Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
|
@@ -97,40 +97,26 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, in
|
|
| 97 |
progress(0.9, desc="Done!")
|
| 98 |
return qa_chain
|
| 99 |
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
|
| 114 |
-
|
| 115 |
-
|
| 116 |
-
|
| 117 |
-
|
| 118 |
-
def initialize_database(list_file_obj, chunk_size, chunk_overlap, db_type, progress=gr.Progress()):
|
| 119 |
-
list_file_path = [x.name for x in list_file_obj if x is not None]
|
| 120 |
-
progress(0.1, desc="Creating collection name...")
|
| 121 |
-
collection_name = create_collection_name(list_file_path[0])
|
| 122 |
-
progress(0.25, desc="Loading document...")
|
| 123 |
-
doc_splits = load_doc(list_file_path, chunk_size, chunk_overlap)
|
| 124 |
-
progress(0.5, desc="Generating vector database...")
|
| 125 |
-
vector_db = create_db(doc_splits, collection_name, db_type)
|
| 126 |
progress(0.9, desc="Done!")
|
| 127 |
-
return
|
| 128 |
-
|
| 129 |
-
def initialize_LLM(llm_option, llm_temperature, max_tokens, top_k, vector_db, initial_prompt, progress=gr.Progress()):
|
| 130 |
-
llm_name = list_llm[llm_option]
|
| 131 |
-
print("llm_name: ", llm_name)
|
| 132 |
-
qa_chain = initialize_llmchain(llm_name, llm_temperature, max_tokens, top_k, vector_db, initial_prompt, progress)
|
| 133 |
-
return qa_chain, "Complete!"
|
| 134 |
|
| 135 |
def format_chat_history(message, chat_history):
|
| 136 |
formatted_chat_history = []
|
|
@@ -156,27 +142,6 @@ def conversation(qa_chain, message, history):
|
|
| 156 |
new_history = history + [(message, response_answer)]
|
| 157 |
return qa_chain, gr.update(value=""), new_history, response_source1, response_source1_page, response_source2, response_source2_page, response_source3, response_source3_page
|
| 158 |
|
| 159 |
-
def initialize_llm_no_doc(llm_model, temperature, max_tokens, top_k, initial_prompt, progress=gr.Progress()):
|
| 160 |
-
progress(0.1, desc="Initializing HF tokenizer...")
|
| 161 |
-
progress(0.5, desc="Initializing HF Hub...")
|
| 162 |
-
llm = HuggingFaceEndpoint(
|
| 163 |
-
repo_id=llm_model,
|
| 164 |
-
huggingfacehub_api_token=api_token,
|
| 165 |
-
temperature=temperature,
|
| 166 |
-
max_new_tokens=max_tokens,
|
| 167 |
-
top_k=top_k,
|
| 168 |
-
)
|
| 169 |
-
progress(0.75, desc="Defining buffer memory...")
|
| 170 |
-
memory = ConversationBufferMemory(
|
| 171 |
-
memory_key="chat_history",
|
| 172 |
-
output_key='answer',
|
| 173 |
-
return_messages=True
|
| 174 |
-
)
|
| 175 |
-
conversation_chain = ConversationChain(llm=llm, memory=memory, verbose=False)
|
| 176 |
-
conversation_chain({"question": initial_prompt})
|
| 177 |
-
progress(0.9, desc="Done!")
|
| 178 |
-
return conversation_chain
|
| 179 |
-
|
| 180 |
def conversation_no_doc(llm, message, history):
|
| 181 |
formatted_chat_history = format_chat_history(message, history)
|
| 182 |
response = llm({"question": message, "chat_history": formatted_chat_history})
|
|
@@ -222,19 +187,13 @@ def demo():
|
|
| 222 |
db_progress = gr.Textbox(label="Vector database initialization", value="None")
|
| 223 |
with gr.Row():
|
| 224 |
db_btn = gr.Button("Generate vector database")
|
| 225 |
-
|
| 226 |
-
initial_prompt = gr.State("")
|
| 227 |
-
|
| 228 |
-
# Define a aba "Set Initial Prompt"
|
| 229 |
with gr.Tab("Step 3 - Set Initial Prompt"):
|
| 230 |
with gr.Row():
|
| 231 |
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.")
