gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Some enhancements:

Browse Source 
- Enable the llm_cache storage to support get_by_mode_and_id, to improve the performance for using real KV server
- Provide an option for the developers to cache the LLM response when extracting entities for a document. Solving the paint point that sometimes the process failed, the processed chunks we need to call LLM again, money and time wasted. With the new option (by default not enabled) enabling, we can cache that result, can significantly save the time and money for beginners.
This commit is contained in:
Samuel Chan
2025-01-06 12:50:05 +08:00
parent 6c1b669f0f
commit 6ae27d8f06
7 changed files with 182 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

62
README.md
View File

@@ -26,6 +26,7 @@ This repository hosts the code of LightRAG. The structure of this code is based
</div>
## 🎉 News
- [x] [2025.01.06]🎯📢You can now [use PostgreSQL for Storage](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#using-postgres-for-storage).
- [x] [2024.12.31]🎯📢LightRAG now supports [deletion by document ID](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#delete).
- [x] [2024.11.25]🎯📢LightRAG now supports seamless integration of [custom knowledge graphs](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#insert-custom-kg), empowering users to enhance the system with their own domain expertise.
- [x] [2024.11.19]🎯📢A comprehensive guide to LightRAG is now available on [LearnOpenCV](https://learnopencv.com/lightrag). Many thanks to the blog author.
@@ -356,6 +357,11 @@ rag = LightRAG(
```
see test_neo4j.py for a working example.
### Using PostgreSQL for Storage
For production level scenarios you will most likely want to leverage an enterprise solution. PostgreSQL can provide a one-stop solution for you as KV store, VectorDB (pgvector) and GraphDB (apache AGE).
* PostgreSQL is lightweight,the whole binary distribution including all necessary plugins can be zipped to 40MB: Ref to [Windows Release](https://github.com/ShanGor/apache-age-windows/releases/tag/PG17%2Fv1.5.0-rc0) as it is easy to install for Linux/Mac.
* How to start? Ref to: [examples/lightrag_zhipu_postgres_demo.py](https://github.com/HKUDS/LightRAG/blob/main/examples/lightrag_zhipu_postgres_demo.py)
### Insert Custom KG
```python
@@ -602,33 +608,34 @@ if __name__ == "__main__":
### LightRAG init parameters
| **Parameter** | **Type** | **Explanation** | **Default** |
| --- | --- | --- | --- |
| **working\_dir** | `str` | Directory where the cache will be stored | `lightrag_cache+timestamp` |
| **kv\_storage** | `str` | Storage type for documents and text chunks. Supported types: `JsonKVStorage`, `OracleKVStorage` | `JsonKVStorage` |
| **vector\_storage** | `str` | Storage type for embedding vectors. Supported types: `NanoVectorDBStorage`, `OracleVectorDBStorage` | `NanoVectorDBStorage` |
| **graph\_storage** | `str` | Storage type for graph edges and nodes. Supported types: `NetworkXStorage`, `Neo4JStorage`, `OracleGraphStorage` | `NetworkXStorage` |
| **log\_level** | | Log level for application runtime | `logging.DEBUG` |
| **chunk\_token\_size** | `int` | Maximum token size per chunk when splitting documents | `1200` |
| **chunk\_overlap\_token\_size** | `int` | Overlap token size between two chunks when splitting documents | `100` |
| **tiktoken\_model\_name** | `str` | Model name for the Tiktoken encoder used to calculate token numbers | `gpt-4o-mini` |
| **entity\_extract\_max\_gleaning** | `int` | Number of loops in the entity extraction process, appending history messages | `1` |
| **entity\_summary\_to\_max\_tokens** | `int` | Maximum token size for each entity summary | `500` |
| **node\_embedding\_algorithm** | `str` | Algorithm for node embedding (currently not used) | `node2vec` |
| **node2vec\_params** | `dict` | Parameters for node embedding | `{"dimensions": 1536,"num_walks": 10,"walk_length": 40,"window_size": 2,"iterations": 3,"random_seed": 3,}` |
| **embedding\_func** | `EmbeddingFunc` | Function to generate embedding vectors from text | `openai_embedding` |
| **embedding\_batch\_num** | `int` | Maximum batch size for embedding processes (multiple texts sent per batch) | `32` |
| **embedding\_func\_max\_async** | `int` | Maximum number of concurrent asynchronous embedding processes | `16` |
| **llm\_model\_func** | `callable` | Function for LLM generation | `gpt_4o_mini_complete` |
| **llm\_model\_name** | `str` | LLM model name for generation | `meta-llama/Llama-3.2-1B-Instruct` |
| **llm\_model\_max\_token\_size** | `int` | Maximum token size for LLM generation (affects entity relation summaries) | `32768` |
| **llm\_model\_max\_async** | `int` | Maximum number of concurrent asynchronous LLM processes | `16` |
| **llm\_model\_kwargs** | `dict` | Additional parameters for LLM generation | |
| **vector\_db\_storage\_cls\_kwargs** | `dict` | Additional parameters for vector database (currently not used) | |
| **enable\_llm\_cache** | `bool` | If `TRUE`, stores LLM results in cache; repeated prompts return cached responses | `TRUE` |
| **addon\_params** | `dict` | Additional parameters, e.g., `{"example_number": 1, "language": "Simplified Chinese", "entity_types": ["organization", "person", "geo", "event"], "insert_batch_size": 10}`: sets example limit, output language, and batch size for document processing | `example_number: all examples, language: English, insert_batch_size: 10` |
