Put it first so that "make" without argument is like "make help".
+help:
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+.PHONY: help Makefile
+
+# Catch-all target: route all unknown targets to Sphinx using the new
+# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
+%: Makefile
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

docs/make.bat

@@ -0,0 +1,35 @@
+@ECHO OFF
+
+pushd %~dp0
+
+REM Command file for Sphinx documentation
+
+if "%SPHINXBUILD%" == "" (
+	set SPHINXBUILD=sphinx-build
+)
+set SOURCEDIR=source
+set BUILDDIR=build
+
+%SPHINXBUILD% >NUL 2>NUL
+if errorlevel 9009 (
+	echo.
+	echo.The 'sphinx-build' command was not found. Make sure you have Sphinx
+	echo.installed, then set the SPHINXBUILD environment variable to point
+	echo.to the full path of the 'sphinx-build' executable. Alternatively you
+	echo.may add the Sphinx directory to PATH.
+	echo.
+	echo.If you don't have Sphinx installed, grab it from
+	echo.https://www.sphinx-doc.org/
+	exit /b 1
+)
+
+if "%1" == "" goto help
+
+%SPHINXBUILD% -M %1 %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
+goto end
+
+:help
+%SPHINXBUILD% -M help %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
+
+:end
+popd

docs/source/.gitignore

@@ -0,0 +1,3 @@
+chromadb/
+index.md
+Untitled.ipynb

docs/source/conf.py

@@ -0,0 +1,113 @@
+# Configuration file for the Sphinx documentation builder.
+#
+# This file only contains a selection of the most common options. For a full
+# list see the documentation:
+# https://www.sphinx-doc.org/en/master/usage/configuration.html
+
+# -- Path setup --------------------------------------------------------------
+
+# If extensions (or modules to document with autodoc) are in another directory,
+# add these directories to sys.path here. If the directory is relative to the
+# documentation root, use os.path.abspath to make it absolute, like shown here.
+#
+# import os
+# import sys
+# sys.path.insert(0, os.path.abspath('.'))
+
+
+# -- Project information -----------------------------------------------------
+import pathlib
+import sys
+from typing import List
+
+import toml
+
+ROOT_FOLDER = str(pathlib.Path(__file__).parent.parent.parent)
+
+# Add the project root to the path
+sys.path.insert(0, ROOT_FOLDER)
+
+with open("../../pyproject.toml") as f:
+    data = toml.load(f)
+
+project = "LangChain Benchmarks"
+copyright = "2023, Langchain AI"
+author = "Langchain AI"
+
+version = data["tool"]["poetry"]["version"]
+release = version
+
+html_title = project + " " + version
+
+
+# -- General configuration ---------------------------------------------------
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
+# ones.
+extensions = [
+    "sphinx.ext.autodoc",
+    "sphinx.ext.autodoc.typehints",
+    "sphinx.ext.autosummary",
+    "sphinx.ext.napoleon",
+    "sphinx.ext.viewcode",
+    "myst_nb",
+    "sphinx_copybutton",
+    "IPython.sphinxext.ipython_console_highlighting",
+]
+source_suffix = [".ipynb", ".html", ".md", ".rst"]
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ["_templates"]
+
+# List of patterns, relative to source directory, that match files and
+# directories to ignore when looking for source files.
+# This pattern also affects html_static_path and html_extra_path.
+exclude_patterns: List[str] = []
+
+
+# -- Options for HTML output -------------------------------------------------
+
+# The theme to use for HTML and HTML Help pages.  See the documentation for
+# a list of builtin themes.
+#
+html_theme = "sphinx_book_theme"
+
+html_theme_options = {
+    "path_to_docs": "docs/source",
+    "repository_url": "https://github.com/langchain-ai/langchain-benchmarks",
+    "home_page_in_toc": True,
+    "show_navbar_depth": 2,
+    "use_sidenotes": True,
+    "use_repository_button": True,
+    "use_issues_button": True,
+    "use_source_button": True,
+    "use_fullscreen_button": True,
+    "repository_branch": "main",
+    "launch_buttons": {
+        "notebook_interface": "jupyterlab",
+        "colab_url": "https://colab.research.google.com",
+    },
+}
+
+html_context = {
+    "display_github": True,  # Integrate GitHub
+    "github_user": "langchain-ai",  # Username
+    "github_repo": "langchain-benchmarks",  # Repo name
+    "github_version": "main",  # Version
+    "conf_py_path": "/docs/",  # Path in the checkout to the docs root
+}
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ["_static"]
+
+# These paths are either relative to html_static_path
+# or fully qualified paths (eg. https://...)
+html_css_files = [
+    "css/custom.css",
+]
+
+nb_execution_mode = "off"
+autosummary_generate = True

