├── .github └── workflows │ ├── publish.yml │ └── test.yml ├── .gitignore ├── LICENSE ├── README.md ├── llm_cluster.py ├── pyproject.toml └── tests ├── conftest.py └── test_llm_cluster.py /.github/workflows/publish.yml: -------------------------------------------------------------------------------- 1 | name: Publish Python Package 2 | 3 | on: 4 | release: 5 | types: [created] 6 | 7 | permissions: 8 | contents: read 9 | 10 | jobs: 11 | test: 12 | runs-on: ubuntu-latest 13 | strategy: 14 | matrix: 15 | python-version: ["3.8", "3.9", "3.10", "3.11"] 16 | steps: 17 | - uses: actions/checkout@v3 18 | - name: Set up Python ${{ matrix.python-version }} 19 | uses: actions/setup-python@v4 20 | with: 21 | python-version: ${{ matrix.python-version }} 22 | cache: pip 23 | cache-dependency-path: setup.py 24 | - name: Cache models 25 | uses: actions/cache@v3 26 | with: 27 | path: ~/.cache/torch 28 | key: ${{ runner.os }}-torch- 29 | - name: Install dependencies 30 | run: | 31 | pip install -e '.[test]' 32 | - name: Run tests 33 | run: | 34 | pytest -s 35 | deploy: 36 | runs-on: ubuntu-latest 37 | needs: [test] 38 | steps: 39 | - uses: actions/checkout@v3 40 | - name: Set up Python 41 | uses: actions/setup-python@v4 42 | with: 43 | python-version: "3.11" 44 | cache: pip 45 | cache-dependency-path: setup.py 46 | - name: Install dependencies 47 | run: | 48 | pip install setuptools wheel twine build 49 | - name: Publish 50 | env: 51 | TWINE_USERNAME: __token__ 52 | TWINE_PASSWORD: ${{ secrets.PYPI_TOKEN }} 53 | run: | 54 | python -m build 55 | twine upload dist/* 56 | 57 | -------------------------------------------------------------------------------- /.github/workflows/test.yml: -------------------------------------------------------------------------------- 1 | name: Test 2 | 3 | on: [push, pull_request] 4 | 5 | permissions: 6 | contents: read 7 | 8 | jobs: 9 | test: 10 | runs-on: ubuntu-latest 11 | strategy: 12 | matrix: 13 | python-version: ["3.8", "3.9", "3.10", "3.11"] 14 | steps: 15 | - uses: actions/checkout@v3 16 | - name: Set up Python ${{ matrix.python-version }} 17 | uses: actions/setup-python@v4 18 | with: 19 | python-version: ${{ matrix.python-version }} 20 | cache: pip 21 | cache-dependency-path: setup.py 22 | - name: Cache models 23 | uses: actions/cache@v3 24 | with: 25 | path: ~/.cache/torch 26 | key: ${{ runner.os }}-torch- 27 | - name: Install dependencies 28 | run: | 29 | pip install -e '.[test]' 30 | - name: Run tests 31 | run: | 32 | pytest -s 33 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | .venv 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | venv 6 | .eggs 7 | .pytest_cache 8 | *.egg-info 9 | .DS_Store 10 | .vscode 11 | dist 12 | build 13 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | Apache License 2 | Version 2.0, January 2004 3 | http://www.apache.org/licenses/ 4 | 5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 6 | 7 | 1. Definitions. 8 | 9 | "License" shall mean the terms and conditions for use, reproduction, 10 | and distribution as defined by Sections 1 through 9 of this document. 11 | 12 | "Licensor" shall mean the copyright owner or entity authorized by 13 | the copyright owner that is granting the License. 14 | 15 | "Legal Entity" shall mean the union of the acting entity and all 16 | other entities that control, are controlled by, or are under common 17 | control with that entity. For the purposes of this definition, 18 | "control" means (i) the power, direct or indirect, to cause the 19 | direction or management of such entity, whether by contract or 20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the 21 | outstanding shares, or (iii) beneficial ownership of such entity. 22 | 23 | "You" (or "Your") shall mean an individual or Legal Entity 24 | exercising permissions granted by this License. 25 | 26 | "Source" form shall mean the preferred form for making modifications, 27 | including but not limited to software source code, documentation 28 | source, and configuration files. 29 | 30 | "Object" form shall mean any form resulting from mechanical 31 | transformation or translation of a Source form, including but 32 | not limited to compiled object code, generated documentation, 33 | and conversions to other media types. 