Update app.py

app.py

@@ -45,15 +45,26 @@ from llama_index.core.node_parser.relational.base_element import (
 from llama_index.core.schema import BaseNode, TextNode
 
 
+# Obtenha o token da variável de ambiente
 api_token = os.getenv("HF_TOKEN")
-# Implementations
+
+# Verifique se o token foi obtido corretamente
+if api_token is None:
+    raise ValueError("O token de API não foi encontrado. Verifique se a variável de ambiente HF_TOKEN está configurada corretamente.")
+
+# Função para ofuscar o token
+def obscure_token(token, num_visible=4):
+    return '*' * (len(token) - num_visible) + token[-num_visible:]
+
+# Exibir o token de API ofuscado (apenas para debug; remova em produção)
+print(f"Token de API: {obscure_token(api_token)}")
 
 # default_persist_directory = './chroma_HF/'
-list_llm = ["mistralai/Mistral-7B-Instruct-v0.3", "mistralai/Mixtral-8x7B-Instruct-v0.1", "mistralai/Mistral-7B-Instruct-v0.1", \
-    "google/gemma-7b-it","google/gemma-2b-it", \
-    "HuggingFaceH4/zephyr-7b-beta", "HuggingFaceH4/zephyr-7b-gemma-v0.1", \
-    "meta-llama/Llama-2-7b-chat-hf", "microsoft/phi-2", \
-    "TinyLlama/TinyLlama-1.1B-Chat-v1.0", "mosaicml/mpt-7b-instruct", "tiiuae/falcon-7b-instruct", \
+list_llm = ["mistralai/Mistral-7B-Instruct-v0.3", "mistralai/Mixtral-8x7B-Instruct-v0.1", "mistralai/Mistral-7B-Instruct-v0.1", 
+    "google/gemma-7b-it","google/gemma-2b-it", 
+    "HuggingFaceH4/zephyr-7b-beta", "HuggingFaceH4/zephyr-7b-gemma-v0.1", 
+    "meta-llama/Llama-2-7b-chat-hf", "microsoft/phi-2", 
+    "TinyLlama/TinyLlama-1.1B-Chat-v1.0", "mosaicml/mpt-7b-instruct", "tiiuae/falcon-7b-instruct", 
     "google/flan-t5-xxl"
 ]
 list_llm_simple = [os.path.basename(llm) for llm in list_llm]
@@ -96,7 +107,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
     if llm_model == "mistralai/Mixtral-8x7B-Instruct-v0.3":
         llm = HuggingFaceEndpoint(
             repo_id=llm_model, 
-            huggingfacehub_api_token = api_token,
+            api_key=api_token,
             temperature = temperature,
             max_new_tokens = max_tokens,
             top_k = top_k,
@@ -106,7 +117,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
         raise gr.Error("LLM model is too large to be loaded automatically on free inference endpoint")
         llm = HuggingFaceEndpoint(
             repo_id=llm_model, 
-            huggingfacehub_api_token = api_token,
+            api_key=api_token,
             temperature = temperature,
             max_new_tokens = max_tokens,
             top_k = top_k,
@@ -114,7 +125,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
     elif llm_model == "microsoft/phi-2":
         llm = HuggingFaceEndpoint(
             repo_id=llm_model, 
-            huggingfacehub_api_token = api_token,
+            api_key=api_token,
             temperature = temperature,
             max_new_tokens = max_tokens,
             top_k = top_k,
@@ -124,7 +135,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
     elif llm_model == "TinyLlama/TinyLlama-1.1B-Chat-v1.0":
         llm = HuggingFaceEndpoint(
             repo_id=llm_model,
-            huggingfacehub_api_token = api_token,
+            api_key=api_token,
             temperature = temperature,
             max_new_tokens = 250,
             top_k = top_k,
@@ -133,7 +144,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
         raise gr.Error("Llama-2-7b-chat-hf model requires a Pro subscription...")
         llm = HuggingFaceEndpoint(
             repo_id=llm_model, 
-            huggingfacehub_api_token = api_token,
+            api_key=api_token,
             temperature = temperature,
             max_new_tokens = max_tokens,
             top_k = top_k,
@@ -141,7 +152,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
     else:
         llm = HuggingFaceEndpoint(
             repo_id=llm_model, 
-            huggingfacehub_api_token = api_token,
+            api_key=api_token,
             temperature = temperature,
             max_new_tokens = max_tokens,
             top_k = top_k,
@@ -235,11 +246,11 @@ def upload_file(file_obj):
 
 # Initialize LlamaIndex parsing
 def initialize_llama_index(file_obj):
-    documents = LlamaParse(result_type="markdown",api_key=secret_value_0).load_data(file_obj.name)
-    node_parser = MarkdownElementNodeParser(llm = None, num_workers=8)
+    documents = LlamaParse(result_type="markdown", api_key=api_token).load_data(file_obj[0].name)
+    node_parser = MarkdownElementNodeParser(llm=None, num_workers=8)
     nodes = node_parser.get_nodes_from_documents(documents)
     base_nodes, objects = node_parser.get_nodes_and_objects(nodes)
-    index_with_obj = VectorStoreIndex(nodes=base_nodes+objects)
+    index_with_obj = VectorStoreIndex(nodes=base_nodes + objects)
     index_ret = index_with_obj.as_retriever(top_k=15)
     recursive_query_engine = RetrieverQueryEngine.from_args(index_ret, node_postprocessors=[FlagEmbeddingReranker(
         top_n=5,
@@ -268,12 +279,12 @@ def demo():
         
