Spaces:
Sleeping
Sleeping
File size: 2,618 Bytes
186f8c5 38c77ce ef19ec7 577ce8d 2671147 ef19ec7 577ce8d 2671147 577ce8d ef19ec7 186f8c5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 |
import streamlit as st
from transformers import AutoTokenizer, AutoModelForCausalLM
import chromadb
from chromadb.config import Settings
from chromadb.utils import embedding_functions
from sentence_transformers import SentenceTransformer
import os
from huggingface_hub import login
from transformers import AutoModel
# Retrieve the API token from the environment variable
os.environ["HF_API_TOKEN"] = os.getenv("HF_TOKEN")
# Set your Hugging Face API token as an environment variable
os.environ["HF_API_TOKEN"] = "your_huggingface_api_token"
# Load the Llama model using Hugging Face Transformers
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
# Initialize ChromaDB
client = chromadb.Client(Settings(chroma_db_impl="duckdb+parquet", persist_directory="./chroma_db"))
# Create a collection for storing supply chain and green environment data
collection = client.get_or_create_collection(
name="supply_chain_green_environment",
embedding_function=embedding_functions.SentenceTransformerEmbeddingFunction(
model_name="all-mpnet-base-v2"
),
)
# Initialize the sentence transformer for generating embeddings
embedding_model = SentenceTransformer("all-mpnet-base-v2")
# Streamlit app title
st.title("Supply Chain & Green Environment Chatbot")
# User input for questions
user_question = st.text_input("Enter your question:")
# Chat history
if "chat_history" not in st.session_state:
st.session_state.chat_history = []
# Process user input and generate response
if user_question:
# Generate embedding for the user question
question_embedding = embedding_model.encode(user_question).tolist()
# Search for relevant information in the ChromaDB collection
results = collection.query(
query_embeddings=question_embedding,
n_results=3,
)
# Construct the context for the Llama model
context = ""
for doc in results["documents"][0]:
context += doc + "\n"
# Generate response from the Llama model
inputs = tokenizer(f"Context: {context}\n\nQuestion: {user_question}\n\nAnswer:", return_tensors="pt")
outputs = model.generate(**inputs, max_length=256)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
# Add user question and bot response to chat history
st.session_state.chat_history.append({"user": user_question, "bot": response})
# Display chat history
for message in st.session_state.chat_history:
st.write(f"**User:** {message['user']}")
st.write(f"**Bot:** {message['bot']}")
|