34 | 35 | "Work" shall mean the work of authorship, whether in Source or 36 | Object form, made available under the License, as indicated by a 37 | copyright notice that is included in or attached to the work 38 | (an example is provided in the Appendix below). 39 | 40 | "Derivative Works" shall mean any work, whether in Source or Object 41 | form, that is based on (or derived from) the Work and for which the 42 | editorial revisions, annotations, elaborations, or other modifications 43 | represent, as a whole, an original work of authorship. For the purposes 44 | of this License, Derivative Works shall not include works that remain 45 | separable from, or merely link (or bind by name) to the interfaces of, 46 | the Work and Derivative Works thereof. 47 | 48 | "Contribution" shall mean any work of authorship, including 49 | the original version of the Work and any modifications or additions 50 | to that Work or Derivative Works thereof, that is intentionally 51 | submitted to Licensor for inclusion in the Work by the copyright owner 52 | or by an individual or Legal Entity authorized to submit on behalf of 53 | the copyright owner. For the purposes of this definition, "submitted" 54 | means any form of electronic, verbal, or written communication sent 55 | to the Licensor or its representatives, including but not limited to 56 | communication on electronic mailing lists, source code control systems, 57 | and issue tracking systems that are managed by, or on behalf of, the 58 | Licensor for the purpose of discussing and improving the Work, but 59 | excluding communication that is conspicuously marked or otherwise 60 | designated in writing by the copyright owner as "Not a Contribution." 61 | 62 | "Contributor" shall mean Licensor and any individual or Legal Entity 63 | on behalf of whom a Contribution has been received by Licensor and 64 | subsequently incorporated within the Work. 65 | 66 | 2. Grant of Copyright License. Subject to the terms and conditions of 67 | this License, each Contributor hereby grants to You a perpetual, 68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 69 | copyright license to reproduce, prepare Derivative Works of, 70 | publicly display, publicly perform, sublicense, and distribute the 71 | Work and such Derivative Works in Source or Object form. 72 | 73 | 3. Grant of Patent License. Subject to the terms and conditions of 74 | this License, each Contributor hereby grants to You a perpetual, 75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 76 | (except as stated in this section) patent license to make, have made, 77 | use, offer to sell, sell, import, and otherwise transfer the Work, 78 | where such license applies only to those patent claims licensable 79 | by such Contributor that are necessarily infringed by their 80 | Contribution(s) alone or by combination of their Contribution(s) 81 | with the Work to which such Contribution(s) was submitted. If You 82 | institute patent litigation against any entity (including a 83 | cross-claim or counterclaim in a lawsuit) alleging that the Work 84 | or a Contribution incorporated within the Work constitutes direct 85 | or contributory patent infringement, then any patent licenses 86 | granted to You under this License for that Work shall terminate 87 | as of the date such litigation is filed. 88 | 89 | 4. Redistribution. You may reproduce and distribute copies of the 90 | Work or Derivative Works thereof in any medium, with or without 91 | modifications, and in Source or Object form, provided that You 92 | meet the following conditions: 93 | 94 | (a) You must give any other recipients of the Work or 95 | Derivative Works a copy of this License; and 96 | 97 | (b) You must cause any modified files to carry prominent notices 98 | stating that You changed the files; and 99 | 100 | (c) You must retain, in the Source form of any Derivative Works 101 | that You distribute, all copyright, patent, trademark, and 102 | attribution notices from the Source form of the Work, 103 | excluding those notices that do not pertain to any part