llama-index-packs-fuzzy-citation


Namellama-index-packs-fuzzy-citation JSON
Version 0.1.0 PyPI version JSON
download
home_page
Summaryllama-index packs fuzzy_citation integration
upload_time2024-02-10 23:14:43
maintainer
docs_urlNone
authorYour Name
requires_python>=3.8.1,<3.12
licenseMIT
keywords
VCS
bugtrack_url
requirements No requirements were recorded.
Travis-CI No Travis.
coveralls test coverage No coveralls.
            # Fuzzy Citation Query Engine Pack

Creates and runs a `CustomQueryEngine` -- `FuzzCitationQueryEngine` -- which post-processes response objects to identify source sentences using fuzzy matching.

The identified sentences are available in the `response.metadata` dictionary, containing a mapping of `(response_sentence, source_chunk)` -> `{"start_char_idx": idx, "end_char_idx": idx, "node" node}`. The start/end idxs represent the character indexes in the node text that the source chunk comes from.

The fuzzy matching uses `fuzz.ratio()` to compare sentences. The default threshold score is 50.

## CLI Usage

You can download llamapacks directly using `llamaindex-cli`, which comes installed with the `llama-index` python package:

```bash
llamaindex-cli download-llamapack FuzzyCitationEnginePack --download-dir ./fuzzy_citation_pack
```

You can then inspect the files at `./fuzzy_citation_pack` and use them as a template for your own project!

## Code Usage

You can download the pack to a the `./fuzzy_citation_pack` directory:

```python
from llama_index import Document, VectorStoreIndex
from llama_index.core.llama_pack import download_llama_pack

# download and install dependencies
FuzzyCitationEnginePack = download_llama_pack(
    "FuzzyCitationEnginePack", "./fuzzy_citation_pack"
)

index = VectorStoreIndex.from_documents([Document.example()])
query_engine = index.as_query_engine()

fuzzy_engine = FuzzyCitationEnginePack(query_engine, threshold=50)
```

The `run()` function is a light wrapper around `query_engine.query()`. The response will have metadata attached to it indicating the fuzzy citations.

```python
response = fuzzy_engine.run("What can you tell me about LLMs?")

# print source sentences
print(response.metadata.keys())

# print full source sentence info
print(response.metadata)
```

See the [notebook on llama-hub](https://github.com/run-llama/llama-hub/blob/main/llama_hub/llama_packs/fuzzy_citation/fuzzy_citation_example.ipynb) for a full example.

            

Raw data

            {
    "_id": null,
    "home_page": "",
    "name": "llama-index-packs-fuzzy-citation",
    "maintainer": "",
    "docs_url": null,
    "requires_python": ">=3.8.1,<3.12",
    "maintainer_email": "",
    "keywords": "",
    "author": "Your Name",
    "author_email": "you@example.com",
    "download_url": "https://files.pythonhosted.org/packages/30/a8/0e802896455c9992d62a21e9155c7bd582269a845bfcfb0d7162f480c517/llama_index_packs_fuzzy_citation-0.1.0.tar.gz",
    "platform": null,
    "description": "# Fuzzy Citation Query Engine Pack\n\nCreates and runs a `CustomQueryEngine` -- `FuzzCitationQueryEngine` -- which post-processes response objects to identify source sentences using fuzzy matching.\n\nThe identified sentences are available in the `response.metadata` dictionary, containing a mapping of `(response_sentence, source_chunk)` -> `{\"start_char_idx\": idx, \"end_char_idx\": idx, \"node\" node}`. The start/end idxs represent the character indexes in the node text that the source chunk comes from.\n\nThe fuzzy matching uses `fuzz.ratio()` to compare sentences. The default threshold score is 50.\n\n## CLI Usage\n\nYou can download llamapacks directly using `llamaindex-cli`, which comes installed with the `llama-index` python package:\n\n```bash\nllamaindex-cli download-llamapack FuzzyCitationEnginePack --download-dir ./fuzzy_citation_pack\n```\n\nYou can then inspect the files at `./fuzzy_citation_pack` and use them as a template for your own project!\n\n## Code Usage\n\nYou can download the pack to a the `./fuzzy_citation_pack` directory:\n\n```python\nfrom llama_index import Document, VectorStoreIndex\nfrom llama_index.core.llama_pack import download_llama_pack\n\n# download and install dependencies\nFuzzyCitationEnginePack = download_llama_pack(\n    \"FuzzyCitationEnginePack\", \"./fuzzy_citation_pack\"\n)\n\nindex = VectorStoreIndex.from_documents([Document.example()])\nquery_engine = index.as_query_engine()\n\nfuzzy_engine = FuzzyCitationEnginePack(query_engine, threshold=50)\n```\n\nThe `run()` function is a light wrapper around `query_engine.query()`. The response will have metadata attached to it indicating the fuzzy citations.\n\n```python\nresponse = fuzzy_engine.run(\"What can you tell me about LLMs?\")\n\n# print source sentences\nprint(response.metadata.keys())\n\n# print full source sentence info\nprint(response.metadata)\n```\n\nSee the [notebook on llama-hub](https://github.com/run-llama/llama-hub/blob/main/llama_hub/llama_packs/fuzzy_citation/fuzzy_citation_example.ipynb) for a full example.\n",
    "bugtrack_url": null,
    "license": "MIT",
    "summary": "llama-index packs fuzzy_citation integration",
    "version": "0.1.0",
    "project_urls": null,
    "split_keywords": [],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "d06b403d076b8e8bf3566ced133e5f3324a2e6333b38f1eff01f6e3b15df5975",
                "md5": "0c770194717a6e63110c2cde1f63da83",
                "sha256": "22867a10d329d6d2afb5dd649e033165d4ced2edfef962fc6f71437226f2b7c7"
            },
            "downloads": -1,
            "filename": "llama_index_packs_fuzzy_citation-0.1.0-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "0c770194717a6e63110c2cde1f63da83",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": ">=3.8.1,<3.12",
            "size": 4173,
            "upload_time": "2024-02-10T23:14:42",
            "upload_time_iso_8601": "2024-02-10T23:14:42.895239Z",
            "url": "https://files.pythonhosted.org/packages/d0/6b/403d076b8e8bf3566ced133e5f3324a2e6333b38f1eff01f6e3b15df5975/llama_index_packs_fuzzy_citation-0.1.0-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        },
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "30a80e802896455c9992d62a21e9155c7bd582269a845bfcfb0d7162f480c517",
                "md5": "f7e662c7eab5f4b87897e94c39ae0e0c",
                "sha256": "e6fd5a50e50037cb49db7c3e9c882499b3667953d18a87a7bc79880238aea43b"
            },
            "downloads": -1,
            "filename": "llama_index_packs_fuzzy_citation-0.1.0.tar.gz",
            "has_sig": false,
            "md5_digest": "f7e662c7eab5f4b87897e94c39ae0e0c",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": ">=3.8.1,<3.12",
            "size": 3609,
            "upload_time": "2024-02-10T23:14:43",
            "upload_time_iso_8601": "2024-02-10T23:14:43.844926Z",
            "url": "https://files.pythonhosted.org/packages/30/a8/0e802896455c9992d62a21e9155c7bd582269a845bfcfb0d7162f480c517/llama_index_packs_fuzzy_citation-0.1.0.tar.gz",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2024-02-10 23:14:43",
    "github": false,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "lcname": "llama-index-packs-fuzzy-citation"
}
        
Elapsed time: 2.48591s