Testing in a macbook pro

[@alonso_isidoro]

Thanks for sharing the model, I have been able to test it on my macbook pro, i9 with 32 GB of ram. I notice that the cpu goes to 400% when inferring the answer, and the gpu goes to 0%. Is it possible to make the model use the gpu? Radeon Pro Vega 20 4GB

[TouchstoneTheDev]

The required package sentence-transformers is available. There is a hyphen instead of underscore in the package name.
Currently the model is the quantized version of the mpt-30b-chat model which is being loaded with Ctransformers. You can use the original huggingface llm with this code

import transformers
llm = transformers.AutoModelForCausalLM.from_pretrained(
  'mosaicml/mpt-30b-chat',
  trust_remote_code=True
)

You should replace lines 73 to 78 in question_answer_docs.py with the above code. It should work fine ( I haven't tried) with GPU if you have the GPU which can hold the 30b parameter model.

refer this "#1 (comment)"

[@alonso_isidoro]

Hi!, i have tried with your suggestion, but i am getting this error:

pip --version
pip 23.2.1 from /usr/local/lib/python3.11/site-packages/pip (python 3.11)
python3.11 -m pip install --upgrade pip
Requirement already satisfied: pip in /usr/local/lib/python3.11/site-packages (23.2.1)
python3.11 -m pip install einops
Requirement already satisfied: einops in /usr/local/lib/python3.11/site-packages (0.6.1)
make server
poetry run python question_answer_docs_server.py
embeddings_model_name is all-MiniLM-L6-v2
persist_directory is db
model_path is models/mpt-30b-chat.ggmlv0.q4_1.bin
target_source_chunks is 4
Initializing model for the first time...
Loading model... models/mpt-30b-chat.ggmlv0.q4_1.bin
This modeling file requires the following packages that were not found in your environment: einops. Run pip install einops
Traceback (most recent call last):
File "/Users/aironman/git/private-chatbot-mpt30b-langchain/question_answer_docs_server.py", line 139, in
init()
File "/Users/aironman/git/private-chatbot-mpt30b-langchain/question_answer_docs_server.py", line 127, in init
llm=load_model(),
^^^^^^^^^^^^
File "/Users/aironman/git/private-chatbot-mpt30b-langchain/question_answer_docs_server.py", line 58, in load_model
llm = transformers.AutoModelForCausalLM.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/aironman/Library/Caches/pypoetry/virtualenvs/private-chatbot-mpt30b-aREpNEbB-py3.11/lib/python3.11/site-packages/transformers/models/auto/auto_factory.py", line 475, in from_pretrained
model_class = get_class_from_dynamic_module(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/aironman/Library/Caches/pypoetry/virtualenvs/private-chatbot-mpt30b-aREpNEbB-py3.11/lib/python3.11/site-packages/transformers/dynamic_module_utils.py", line 431, in get_class_from_dynamic_module
final_module = get_cached_module_file(
^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/aironman/Library/Caches/pypoetry/virtualenvs/private-chatbot-mpt30b-aREpNEbB-py3.11/lib/python3.11/site-packages/transformers/dynamic_module_utils.py", line 306, in get_cached_module_file
get_cached_module_file(
File "/Users/aironman/Library/Caches/pypoetry/virtualenvs/private-chatbot-mpt30b-aREpNEbB-py3.11/lib/python3.11/site-packages/transformers/dynamic_module_utils.py", line 268, in get_cached_module_file
modules_needed = check_imports(resolved_module_file)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/aironman/Library/Caches/pypoetry/virtualenvs/private-chatbot-mpt30b-aREpNEbB-py3.11/lib/python3.11/site-packages/transformers/dynamic_module_utils.py", line 151, in check_imports
raise ImportError(
ImportError: This modeling file requires the following packages that were not found in your environment: einops. Run pip install einops
make: *** [server] Error 1

i have tried to install einops, but, as you can see i am using poetry