of 104 | the Derivative Works; and 105 | 106 | (d) If the Work includes a "NOTICE" text file as part of its 107 | distribution, then any Derivative Works that You distribute must 108 | include a readable copy of the attribution notices contained 109 | within such NOTICE file, excluding those notices that do not 110 | pertain to any part of the Derivative Works, in at least one 111 | of the following places: within a NOTICE text file distributed 112 | as part of the Derivative Works; within the Source form or 113 | documentation, if provided along with the Derivative Works; or, 114 | within a display generated by the Derivative Works, if and 115 | wherever such third-party notices normally appear. The contents 116 | of the NOTICE file are for informational purposes only and 117 | do not modify the License. You may add Your own attribution 118 | notices within Derivative Works that You distribute, alongside 119 | or as an addendum to the NOTICE text from the Work, provided 120 | that such additional attribution notices cannot be construed 121 | as modifying the License. 122 | 123 | You may add Your own copyright statement to Your modifications and 124 | may provide additional or different license terms and conditions 125 | for use, reproduction, or distribution of Your modifications, or 126 | for any such Derivative Works as a whole, provided Your use, 127 | reproduction, and distribution of the Work otherwise complies with 128 | the conditions stated in this License. 129 | 130 | 5. Submission of Contributions. Unless You explicitly state otherwise, 131 | any Contribution intentionally submitted for inclusion in the Work 132 | by You to the Licensor shall be under the terms and conditions of 133 | this License, without any additional terms or conditions. 134 | Notwithstanding the above, nothing herein shall supersede or modify 135 | the terms of any separate license agreement you may have executed 136 | with Licensor regarding such Contributions. 137 | 138 | 6. Trademarks. This License does not grant permission to use the trade 139 | names, trademarks, service marks, or product names of the Licensor, 140 | except as required for reasonable and customary use in describing the 141 | origin of the Work and reproducing the content of the NOTICE file. 142 | 143 | 7. Disclaimer of Warranty. Unless required by applicable law or 144 | agreed to in writing, Licensor provides the Work (and each 145 | Contributor provides its Contributions) on an "AS IS" BASIS, 146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or 147 | implied, including, without limitation, any warranties or conditions 148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A 149 | PARTICULAR PURPOSE. You are solely responsible for determining the 150 | appropriateness of using or redistributing the Work and assume any 151 | risks associated with Your exercise of permissions under this License. 152 | 153 | 8. Limitation of Liability. In no event and under no legal theory, 154 | whether in tort (including negligence), contract, or otherwise, 155 | unless required by applicable law (such as deliberate and grossly 156 | negligent acts) or agreed to in writing, shall any Contributor be 157 | liable to You for damages, including any direct, indirect, special, 158 | incidental, or consequential damages of any character arising as a 159 | result of this License or out of the use or inability to use the 160 | Work (including but not limited to damages for loss of goodwill, 161 | work stoppage, computer failure or malfunction, or any and all 162 | other commercial damages or losses), even if such Contributor 163 | has been advised of the possibility of such damages. 164 | 165 | 9. Accepting Warranty or Additional Liability. While redistributing 166 | the Work or Derivative Works thereof, You may choose to offer, 167 | and charge a fee for, acceptance of support, warranty, indemnity, 168 | or other liability obligations and/or rights consistent with this 169 | License. However, in accepting such obligations, You may act only 170 | on Your own behalf and on Your sole responsibility, not on behalf 171 | of any other Contributor, and only if You agree to indemnify, 172 | defend, and hold each Contributor harmless for any liability 173 | incurred by, or claims asserted against, such Contributor by reason 174 | of your accepting any such warranty or additional liability. 