search

marella / chatdocs

Chat with your documents offline using AI.
MIT License
704 stars 100 forks source link

index not found #57

Open abhishekrai43 opened 1 year ago

abhishekrai43 commented 1 year ago

Not sure what is wrong here. the db and index folders are at their default locations. Even so I copied them to each and evey folder possible. But still it is giving the same error. 

File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\langchain\vectorstores\chroma.py", line 312, in similarity_search_with_score
    results = self.__query_collection(
              ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\langchain\utils\utils.py", line 30, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\langchain\vectorstores\chroma.py", line 145, in __query_collection
    return self._collection.query(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\chromadb\api\models\Collection.py", line 227, in query
    return self._client._query(
           ^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\chromadb\api\local.py", line 437, in _query
    uuids, distances = self._db.get_nearest_neighbors(
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\chromadb\db\clickhouse.py", line 585, in get_nearest_neighbors
    uuids, distances = index.get_nearest_neighbors(embeddings, n_results, ids)
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Documents\chatdocs\chatdocs\Lib\site-packages\chromadb\db\index\hnswlib.py", line 240, in get_nearest_neighbors
    raise NoIndexException(
chromadb.errors.NoIndexException: Index not found, please create an instance before querying
abhishekrai43 commented 1 year ago

load INSTRUCTOR_Transformer max_seq_length 512 Using embedded DuckDB with persistence: data will be stored in: db CUDA extension not installed. The safetensors archive passed at C:\Users\Administrator/.cache\huggingface\hub\models--TheBloke--Llama-2-7b-Chat-GPTQ\snapshots\b7ee6c20ac0bba85a310dc699d6bb4c845811608\gptq_model-4bit-128g.safetensors does not contain metadata. Make sure to save your model with the save_pretrained method. Defaulting to 'pt' metadata. skip module injection for FusedLlamaMLPForQuantizedModel not support integrate without triton yet. A matching Triton is not available, some optimizations will not be enabled. Error caught was: No module named 'triton'

llm: gptq

gptq: model: TheBloke/Llama-2-7b-Chat-GPTQ model_file: C:\Users\Administrator\Downloads\gptq_model-4bit-128g.safetensors device: 0 pipeline_kwargs: max_new_tokens: 256

download: false

host: localhost port: 5000 auth: false

chroma: persist_directory: db chroma_db_impl: duckdb+parquet anonymized_telemetry: false

retriever: search_kwargs: k: 4 , could you please guide me a little on how to get this "lighting fast" performance out of this? I am on a Tesla T4 AWS EC-2. If there is no other option than to move to linux, I will do that as well, but it will increase my work manyfold, so tell me if there is a way to get this done on the same machine please.

vmirea commented 1 year ago

Got the same error: File "/media/vmirea/NTFS_8TB/projects/chatdocs/venv/lib/python3.11/site-packages/chromadb/db/index/hnswlib.py", line 240, in get_nearest_neighbors raise NoIndexException( chromadb.errors.NoIndexException: Index not found, please create an instance before querying

(venv) (base) vmirea@vmirea-Z390-GAMING-SLI:/media/vmirea/NTFS_8TB/projects/chatdocs$ chatdocs ui load INSTRUCTOR_Transformer max_seq_length 512 Using embedded DuckDB with persistence: data will be stored in: db