DMEvanCT / convesational_ai

Conversational AI application built using LangChain, Llama2 and Streamlit

Convesational AI

Conversational AI application: Chat with your documents 📚 using Llama2 🦙, AWS SageMaker 🧠, LangChain 🦜️🔗 and Streamlit 🔥

Running this application

1. Clone this repo ⏬

   git clone https://github.com/vas610/convesational_ai.git
   
   cd convesational_ai

2. Create conda 🐍 environment

   conda env create -f environment.yml python=3.10  # conda 22.9.0
   
   conda activate docai
   
   pip install -r requirements.txt
   
   python -m ipykernel install --user --name=conda_docai

3. Download required data 🔠

   wget --quiet https://docs.aws.amazon.com/sagemaker/latest/dg/sitemap.xml --output-document - | egrep -o "https://[^<]+" | wget --directory-prefix=./aws_docs/sagemaker/ -i -

4. Create and Store Embeddings 1️⃣0️⃣

   ./dataprep.py

5. Setup a SageMaker Endpoint with Llama2 🦙 by following this blog. I have used the meta-textgeneration-llama-2-7b-f model . Also, update the endpoint name in the .env file

6. Run the below command to start the streamlit app 🔥

   streamlit run streamlit_app.py --server.address 0.0.0.0 --server.port 8080 --server.fileWatcherType none --browser.gatherUsageStats False

2023-07-30_13-22-58-3.mp4

---