175 | 176 | END OF TERMS AND CONDITIONS 177 | 178 | APPENDIX: How to apply the Apache License to your work. 179 | 180 | To apply the Apache License to your work, attach the following 181 | boilerplate notice, with the fields enclosed by brackets "[]" 182 | replaced with your own identifying information. (Don't include 183 | the brackets!) The text should be enclosed in the appropriate 184 | comment syntax for the file format. We also recommend that a 185 | file or class name and description of purpose be included on the 186 | same "printed page" as the copyright notice for easier 187 | identification within third-party archives. 188 | 189 | Copyright [yyyy] [name of copyright owner] 190 | 191 | Licensed under the Apache License, Version 2.0 (the "License"); 192 | you may not use this file except in compliance with the License. 193 | You may obtain a copy of the License at 194 | 195 | http://www.apache.org/licenses/LICENSE-2.0 196 | 197 | Unless required by applicable law or agreed to in writing, software 198 | distributed under the License is distributed on an "AS IS" BASIS, 199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 200 | See the License for the specific language governing permissions and 201 | limitations under the License. 202 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # llm-cluster 2 | 3 | [![PyPI](https://img.shields.io/pypi/v/llm-cluster.svg)](https://pypi.org/project/llm-cluster/) 4 | [![Changelog](https://img.shields.io/github/v/release/simonw/llm-cluster?include_prereleases&label=changelog)](https://github.com/simonw/llm-cluster/releases) 5 | [![Tests](https://github.com/simonw/llm-cluster/workflows/Test/badge.svg)](https://github.com/simonw/llm-cluster/actions?query=workflow%3ATest) 6 | [![License](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](https://github.com/simonw/llm-cluster/blob/main/LICENSE) 7 | 8 | [LLM](https://llm.datasette.io/) plugin for clustering embeddings 9 | 10 | Background on this project: [Clustering with llm-cluster](https://simonwillison.net/2023/Sep/4/llm-embeddings/#llm-cluster). 11 | 12 | ## Installation 13 | 14 | Install this plugin in the same environment as LLM. 15 | ```bash 16 | llm install llm-cluster 17 | ``` 18 | 19 | ## Usage 20 | 21 | The plugin adds a new command, `llm cluster`. This command takes the name of an [embedding collection](https://llm.datasette.io/en/stable/embeddings/cli.html#storing-embeddings-in-sqlite) and the number of clusters to return. 22 | 23 | First, use [paginate-json](https://github.com/simonw/paginate-json) and [jq](https://stedolan.github.io/jq/) to populate a collection. I this case we are embedding the title and body of every issue in the [llm repository](https://github.com/simonw/llm), and storing the result in a `issues.db` database: 24 | ```bash 25 | paginate-json 'https://api.github.com/repos/simonw/llm/issues?state=all&filter=all' \ 26 | | jq '[.[] | {id: .id, title: .title}]' \ 27 | | llm embed-multi llm-issues - \ 28 | --database issues.db --store 29 | ``` 30 | The `--store` flag causes the content to be stored in the database along with the embedding vectors. 31 | 32 | Now we can cluster those embeddings into 10 groups: 33 | ```bash 34 | llm cluster llm-issues 10 \ 35 | -d issues.db 36 | ``` 37 | If you omit the `-d` option the default embeddings database will be used. 38 | 39 | The output should look something like this (truncated): 40 | ```json 41 | [ 42 | { 43 | "id": "2", 44 | "items": [ 45 | { 46 | "id": "1650662628", 47 | "content": "Initial design" 48 | }, 49 | { 50 | "id": "1650682379", 51 | "content": "Log prompts and responses to SQLite" 52 | } 53 | ] 54 | }, 55 | { 56 | "id": "4", 57 | "items": [ 58 | { 59 | "id": "1650760699", 60 | "content": "llm web command - launches a web server" 61 | }, 62 | { 63 | "id": "1759659476", 64 | "content": "`llm models` command" 65 | }, 66 | { 67 | "id": "1784156919", 68 | "content": "`llm.get_model(alias)` helper" 69 | } 70 | ] 71 | }, 72 | { 73 | "id": "7", 74 | "items": [ 75 | { 76 | "id": "1650765575", 77 | "content": "--code mode for outputting code" 78 | }, 79 | { 80 | "id": "1659086298", 81 | "content": "Accept PROMPT from --stdin" 82 | }, 83 | { 84 | "id": "1714651657", 85 | "content": "Accept input from standard in" 86 | } 87 | ] 88 | } 89 | ] 90 | ``` 91 | The content displayed is truncated to 100 characters. Pass `--truncate 0` to disable truncation, or `--truncate X` to truncate to X characters. 