docs/source/notebooks/datasets.ipynb

@@ -0,0 +1,226 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "033684fb-65b2-4586-a959-68c614741ca2",
+   "metadata": {},
+   "source": [
+    "# Datasets\n",
+    "\n",
+    "Here, we'll see how to work with LangSmith datasets."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "474292e6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%pip install -U langchain-benchmarks"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "6d272fbf-710e-4a49-a0da-67e010541905",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from langchain_benchmarks import clone_public_dataset, download_public_dataset"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "18ee0f96-e5c4-4ae9-aebf-7d8b88c51662",
+   "metadata": {},
+   "source": [
+    "Let's first download the dataset to the local file system"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "58b94f6d-0c91-4361-9b22-f758ffaa150a",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Fetching examples...\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5a2fad8c0c3549ec96a3b38fe8a002b0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/21 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Done fetching examples.\n"
+     ]
+    }
+   ],
+   "source": [
+    "download_public_dataset(\n",
+    "    \"https://smith.langchain.com/public/452ccafc-18e1-4314-885b-edd735f17b9d/examples\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "841db832-b0d3-4fd1-8531-1154ec9b3caa",
+   "metadata": {},
+   "source": [
+    "we can take a look at the first two examples"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "664e90fc-af84-4c5f-a3dd-5d9ffe649650",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[\n",
+      "  {\n",
+      "    \"created_at\": \"2023-11-15T15:26:53.511629\",\n",
+      "    \"dataset_id\": \"9f73165c-d333-4d14-8f59-bd7eede5db08\",\n",
+      "    \"id\": \"0703a989-2693-4039-a1f6-7281fc1b4cb0\",\n",
+      "    \"inputs\": {\n",
+      "      \"question\": \"do bob and alice live in the same city?\"\n",
+      "    },\n",
+      "    \"modified_at\": \"2023-11-15T15:26:53.511629\",\n",
+      "    \"outputs\": {\n",
+      "      \"expected_steps\": [\n",
+      "        \"find_users_by_name\",\n",
+      "        \"get_user_location\",\n",
+      "        \"get_city_for_location\",\n",
+      "        \"get_user_location\",\n",
+      "        \"get_city_for_location\"\n",
+      "      ],\n",
+      "      \"order_matters\": false,\n",
+      "      \"reference\": \"no\"\n",
+      "    },\n",
+      "    \"runs\": []\n",
+      "  },\n",
+      "  {\n",
+      "    \"created_at\": \"2023-11-15T15:26:53.491359\",\n",
+      "    \"dataset_id\": \"9f73165c-d333-4d14-8f59-bd7eede5db08\",\n",
+      "    \"id\": \"b258b95a-9524-4da7-b758-c5481109322d\",\n",
+      "    \"inputs\": {\n",
+      "      \"question\": \"Is it likely that Donna is outside with an umbrella at this time?\"\n",
+      "    },\n",
+      "    \"modified_at\": \"2023-11-15T15:26:53.491359\",\n",
+      "    \"outputs\": {\n",
+      "      \"expected_steps\": [\n",
+      "        \"find_users_by_name\",\n",
+      "        \"get_user_location\",\n",
+      "        \"get_current_time_for_location\",\n",
+      "        \"get_current_weather_for_location\"\n",
+      "      ],\n",
+      "      \"order_matters\": false,\n",
+      "      \"reference\": \"yes\"\n",
+      "    },\n",
+      "    \"runs\": []\n",
+      "  }\n",
+      "]\n"
+     ]
+    }
+   ],
+   "source": [
+    "import json\n",
+    "\n",
+    "with open(\"./e95d45da-aaa3-44b3-ba2b-7c15ff6e46f5.json\", \"r\", encoding=\"utf-8\") as f:\n",
+    "    print(json.dumps(json.load(f)[:2], indent=2, sort_keys=True))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2c6cf01f-466b-406d-b4c7-2395747780fd",
+   "metadata": {},
+   "source": [
+    "We can also clone the dataset to our local tenant"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e4dea4df-2f1c-436b-a71c-49ffb2295ccc",
+   "metadata": {},
+   "source": [
+    "Executing this command will clone the dataset to your own LangSmith tenant. \n",
+    "For this to work you must have a [LangSmith account](https://smith.langchain.com/) set up."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7eb38ea6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "\n",
+    "# Get from https://smith.langchain.com/settings\n",
+    "os.environ[\"LANGCHAIN_API_KEY\"] = \"ls_...\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "18d0b905-2a6a-4752-a7cb-8653bd9049e3",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "clone_public_dataset(\n",
+    "    \"https://smith.langchain.com/public/452ccafc-18e1-4314-885b-edd735f17b9d/examples\",\n",
+    "    dataset_name=\"Agent Dataset\",\n",
+    ")"
+   ]
+  }
+ ],
+ "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": "3.11.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