|
| 232 |
with gr.Row():
|
| 233 |
set_prompt_btn = gr.Button("Set Prompt")
|
| 234 |
|
| 235 |
-
# Atualiza o estado do prompt inicial ao clicar no botão "Set Prompt"
|
| 236 |
-
set_prompt_btn.click(fn=lambda prompt: prompt, inputs=prompt_input, outputs=initial_prompt)
|
| 237 |
-
|
| 238 |
with gr.Tab("Step 4 - Initialize QA chain"):
|
| 239 |
with gr.Row():
|
| 240 |
llm_btn = gr.Radio(list_llm_simple,
|
|
@@ -295,7 +254,7 @@ def demo():
|
|
| 295 |
db_btn.click(initialize_database,
|
| 296 |
inputs=[document, slider_chunk_size, slider_chunk_overlap, db_type_radio],
|
| 297 |
outputs=[vector_db, collection_name, db_progress])
|
| 298 |
-
set_prompt_btn.click(lambda prompt: prompt,
|
| 299 |
inputs=prompt_input,
|
| 300 |
outputs=initial_prompt)
|
| 301 |
qachain_btn.click(initialize_LLM,
|
|
|
|
| 97 |
progress(0.9, desc="Done!")
|
| 98 |
return qa_chain
|
| 99 |
|
| 100 |
+
def initialize_llm_no_doc(llm_model, temperature, max_tokens, top_k, initial_prompt, progress=gr.Progress()):
|
| 101 |
+
progress(0.1, desc="Initializing HF tokenizer...")
|
| 102 |
+
progress(0.5, desc="Initializing HF Hub...")
|
| 103 |
+
llm = HuggingFaceEndpoint(
|
| 104 |
+
repo_id=llm_model,
|
| 105 |
+
huggingfacehub_api_token=api_token,
|
| 106 |
+
temperature=temperature,
|
| 107 |
+
max_new_tokens=max_tokens,
|
| 108 |
+
top_k=top_k,
|
| 109 |
+
)
|
| 110 |
+
progress(0.75, desc="Defining buffer memory...")
|
| 111 |
+
memory = ConversationBufferMemory(
|
| 112 |
+
memory_key="chat_history",
|
| 113 |
+
output_key='answer',
|
| 114 |
+
return_messages=True
|
| 115 |
+
)
|
| 116 |
+
conversation_chain = ConversationChain(llm=llm, memory=memory, verbose=False)
|
| 117 |
+
conversation_chain({"question": initial_prompt})
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
progress(0.9, desc="Done!")
|
| 119 |
+
return conversation_chain
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
|
| 121 |
def format_chat_history(message, chat_history):
|
| 122 |
formatted_chat_history = []
|
|
|
|
| 142 |
new_history = history + [(message, response_answer)]
|
| 143 |
return qa_chain, gr.update(value=""), new_history, response_source1, response_source1_page, response_source2, response_source2_page, response_source3, response_source3_page
|
| 144 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 145 |
def conversation_no_doc(llm, message, history):
|
| 146 |
formatted_chat_history = format_chat_history(message, history)
|
| 147 |
response = llm({"question": message, "chat_history": formatted_chat_history})
|
|
|
|
| 187 |
db_progress = gr.Textbox(label="Vector database initialization", value="None")
|
| 188 |
with gr.Row():
|
| 189 |
db_btn = gr.Button("Generate vector database")
|
| 190 |
+
|
|
|
|
|
|
|
|
|
|
| 191 |
with gr.Tab("Step 3 - Set Initial Prompt"):
|
| 192 |
with gr.Row():
|
| 193 |
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.")
|
| 194 |
with gr.Row():
|
| 195 |
set_prompt_btn = gr.Button("Set Prompt")
|
| 196 |
|
|
|
|
|
|
|
|
|
|
| 197 |
with gr.Tab("Step 4 - Initialize QA chain"):
|
| 198 |
with gr.Row():
|
| 199 |
llm_btn = gr.Radio(list_llm_simple,
|
|
|
|
| 254 |
db_btn.click(initialize_database,
|
| 255 |
inputs=[document, slider_chunk_size, slider_chunk_overlap, db_type_radio],
|
| 256 |
outputs=[vector_db, collection_name, db_progress])
|
| 257 |
+
set_prompt_btn.click(lambda prompt: gr.update(value=prompt),
|
| 258 |
inputs=prompt_input,
|
| 259 |
outputs=initial_prompt)
|
| 260 |
qachain_btn.click(initialize_LLM,
|