92 | 93 | ## Generating summaries for each cluster 94 | 95 | The `--summary` flag will cause the plugin to generate a summary for each cluster, by passing the content of the items (truncated according to the `--truncate` option) through a prompt to a Large Language Model. 96 | 97 | This feature is still experimental. You should experiment with custom prompts to improve the quality of your summaries. 98 | 99 | Since this can run a large amount of text through a LLM this can be expensive, depending on which model you are using. 100 | 101 | This feature only works for embeddings that have had their associated content stored in the database using the `--store` flag. 102 | 103 | You can use it like this: 104 | 105 | ```bash 106 | llm cluster llm-issues 10 \ 107 | -d issues.db \ 108 | --summary 109 | ``` 110 | This uses the default prompt and the default model. 111 | 112 | Partial example output: 113 | ```json 114 | [ 115 | { 116 | "id": "5", 117 | "items": [ 118 | { 119 | "id": "1650682379", 120 | "content": "Log prompts and responses to SQLite" 121 | }, 122 | { 123 | "id": "1650757081", 124 | "content": "Command for browsing captured logs" 125 | } 126 | ], 127 | "summary": "Log Management and Interactive Prompt Tracking" 128 | }, 129 | { 130 | "id": "6", 131 | "items": [ 132 | { 133 | "id": "1650771320", 134 | "content": "Mechanism for continuing an existing conversation" 135 | }, 136 | { 137 | "id": "1740090291", 138 | "content": "-c option for continuing a chat (using new chat_id column)" 139 | }, 140 | { 141 | "id": "1784122278", 142 | "content": "Figure out truncation strategy for continue conversation mode" 143 | } 144 | ], 145 | "summary": "Continuing Conversation Mechanism and Management" 146 | } 147 | ] 148 | ``` 149 | 150 | To use a different model, e.g. GPT-4, pass the `--model` option: 151 | ```bash 152 | llm cluster llm-issues 10 \ 153 | -d issues.db \ 154 | --summary \ 155 | --model gpt-4 156 | ``` 157 | The default prompt used is: 158 | 159 | > Short, concise title for this cluster of related documents. 160 | 161 | To use a custom prompt, pass `--prompt`: 162 | 163 | ```bash 164 | llm cluster llm-issues 10 \ 165 | -d issues.db \ 166 | --summary \ 167 | --model gpt-4 \ 168 | --prompt 'Summarize this in a short line in the style of a bored, angry panda' 169 | ``` 170 | A `"summary"` key will be added to each cluster, containing the generated summary. 171 | 172 | ## Development 173 | 174 | To set up this plugin locally, first checkout the code. Then create a new virtual environment: 175 | ```bash 176 | cd llm-cluster 177 | python3 -m venv venv 178 | source venv/bin/activate 179 | ``` 180 | Now install the dependencies and test dependencies: 181 | ```bash 182 | pip install -e '.[test]' 183 | ``` 184 | To run the tests: 185 | ```bash 186 | pytest 187 | ``` 188 | -------------------------------------------------------------------------------- /llm_cluster.py: -------------------------------------------------------------------------------- 1 | import click 2 | import json 3 | import llm 4 | import numpy as np 5 | import sklearn.cluster 6 | import sqlite_utils 7 | import textwrap 8 | 9 | DEFAULT_SUMMARY_PROMPT = """ 10 | Short, concise title for this cluster of related documents. 