docs/source/notebooks/extraction/intro.ipynb

@@ -0,0 +1,130 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "7e8fc49a-e8b2-404b-a059-e9f668c460e5",
+   "metadata": {},
+   "source": [
+    "# Introduction\n",
+    "\n",
+    "\n",
+    "\n",
+    "These tasks refer to an LLM's ability to extract structured output from an unstructured source, such as emails, websites, or other text. Below are a list of supported datasets.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "86912590-a90a-4351-8ab4-89192cdee1e7",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table>\n",
+       "<thead>\n",
+       "<tr><th>Name            </th><th>Type          </th><th>Dataset ID                                                                                                                                                 </th><th>Description  </th></tr>\n",
+       "</thead>\n",
+       "<tbody>\n",
+       "<tr><td>Email Extraction</td><td>ExtractionTask</td><td><a href=\"https://smith.langchain.com/public/a1742786-bde5-4f51-a1d8-e148e5251ddb/d\" target=\"_blank\" rel=\"noopener\">a1742786-bde5-4f51-a1d8-e148e5251ddb</a></td><td>A dataset of 42 real emails deduped from a spam folder, with semantic HTML tags removed, as well as a script for initial extraction and formatting of other emails from an arbitrary .mbox file like the one exported by Gmail.\n",
+       "\n",
+       "Some additional cleanup of the data was done by hand after the initial pass.\n",
+       "\n",
+       "See https://github.com/jacoblee93/oss-model-extraction-evals.              </td></tr>\n",
+       "<tr><td>Chat Extraction </td><td>ExtractionTask</td><td><a href=\"https://smith.langchain.com/public/00f4444c-9460-4a82-b87a-f50096f1cfef/d\" target=\"_blank\" rel=\"noopener\">00f4444c-9460-4a82-b87a-f50096f1cfef</a></td><td>A dataset meant to test the ability of an LLM to extract and infer\n",
+       "structured information from a dialogue. The dialogue is between a user and a support\n",
+       "engineer. Outputs should be structured as a JSON object and test both the ability\n",
+       "of the LLM to correctly structure the information and its ability to perform simple \n",
+       "classification tasks.              </td></tr>\n",
+       "</tbody>\n",
+       "</table>"
+      ],
+      "text/plain": [
+       "Registry(tasks=[ExtractionTask(name='Email Extraction', dataset_id='https://smith.langchain.com/public/a1742786-bde5-4f51-a1d8-e148e5251ddb/d', description='A dataset of 42 real emails deduped from a spam folder, with semantic HTML tags removed, as well as a script for initial extraction and formatting of other emails from an arbitrary .mbox file like the one exported by Gmail.\\n\\nSome additional cleanup of the data was done by hand after the initial pass.\\n\\nSee https://github.com/jacoblee93/oss-model-extraction-evals.\\n    ', schema=<class 'langchain_benchmarks.extraction.tasks.email_task.Email'>, instructions=ChatPromptTemplate(input_variables=['input'], messages=[SystemMessagePromptTemplate(prompt=PromptTemplate(input_variables=[], template='You are an expert researcher.')), HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['input'], template='What can you tell me about the following email? Make sure to extract the question in the correct format. Here is the email:\\n ```\\n{input}\\n```'))])), ExtractionTask(name='Chat Extraction', dataset_id='https://smith.langchain.com/public/00f4444c-9460-4a82-b87a-f50096f1cfef/d', description='A dataset meant to test the ability of an LLM to extract and infer\\nstructured information from a dialogue. The dialogue is between a user and a support\\nengineer. Outputs should be structured as a JSON object and test both the ability\\nof the LLM to correctly structure the information and its ability to perform simple \\nclassification tasks.', schema=<class 'langchain_benchmarks.extraction.tasks.chat_extraction.schema.GenerateTicket'>, instructions=ChatPromptTemplate(input_variables=['dialogue'], messages=[SystemMessagePromptTemplate(prompt=PromptTemplate(input_variables=[], template='You are a helpdesk assistant responsible with extracting information and generating tickets. Dialogues are between a user and a support engineer.')), HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['dialogue'], template='Generate a ticket for the following question-response pair:\\n<Dialogue>\\n{dialogue}\\n</Dialogue>'))]))])"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from langchain_benchmarks import registry\n",
+    "\n",
+    "registry.filter(Type=\"ExtractionTask\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e771e544-b8f5-4359-8fd7-c89b71fbe460",
+   "metadata": {},
+   "source": [
+    "### Task resources\n",
+    "\n",
+    "In addition to the dataset_id, name, and description, each extraction task provides the following:\n",
+    "\n",
+    "- `schema` - a pydantic base model defining the schema (or schemas) the model should extract\n",
+    "\n",
+    "\n",
+    "### Dataset schema\n",
+    "\n",
+    "Each task corresponds to a LangSmith dataset with the following schema:\n",
+    "\n",
+    "Inputs:\n",
+    "- `input: str` - the input text\n",
+    "\n",
+    "Outputs\n",
+    "- `output: str` - the expected extraction result, as a json object\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d04e05f3-3f20-4fed-bb98-3eb072213bbd",
+   "metadata": {},
+   "source": [
+    "### Evaluation\n",
+    "\n",
+    "The extraction tasks also have an evaluation config, which defines default LangSmith evaluators to apply when benchmarking your architecture.\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "9c7865bd-8251-4579-85a3-f9085d96f497",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from langchain.chat_models import ChatOpenAI\n",
+    "\n",
+    "from langchain_benchmarks.extraction import get_eval_config\n",
+    "\n",
+    "eval_llm = ChatOpenAI(model=\"gpt-4\", model_kwargs={\"seed\": 42})\n",
+    "eval_config = get_eval_config(eval_llm)"
+   ]
+  }
+ ],
+ "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": "3.11.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}