| **convert\_response\_to\_json\_func** | `callable` | Not used | `convert_response_to_json` |
| **embedding\_cache\_config** | `dict` | Configuration for question-answer caching. Contains three parameters:<br>- `enabled`: Boolean value to enable/disable cache lookup functionality. When enabled, the system will check cached responses before generating new answers.<br>- `similarity_threshold`: Float value (0-1), similarity threshold. When a new question's similarity with a cached question exceeds this threshold, the cached answer will be returned directly without calling the LLM.<br>- `use_llm_check`: Boolean value to enable/disable LLM similarity verification. When enabled, LLM will be used as a secondary check to verify the similarity between questions before returning cached answers. | Default: `{"enabled": False, "similarity_threshold": 0.95, "use_llm_check": False}` |
| **Parameter** | **Type** | **Explanation** | **Default** |
|----------------------------------------------| --- |-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------|
| **working\_dir** | `str` | Directory where the cache will be stored | `lightrag_cache+timestamp` |
| **kv\_storage** | `str` | Storage type for documents and text chunks. Supported types: `JsonKVStorage`, `OracleKVStorage` | `JsonKVStorage` |
| **vector\_storage** | `str` | Storage type for embedding vectors. Supported types: `NanoVectorDBStorage`, `OracleVectorDBStorage` | `NanoVectorDBStorage` |
| **graph\_storage** | `str` | Storage type for graph edges and nodes. Supported types: `NetworkXStorage`, `Neo4JStorage`, `OracleGraphStorage` | `NetworkXStorage` |
| **log\_level** | | Log level for application runtime | `logging.DEBUG` |
| **chunk\_token\_size** | `int` | Maximum token size per chunk when splitting documents | `1200` |
| **chunk\_overlap\_token\_size** | `int` | Overlap token size between two chunks when splitting documents | `100` |
| **tiktoken\_model\_name** | `str` | Model name for the Tiktoken encoder used to calculate token numbers | `gpt-4o-mini` |
| **entity\_extract\_max\_gleaning** | `int` | Number of loops in the entity extraction process, appending history messages | `1` |
| **entity\_summary\_to\_max\_tokens** | `int` | Maximum token size for each entity summary | `500` |
| **node\_embedding\_algorithm** | `str` | Algorithm for node embedding (currently not used) | `node2vec` |
| **node2vec\_params** | `dict` | Parameters for node embedding | `{"dimensions": 1536,"num_walks": 10,"walk_length": 40,"window_size": 2,"iterations": 3,"random_seed": 3,}` |
| **embedding\_func** | `EmbeddingFunc` | Function to generate embedding vectors from text | `openai_embedding` |
| **embedding\_batch\_num** | `int` | Maximum batch size for embedding processes (multiple texts sent per batch) | `32` |
| **embedding\_func\_max\_async** | `int` | Maximum number of concurrent asynchronous embedding processes | `16` |
| **llm\_model\_func** | `callable` | Function for LLM generation | `gpt_4o_mini_complete` |
| **llm\_model\_name** | `str` | LLM model name for generation | `meta-llama/Llama-3.2-1B-Instruct` |
| **llm\_model\_max\_token\_size** | `int` | Maximum token size for LLM generation (affects entity relation summaries) | `32768` |
| **llm\_model\_max\_async** | `int` | Maximum number of concurrent asynchronous LLM processes | `16` |
| **llm\_model\_kwargs** | `dict` | Additional parameters for LLM generation | |
| **vector\_db\_storage\_cls\_kwargs** | `dict` | Additional parameters for vector database (currently not used) | |
| **enable\_llm\_cache** | `bool` | If `TRUE`, stores LLM results in cache; repeated prompts return cached responses | `TRUE` |
| **enable\_llm\_cache\_for\_entity\_extract** | `bool` | If `TRUE`, stores LLM results in cache for entity extraction; Good for beginners to debug your application | `FALSE` |
| **addon\_params** | `dict` | Additional parameters, e.g., `{"example_number": 1, "language": "Simplified Chinese", "entity_types": ["organization", "person", "geo", "event"], "insert_batch_size": 10}`: sets example limit, output language, and batch size for document processing | `example_number: all examples, language: English, insert_batch_size: 10` |
| **convert\_response\_to\_json\_func** | `callable` | Not used | `convert_response_to_json` |
| **embedding\_cache\_config** | `dict` | Configuration for question-answer caching. Contains three parameters:<br>- `enabled`: Boolean value to enable/disable cache lookup functionality. When enabled, the system will check cached responses before generating new answers.<br>- `similarity_threshold`: Float value (0-1), similarity threshold. When a new question's similarity with a cached question exceeds this threshold, the cached answer will be returned directly without calling the LLM.<br>- `use_llm_check`: Boolean value to enable/disable LLM similarity verification. When enabled, LLM will be used as a secondary check to verify the similarity between questions before returning cached answers. | Default: `{"enabled": False, "similarity_threshold": 0.95, "use_llm_check": False}` |
### Error Handling
<details>
@@ -1206,6 +1213,7 @@ curl "http://localhost:9621/health"
```
## Development
Contribute to the project: [Guide](contributor-readme.MD)
### Running in Development Mode