11 | """.strip() 12 | 13 | 14 | @llm.hookimpl 15 | def register_commands(cli): 16 | @cli.command() 17 | @click.argument("collection") 18 | @click.argument("n", type=int) 19 | @click.option( 20 | "--truncate", 21 | type=int, 22 | default=100, 23 | help="Truncate content to this many characters - 0 for no truncation", 24 | ) 25 | @click.option( 26 | "-d", 27 | "--database", 28 | type=click.Path( 29 | file_okay=True, allow_dash=False, dir_okay=False, writable=True 30 | ), 31 | envvar="LLM_EMBEDDINGS_DB", 32 | help="SQLite database file containing embeddings", 33 | ) 34 | @click.option( 35 | "--summary", is_flag=True, help="Generate summary title for each cluster" 36 | ) 37 | @click.option("-m", "--model", help="LLM model to use for the summary") 38 | @click.option("--prompt", help="Custom prompt to use for the summary") 39 | def cluster(collection, n, truncate, database, summary, model, prompt): 40 | """ 41 | Generate clusters from embeddings in a collection 42 | 43 | Example usage, to create 10 clusters: 44 | 45 | \b 46 | llm cluster my_collection 10 47 | 48 | Outputs a JSON array of {"id": "cluster_id", "items": [list of items]} 49 | 50 | Pass --summary to generate a summary for each cluster, using the default 51 | language model or the model you specify with --model. 52 | """ 53 | from llm.cli import get_default_model, get_key 54 | 55 | clustering_model = sklearn.cluster.MiniBatchKMeans(n_clusters=n, n_init="auto") 56 | if database: 57 | db = sqlite_utils.Database(database) 58 | else: 59 | db = sqlite_utils.Database(llm.user_dir() / "embeddings.db") 60 | rows = [ 61 | (row[0], llm.decode(row[1]), row[2]) 62 | for row in db.execute( 63 | """ 64 | select id, embedding, content from embeddings 65 | where collection_id = ( 66 | select id from collections where name = ? 67 | ) 68 | """, 69 | [collection], 70 | ).fetchall() 71 | ] 72 | to_cluster = np.array([item[1] for item in rows]) 73 | clustering_model.fit(to_cluster) 74 | assignments = clustering_model.labels_ 75 | 76 | def truncate_text(text): 77 | if not text: 78 | return None 79 | if truncate > 0: 80 | return text[:truncate] 81 | else: 82 | return text 83 | 84 | # Each one corresponds to an ID 85 | clusters = {} 86 | for (id, _, content), cluster in zip(rows, assignments): 87 | clusters.setdefault(str(cluster), []).append( 88 | {"id": str(id), "content": truncate_text(content)} 89 | ) 90 | # Re-arrange into a list 91 | output_clusters = [{"id": k, "items": v} for k, v in clusters.items()] 92 | 93 | # Do we need to generate summaries? 94 | if summary: 95 | model = llm.get_model(model or get_default_model()) 96 | if model.needs_key: 97 | model.key = get_key("", model.needs_key, model.key_env_var) 98 | prompt = prompt or DEFAULT_SUMMARY_PROMPT 99 | click.echo("[") 100 | for cluster, is_last in zip( 101 | output_clusters, [False] * (len(output_clusters) - 1) + [True] 102 | ): 103 | click.echo(" {") 104 | click.echo(' "id": {},'.format(json.dumps(cluster["id"]))) 105 | click.echo( 106 | ' "items": ' 107 | + textwrap.indent( 108 | json.dumps(cluster["items"], indent=2), " " 109 | ).lstrip() 110 | + "," 111 | ) 112 | prompt_content = "\n".join( 113 | [item["content"] for item in cluster["items"] if item["content"]] 114 | ) 115 | if prompt_content.strip(): 116 | summary = model.prompt( 117 | prompt_content, 118 | system=prompt, 119 | ).text() 120 | else: 121 | summary = None 122 | click.echo(' "summary": {}'.format(json.dumps(summary))) 123 | click.echo(" }" + ("," if not is_last else "")) 124 | click.echo("]") 125 | else: 126 | click.echo(json.dumps(output_clusters, indent=4)) 127 | -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- 1 | [project] 2 | name = "llm-cluster" 3 | version = "0.2" 4 | description = "LLM plugin for clustering embeddings" 5 | readme = "README.md" 6 | authors = [{name = "Simon Willison"}] 7 | license = {text = "Apache-2.0"} 8 | classifiers = [ 9 | "License :: OSI Approved :: Apache Software License" 10 | ] 11 | dependencies = [ 12 | "llm>=0.9", 13 | "scikit-learn" 14 | ] 15 | 16 | [project.urls] 17 | Homepage = "https://github.com/simonw/llm-cluster" 18 | Changelog = "https://github.com/simonw/llm-cluster/releases" 19 | Issues = "https://github.com/simonw/llm-cluster/issues" 20 | CI = "https://github.com/simonw/llm-cluster/actions" 21 | 22 | [project.entry-points.llm] 23 | cluster = "llm_cluster" 24 | 25 | [project.optional-dependencies] 26 | test = ["pytest"] 27 | -------------------------------------------------------------------------------- /tests/conftest.py: -------------------------------------------------------------------------------- 