         with gr.Tab("Step 2 - Process document"):
             with gr.Row():
-                db_btn = gr.Radio(["ChromaDB"], label="Vector database type", value = "ChromaDB", type="index", info="Choose your vector database")
+                db_btn = gr.Radio(["ChromaDB"], label="Vector database type", value="ChromaDB", type="index", info="Choose your vector database")
             with gr.Accordion("Advanced options - Document text splitter", open=False):
                 with gr.Row():
-                    slider_chunk_size = gr.Slider(minimum = 100, maximum = 1000, value=600, step=20, label="Chunk size", info="Chunk size", interactive=True)
+                    slider_chunk_size = gr.Slider(minimum=100, maximum=1000, value=600, step=20, label="Chunk size", info="Chunk size", interactive=True)
                 with gr.Row():
-                    slider_chunk_overlap = gr.Slider(minimum = 10, maximum = 200, value=40, step=10, label="Chunk overlap", info="Chunk overlap", interactive=True)
+                    slider_chunk_overlap = gr.Slider(minimum=10, maximum=200, value=40, step=10, label="Chunk overlap", info="Chunk overlap", interactive=True)
             with gr.Row():
                 db_progress = gr.Textbox(label="Vector database initialization", value="None")
             with gr.Row():
@@ -281,17 +292,17 @@ def demo():
         
         with gr.Tab("Step 3 - Initialize QA chain"):
             with gr.Row():
-                llm_btn = gr.Radio(list_llm_simple, \
-                    label="LLM models", value = list_llm_simple[0], type="index", info="Choose your LLM model")
+                llm_btn = gr.Radio(list_llm_simple, 
+                    label="LLM models", value=list_llm_simple[0], type="index", info="Choose your LLM model")
             with gr.Accordion("Advanced options - LLM model", open=False):
                 with gr.Row():
-                    slider_temperature = gr.Slider(minimum = 0.01, maximum = 1.0, value=0.7, step=0.1, label="Temperature", info="Model temperature", interactive=True)
+                    slider_temperature = gr.Slider(minimum=0.01, maximum=1.0, value=0.7, step=0.1, label="Temperature", info="Model temperature", interactive=True)
                 with gr.Row():
-                    slider_maxtokens = gr.Slider(minimum = 224, maximum = 4096, value=1024, step=32, label="Max Tokens", info="Model max tokens", interactive=True)
+                    slider_maxtokens = gr.Slider(minimum=224, maximum=4096, value=1024, step=32, label="Max Tokens", info="Model max tokens", interactive=True)
                 with gr.Row():
-                    slider_topk = gr.Slider(minimum = 1, maximum = 10, value=3, step=1, label="top-k samples", info="Model top-k samples", interactive=True)
+                    slider_topk = gr.Slider(minimum=1, maximum=10, value=3, step=1, label="top-k samples", info="Model top-k samples", interactive=True)
             with gr.Row():
-                llm_progress = gr.Textbox(value="None",label="QA chain initialization")
+                llm_progress = gr.Textbox(value="None", label="QA chain initialization")
             with gr.Row():
                 qachain_btn = gr.Button("Initialize Question Answering chain")
 
@@ -320,31 +331,31 @@ def demo():
                 clear_btn = gr.ClearButton([msg, chatbot], value="Clear conversation")
             
         # Preprocessing events
-        db_btn.click(initialize_database, \
-            inputs=[document, slider_chunk_size, slider_chunk_overlap], \
+        db_btn.click(initialize_database, 
+            inputs=[document, slider_chunk_size, slider_chunk_overlap], 
             outputs=[vector_db, collection_name, db_progress])
-        qachain_btn.click(initialize_LLM, \
-            inputs=[llm_btn, slider_temperature, slider_maxtokens, slider_topk, vector_db], \
-            outputs=[qa_chain, llm_progress]).then(lambda:[None,"",0,"",0,"",0], \
-            inputs=None, \
-            outputs=[chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], \
+        qachain_btn.click(initialize_LLM, 
+            inputs=[llm_btn, slider_temperature, slider_maxtokens, slider_topk, vector_db], 
+            outputs=[qa_chain, llm_progress]).then(lambda:[None,"",0,"",0,"",0], 
+            inputs=None, 
+            outputs=[chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], 
             queue=False)
-        llama_index_btn.click(initialize_llama_index, \
-            inputs=[document], \
+        llama_index_btn.click(initialize_llama_index, 
+            inputs=[document], 
             outputs=[llama_index_engine, llama_index_progress])
 
         # Chatbot events
-        msg.submit(conversation, \
-            inputs=[qa_chain, msg, chatbot], \
-            outputs=[qa_chain, msg, chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], \
+        msg.submit(conversation, 
+            inputs=[qa_chain, msg, chatbot], 
+            outputs=[qa_chain, msg, chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], 
             queue=False)
-        submit_btn.click(conversation, \
-            inputs=[qa_chain, msg, chatbot], \
-            outputs=[qa_chain, msg, chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], \
+        submit_btn.click(conversation, 
+            inputs=[qa_chain, msg, chatbot], 
+            outputs=[qa_chain, msg, chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], 
             queue=False)
-        clear_btn.click(lambda:[None,"",0,"",0,"",0], \
-            inputs=None, \
-            outputs=[chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], \
+        clear_btn.click(lambda:[None,"",0,"",0,"",0], 
+            inputs=None, 
+            outputs=[chatbot, doc_source1, source1_page, doc_source2, source2_page, doc_source3, source3_page], 
             queue=False)
     demo.queue().launch(debug=True)