Problem to ingest files

[iRitiLopes]

Hi guys, Im facing a problem when I try to ingest files into. Always gives me this error

18:39:16.700 [INFO    ]            uvicorn.access - 127.0.0.1:9617 - "POST /upload HTTP/1.1" 200
18:39:16.705 [INFO    ]            uvicorn.access - 127.0.0.1:9617 - "POST /queue/join HTTP/1.1" 200
18:39:16.710 [INFO    ]            uvicorn.access - 127.0.0.1:9617 - "GET /queue/data?session_hash=nx8aimvqzq HTTP/1.1" 200
18:39:16.741 [INFO    ] private_gpt.server.ingest.ingest_service - Ingesting file_names=['teste.txt']
Parsing nodes: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 498.79it/s]
Generating embeddings: 100%|█████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:01<00:00, 12.95it/s]
Generating embeddings: 0it [00:00, ?it/s]
Traceback (most recent call last):
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\gradio\queueing.py", line 495, in call_prediction
    output = await route_utils.call_process_api(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\gradio\route_utils.py", line 235, in call_process_api
    output = await app.get_blocks().process_api(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\gradio\blocks.py", line 1627, in process_api
    result = await self.call_function(
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\gradio\blocks.py", line 1173, in call_function
    prediction = await anyio.to_thread.run_sync(
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
           ^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\gradio\utils.py", line 690, in wrapper
    response = f(*args, **kwargs)
               ^^^^^^^^^^^^^^^^^^
  File "H:\privateGPT\privateGPT\private_gpt\ui\ui.py", line 252, in _upload_file
    self._ingest_service.bulk_ingest([(str(path.name), path) for path in paths])
  File "H:\privateGPT\privateGPT\private_gpt\server\ingest\ingest_service.py", line 84, in bulk_ingest
    documents = self.ingest_component.bulk_ingest(files)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\privateGPT\privateGPT\private_gpt\components\ingest\ingest_component.py", line 133, in bulk_ingest
    saved_documents.extend(self._save_docs(documents))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\privateGPT\privateGPT\private_gpt\components\ingest\ingest_component.py", line 140, in _save_docs
    self._index.insert(document, show_progress=True)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\llama_index\core\indices\base.py", line 231, in insert
    self.insert_nodes(nodes, **insert_kwargs)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\llama_index\core\indices\vector_store\base.py", line 320, in insert_nodes
    self._insert(nodes, **insert_kwargs)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\llama_index\core\indices\vector_store\base.py", line 311, in _insert
    self._add_nodes_to_index(self._index_struct, nodes, **insert_kwargs)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\llama_index\core\indices\vector_store\base.py", line 233, in _add_nodes_to_index
    new_ids = self._vector_store.add(nodes_batch, **insert_kwargs)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\llama_index\vector_stores\qdrant\base.py", line 256, in add
    self._client.upload_points(
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\qdrant_client\qdrant_client.py", line 1865, in upload_points
    return self._client.upload_points(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\qdrant_client\local\qdrant_local.py", line 687, in upload_points
    self._upload_points(collection_name, points)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\qdrant_client\local\qdrant_local.py", line 701, in _upload_points
    collection.upsert(
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\qdrant_client\local\local_collection.py", line 1108, in upsert
    self._upsert_point(point)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\qdrant_client\local\local_collection.py", line 1100, in _upsert_point
    self._add_point(point)
  File "C:\Users\richa\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-jkqtB0bC-py3.11\Lib\site-packages\qdrant_client\local\local_collection.py", line 1042, in _add_point
    named_vectors[idx] = vector_np
    ~~~~~~~~~~~~~^^^^^
ValueError: could not broadcast input array from shape (384,) into shape (768,)

I'm running on local profile

[dougbristor]

I had the same problem. This is a workaround as I couldn't fix for qdrant, so I use chroma instead.
I edited both settings.yaml and settings-local.yaml
vectorstore: database: chroma #qdrant
Then ran with
PGPT_PROFILES=local make run

(N.B. I thought there might be some other paths to set, but this seemed to work. during my attempt I did type: -
poetry install --extras "ui llms-llama-cpp embeddings-huggingface vector-stores-chroma"
CMAKE_ARGS='-DLLAMA_CUBLAS=on' poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python
PGPT_PROFILES=local make run
)

[miguel-br-dl]

FIXED for me: I had had the same issue here.

My config IS:

`
vectorstore:
database: qdrant

qdrant:
path: local_data/private_gpt/qdrant
`

For me, the problem was that I tried another embedding (with 768) so qdrant was already created with this space.
Even asking to Delete all files, the path local_data/private_gpt/qdrant remained with 768 config (see metadata.json)

1. So I move it away:

mv local_data/private_gpt/qdrant local_data/private_gpt/oldqdrant

2. And restarted privateGPT.

The folder was restablished and I could ingest new files.

3. Ingestion OK!

Hope this helps!!

[qdm12]

Many thanks, this helped me! 👍

[dbzoo]

The abilty to reset when the dimension change was added in #1783
For example if you are using a local-settings.yaml profile:
% PGPT_PROFILES=local make wipe