1 | import llm 2 | from llm.plugins import pm 3 | import pytest 4 | 5 | 6 | class LengthSummary(llm.Model): 7 | model_id = "length-summary" 8 | 9 | def execute(self, prompt, stream, response, conversation): 10 | return ["Length: {}".format(len(prompt.prompt))] 11 | 12 | 13 | class SimpleEmbeddings(llm.EmbeddingModel): 14 | model_id = "simple-embeddings" 15 | 16 | def embed_batch(self, texts): 17 | for text in texts: 18 | words = text.split()[:16] 19 | embedding = [len(word) for word in words] 20 | # Pad with 0 up to 16 words 21 | embedding += [0] * (16 - len(embedding)) 22 | yield embedding 23 | 24 | 25 | @pytest.fixture(autouse=True) 26 | def register_models(): 27 | class ModelsPlugin: 28 | __name__ = "ModelsPlugin" 29 | 30 | @llm.hookimpl 31 | def register_models(self, register): 32 | register(LengthSummary()) 33 | 34 | @llm.hookimpl 35 | def register_embedding_models(self, register): 36 | register(SimpleEmbeddings()) 37 | 38 | pm.register(ModelsPlugin(), name="undo-demo-plugin") 39 | try: 40 | yield 41 | finally: 42 | pm.unregister(name="undo-demo-plugin") 43 | -------------------------------------------------------------------------------- /tests/test_llm_cluster.py: -------------------------------------------------------------------------------- 1 | from click.testing import CliRunner 2 | import json 3 | from llm.cli import cli 4 | import llm 5 | import pytest 6 | import sqlite_utils 7 | 8 | 9 | @pytest.fixture 10 | def db_path(tmpdir): 11 | db_path = tmpdir / "data.db" 12 | db = sqlite_utils.Database(str(db_path)) 13 | collection = llm.Collection("entries", db, model_id="simple-embeddings") 14 | collection.embed_multi( 15 | [ 16 | (1, "one word"), 17 | (2, "two words"), 18 | (3, "three thing"), 19 | (4, "fourth thing"), 20 | (5, "fifth thing"), 21 | (6, "sixth thing"), 22 | (7, "seventh thing"), 23 | (8, "eighth thing"), 24 | (9, "ninth thing"), 25 | (10, "tenth thing"), 26 | ], 27 | store=True, 28 | ) 29 | return db_path 30 | 31 | 32 | @pytest.mark.parametrize("n", (2, 3)) 33 | def test_cluster(db_path, n): 34 | db = sqlite_utils.Database(str(db_path)) 35 | assert db["embeddings"].count == 10 36 | runner = CliRunner() 37 | result = runner.invoke( 38 | cli, 39 | [ 40 | "cluster", 41 | "entries", 42 | "--database", 43 | str(db_path), 44 | str(n), 45 | ], 46 | ) 47 | assert result.exit_code == 0, result.output 48 | clusters = json.loads(result.output) 49 | assert len(clusters) == n 50 | for cluster in clusters: 51 | assert isinstance(cluster["id"], str) 52 | assert isinstance(cluster["items"], list) 53 | 54 | 55 | @pytest.mark.parametrize("truncate", (0, 3)) 56 | def test_truncate(db_path, truncate): 57 | runner = CliRunner() 58 | result = runner.invoke( 59 | cli, 60 | [ 61 | "cluster", 62 | "entries", 63 | "--database", 64 | str(db_path), 65 | "2", 66 | "--truncate", 67 | str(truncate), 68 | ], 69 | ) 70 | assert result.exit_code == 0, result.output 71 | clusters = json.loads(result.output) 72 | if truncate == 0: 73 | # Should not have been truncated 74 | assert all( 75 | len(item["content"]) > 3 76 | for cluster in clusters 77 | for item in cluster["items"] 78 | ) 79 | else: 80 | assert all( 81 | len(item["content"]) == truncate 82 | for cluster in clusters 83 | for item in cluster["items"] 84 | ) 85 | 86 | 87 | @pytest.mark.parametrize("content_available", (True, False)) 88 | def test_cluster_summary(db_path, content_available): 89 | db = sqlite_utils.Database(str(db_path)) 90 | if not content_available: 91 | with db.conn: 92 | db.execute("update embeddings set content = null") 93 | runner = CliRunner() 94 | result = runner.invoke( 95 | cli, 96 | [ 97 | "cluster", 98 | "entries", 99 | "--database", 100 | str(db_path), 101 | "4", 102 | "--summary", 103 | "--model", 104 | "length-summary", 105 | ], 106 | ) 107 | assert result.exit_code == 0, result.output 108 | clusters = json.loads(result.output) 109 | if content_available: 110 | assert all(cluster["summary"].startswith("Length: ") for cluster in clusters) 111 | else: 112 | assert all(cluster["summary"] is None for cluster in clusters) 113 | --------------------------------------------------------------------------------