Chat interface created

[ombhojane]

I have created a chat interface in Streamlit.
I've connected with our model.

Preview :

Please help to resolve this issue: I am unable to setup this project locally. the 4.7 GB model get installed, when I run main file. config.py also worked. But Builder.py not worked. Heence chormaDB not loaded and unable to run main.py
Since I've chat interface with main.py, It can work in your local system!

Improvements : this is first version of chat interface. we can add voice/enter arrow and make UI better of this.

[Priyamakeshwari]

Can you send what error do you get when you run the builder.py file

[ombhojane]

load INSTRUCTOR_Transformer
max_seq_length 512
Traceback (most recent call last):
File "c:\Users\Om\Desktop\Hactoberfest\TeachGPT\builder.py", line 49, in
builder()
File "c:\Users\Om\Desktop\Hactoberfest\TeachGPT\builder.py", line 37, in builder
db = Chroma.from_documents(
^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Om\AppData\Local\Programs\Python\Python311\Lib\site-packages\langchain\vectorstores\chroma.py", line 646, in from_documents
return cls.from_texts(
^^^^^^^^^^^^^^^
File "C:\Users\Om\AppData\Local\Programs\Python\Python311\Lib\site-packages\langchain\vectorstores\chroma.py", line 610, in from_texts
chroma_collection.add_texts(texts=texts, metadatas=metadatas, ids=ids)
File "C:\Users\Om\AppData\Local\Programs\Python\Python311\Lib\site-packages\langchain\vectorstores\chroma.py", line 188, in add_texts
embeddings = self._embedding_function.embed_documents(texts)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Om\AppData\Local\Programs\Python\Python311\Lib\site-packages\langchain\embeddings\huggingface.py", line 171, in embed_documents
embeddings = self.client.encode(instruction_pairs, **self.encode_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Om\AppData\Local\Programs\Python\Python311\Lib\site-packages\InstructorEmbedding\instructor.py", line 524, in encode
if isinstance(sentences[0],list):
~~~~~~~~~^^^
IndexError: list index out of range

[Priyamakeshwari]

Try installing this library pip install instructorembedding and then run builder.py so vector store is formed and we can use RAG by running the main.py file. I'll update the requirements.txt file soon!

[Priyamakeshwari]

@ombhojane I was also facing the issue this morning. You can try this code as a temporary fix for avoiding that error.
Fix 1 : delete the db folder and then builder.py
Fix 2:

from chromadb.errors import InvalidDimensionException
try:
    docsearch = Chroma.from_documents(documents=..., embedding=...)
except InvalidDimensionException:
    Chroma().delete_collection()
    docsearch = Chroma.from_documents(documents=..., embedding=...)

[Priyamakeshwari]

@ombhojane Any updates on this PR?