Update app.py

app.py

@@ -1,62 +1,31 @@
 import gradio as gr
-from langchain_community.document_loaders import TextLoader
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.vectorstores import FAISS
-from langchain.llms import HuggingFacePipeline
-from langchain.chains import RetrievalQA
-from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+from huggingface_hub import InferenceClient
+import os
 
-# Load and process documents
-doc_loader = TextLoader("dataset.txt")
-docs = doc_loader.load()
-text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
-split_docs = text_splitter.split_documents(docs)
+# Use the Hugging Face Inference API
+client = InferenceClient("01-ai/Yi-Coder-9B-Chat")
 
-# Create vector database
-embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
-vectordb = FAISS.from_documents(split_docs, embeddings)
+# Load the dataset
+with open("dataset.txt", "r", encoding='utf-8') as f:
+    dataset = f.read()
 
-# Load model and create pipeline
-model_name = "01-ai/Yi-Coder-9B-Chat"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto")
-qa_pipeline = pipeline(
-    "text-generation",
-    model=model,
-    tokenizer=tokenizer,
-    max_new_tokens=500,
-    pad_token_id=tokenizer.eos_token_id
-)
+def chatbot_response(user_input):
+    # Combine the dataset and user input
+    prompt = f"""Based on the following dataset, answer the user's question:
 
-# Set up LangChain
-llm = HuggingFacePipeline(pipeline=qa_pipeline)
-retriever = vectordb.as_retriever(search_kwargs={"k": 5})
-qa_chain = RetrievalQA.from_chain_type(
-    retriever=retriever,
-    chain_type="stuff",
-    llm=llm,
-    return_source_documents=False
-)
+Dataset:
+{dataset}
 
-def preprocess_query(query):
-    if "script" in query or "code" in query.lower():
-        return f"Write a CPSL script: {query}"
-    return query
+User question: {user_input}
 
-def clean_response(response):
-    result = response.get("result", "")
-    if "Answer:" in result:
-        return result.split("Answer:")[1].strip()
-    return result.strip()
+Answer:"""
 
-def chatbot_response(user_input):
-    processed_query = preprocess_query(user_input)
-    raw_response = qa_chain.invoke({"query": processed_query})
-    return clean_response(raw_response)
+    # Use the Inference API to generate a response
+    response = client.text_generation(prompt, max_new_tokens=500, temperature=0.7)
+    return response
 
-# Gradio interface
-with gr.Blocks() as demo:
+# Set up the Gradio interface
+with gr.Blocks() as chat_interface:
     gr.Markdown("# CPSL Chatbot")
     chat_history = gr.Chatbot()
     user_input = gr.Textbox(label="Your Message:")
@@ -67,4 +36,8 @@ with gr.Blocks() as demo:
         history.append((user_message, bot_reply))
         return history, history
 
-    send_button.click(interact, inputs=[user_input, chat_history], outputs=[chat_history, chat_history])
+    send_button.click(interact, inputs=[user_input, chat_history], outputs=[chat_history, chat_history])
+
+# Launch the interface
+if __name__ == "__main__":
+    chat_interface.launch()