Update app.py

app.py

@@ -97,40 +97,26 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, in
     progress(0.9, desc="Done!")
     return qa_chain
 
-# Generate collection name for vector database
-def create_collection_name(filepath):
-    collection_name = Path(filepath).stem
-    collection_name = collection_name.replace(" ", "-")
-    collection_name = unidecode(collection_name)
-    collection_name = re.sub('[^A-Za-z0-9]+', '-', collection_name)
-    collection_name = collection_name[:50]
-    if len(collection_name) < 3:
-        collection_name = collection_name + 'xyz'
-    if not collection_name[0].isalnum():
-        collection_name = 'A' + collection_name[1:]
-    if not collection_name[-1].isalnum():
-        collection_name = collection_name[:-1] + 'Z'
-    print('Filepath: ', filepath)
-    print('Collection name: ', collection_name)
-    return collection_name
-
-# Initialize database
-def initialize_database(list_file_obj, chunk_size, chunk_overlap, db_type, progress=gr.Progress()):
-    list_file_path = [x.name for x in list_file_obj if x is not None]
-    progress(0.1, desc="Creating collection name...")
-    collection_name = create_collection_name(list_file_path[0])
-    progress(0.25, desc="Loading document...")
-    doc_splits = load_doc(list_file_path, chunk_size, chunk_overlap)
-    progress(0.5, desc="Generating vector database...")
-    vector_db = create_db(doc_splits, collection_name, db_type)
+def initialize_llm_no_doc(llm_model, temperature, max_tokens, top_k, initial_prompt, progress=gr.Progress()):
+    progress(0.1, desc="Initializing HF tokenizer...")
+    progress(0.5, desc="Initializing HF Hub...")
+    llm = HuggingFaceEndpoint(
+        repo_id=llm_model,
+        huggingfacehub_api_token=api_token,
+        temperature=temperature,
+        max_new_tokens=max_tokens,
+        top_k=top_k,
+    )
+    progress(0.75, desc="Defining buffer memory...")
+    memory = ConversationBufferMemory(
+        memory_key="chat_history",
+        output_key='answer',
+        return_messages=True
+    )
+    conversation_chain = ConversationChain(llm=llm, memory=memory, verbose=False)
+    conversation_chain({"question": initial_prompt})
     progress(0.9, desc="Done!")
-    return vector_db, collection_name, "Complete!"
-
-def initialize_LLM(llm_option, llm_temperature, max_tokens, top_k, vector_db, initial_prompt, progress=gr.Progress()):
-    llm_name = list_llm[llm_option]
-    print("llm_name: ", llm_name)
-    qa_chain = initialize_llmchain(llm_name, llm_temperature, max_tokens, top_k, vector_db, initial_prompt, progress)
-    return qa_chain, "Complete!"
+    return conversation_chain
 
 def format_chat_history(message, chat_history):
     formatted_chat_history = []
@@ -156,27 +142,6 @@ def conversation(qa_chain, message, history):
     new_history = history + [(message, response_answer)]
     return qa_chain, gr.update(value=""), new_history, response_source1, response_source1_page, response_source2, response_source2_page, response_source3, response_source3_page
 
-def initialize_llm_no_doc(llm_model, temperature, max_tokens, top_k, initial_prompt, progress=gr.Progress()):
-    progress(0.1, desc="Initializing HF tokenizer...")
-    progress(0.5, desc="Initializing HF Hub...")
-    llm = HuggingFaceEndpoint(
-        repo_id=llm_model,
-        huggingfacehub_api_token=api_token,
-        temperature=temperature,
-        max_new_tokens=max_tokens,
-        top_k=top_k,
-    )
-    progress(0.75, desc="Defining buffer memory...")
-    memory = ConversationBufferMemory(
-        memory_key="chat_history",
-        output_key='answer',
-        return_messages=True
-    )
-    conversation_chain = ConversationChain(llm=llm, memory=memory, verbose=False)
-    conversation_chain({"question": initial_prompt})
-    progress(0.9, desc="Done!")
-    return conversation_chain
-
 def conversation_no_doc(llm, message, history):
     formatted_chat_history = format_chat_history(message, history)
     response = llm({"question": message, "chat_history": formatted_chat_history})
@@ -222,19 +187,13 @@ def demo():
                 db_progress = gr.Textbox(label="Vector database initialization", value="None")
             with gr.Row():
                 db_btn = gr.Button("Generate vector database")
-        # Define o estado para o prompt inicial
-        initial_prompt = gr.State("")
-
-        # Define a aba "Set Initial Prompt"
+        
         with gr.Tab("Step 3 - Set Initial Prompt"):
             with gr.Row():
                 prompt_input = gr.Textbox(label="Initial Prompt", lines=5, value="Você é um advogado sênior, onde seu papel é analisar e trazer as informações sem inventar, dando a sua melhor opinião sempre trazendo contexto e referência. Aprenda o que é jurisprudência.")
             with gr.Row():
                 set_prompt_btn = gr.Button("Set Prompt")
 
-        # Atualiza o estado do prompt inicial ao clicar no botão "Set Prompt"
-        set_prompt_btn.click(fn=lambda prompt: prompt, inputs=prompt_input, outputs=initial_prompt)
-
         with gr.Tab("Step 4 - Initialize QA chain"):
             with gr.Row():
                 llm_btn = gr.Radio(list_llm_simple, 
@@ -295,7 +254,7 @@ def demo():
         db_btn.click(initialize_database, 
             inputs=[document, slider_chunk_size, slider_chunk_overlap, db_type_radio], 
             outputs=[vector_db, collection_name, db_progress])
-        set_prompt_btn.click(lambda prompt: prompt, 
+        set_prompt_btn.click(lambda prompt: gr.update(value=prompt), 
                              inputs=prompt_input, 
                              outputs=initial_prompt)
         qachain_btn.click(initialize_LLM,