pseudotensor
commented
4 months ago It should look like:
--document_choice="['file2.pdf']"
for Fire to process it into list, the others cause problems in bash command line.
If you are in a different command line, like pycharm, then don't need the extra outer double quotes.
Hi, tried to search relevant issues but did not find any. when trying to run --langchain_mode with my db and with --document_choice like so: python generate.py --base_model=https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/resolve/main/zephyr-7b-beta.Q2_K.gguf --prompt_type=zephyr --langchain_mode=Duck --document_choice=[file2.pdf] --hf_embedding_model=sentence-transformers/all-MiniLM-L6-v2 --score_model=None --llamacpp_dict="{'n_gpu_layers':10}" --max_seq_len=1024 --enable_tts=False --enable_stt=False --enable_transcriptions=False
and like so: --document_choice="file2.pdf" --document_choice=["file2.pdf"] --document_choice=file2.pdf
i get the following error:
File "/home/tomershm/h2ogpt/generate.py", line 20, in
entrypoint_main()
File "/home/tomershm/h2ogpt/generate.py", line 16, in entrypoint_main
H2O_Fire(main)
File "/home/tomershm/h2ogpt/src/utils.py", line 69, in H2O_Fire
fire.Fire(component=component, command=args)
File "/home/tomershm/miniconda3/envs/h2ogpt/lib/python3.10/site-packages/fire/core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "/home/tomershm/miniconda3/envs/h2ogpt/lib/python3.10/site-packages/fire/core.py", line 475, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "/home/tomershm/miniconda3/envs/h2ogpt/lib/python3.10/site-packages/fire/core.py", line 691, in _CallAndUpdateTrace
component = fn(*varargs, kwargs)
File "/home/tomershm/h2ogpt/src/gen.py", line 2203, in main
go_gradio(locals())
File "/home/tomershm/h2ogpt/src/gradio_runner.py", line 872, in go_gradio
docs_state00 = kwargs['document_choice'] + [DocumentChoice.ALL.value]
TypeError: can only concatenate str (not "list") to str
I am basically trying to select specific documents to work with after creating a db via make_db. would love some help or clarification on how to use document selection. thank you!