app.py

app.py

@@ -0,0 +1,131 @@
+import os
+import shutil
+import gradio as gr
+from typing import List
+from llama_index.core import SimpleDirectoryReader, StorageContext, VectorStoreIndex
+from llama_index.core.node_parser import SentenceSplitter
+from llama_index.embeddings.huggingface import HuggingFaceEmbedding
+from llama_index.vector_stores.chroma import ChromaVectorStore
+from llama_index.llms.groq import Groq
+from llama_index.core.memory import ChatSummaryMemoryBuffer
+import chromadb
+from tempfile import TemporaryDirectory
+from PyPDF2 import PdfReader
+
+# Wrapper de embedding compatível com ChromaDB
+class ChromaEmbeddingWrapper:
+    def __init__(self, model_name: str):
+        self.model = HuggingFaceEmbedding(model_name=model_name)
+
+    def __call__(self, input: List[str]) -> List[List[float]]:
+        return self.model.embed_documents(input)
+
+# Inicializa modelos de embedding
+embed_model = HuggingFaceEmbedding(model_name='intfloat/multilingual-e5-large')
+embed_model_chroma = ChromaEmbeddingWrapper(model_name='intfloat/multilingual-e5-large')
+
+# Inicializa ChromaDB
+chroma_client = chromadb.PersistentClient(path='./chroma_db')
+collection_name = 'documentos_serenatto'
+chroma_collection = chroma_client.get_or_create_collection(
+    name=collection_name,
+    embedding_function=embed_model_chroma
+)
+
+vector_store = ChromaVectorStore(chroma_collection=chroma_collection)
+storage_context = StorageContext.from_defaults(vector_store=vector_store)
+
+# Inicializa LLM da Groq
+Groq_api = os.environ.get("GROQ_API_KEY")
+llms = Groq(model='llama3-70b-8192', api_key='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')
+
+# Estados globais
+document_index = None
+chat_engine = None
+
+# Processamento do PDF
+
+def process_pdf(file):
+    global document_index, chat_engine
+
+    try:
+        with TemporaryDirectory() as tmpdir:
+            pdf_path = os.path.join(tmpdir, "upload.pdf")
+            shutil.copy(file.name, pdf_path)
+
+            text = ""
+            reader = PdfReader(pdf_path)
+            for page in reader.pages:
+                text += page.extract_text() or ""
+
+            with open(os.path.join(tmpdir, "temp.txt"), "w", encoding="utf-8") as f:
+                f.write(text)
+
+            documentos = SimpleDirectoryReader(input_dir=tmpdir)
+            docs = documentos.load_data()
+
+            node_parser = SentenceSplitter(chunk_size=1200)
+            nodes = node_parser.get_nodes_from_documents(docs, show_progress=True)
+
+            document_index = VectorStoreIndex(nodes, storage_context=storage_context, embed_model=embed_model)
+
+            memory = ChatSummaryMemoryBuffer(llm=llms, token_limit=256)
+
+            chat_engine = document_index.as_chat_engine(
+                chat_mode='context',
+                llm=llms,
+                memory=memory,
+                system_prompt='''Voce é especialista em cafes da loja Serenatto, uma loja online que vende graos de cafe
+                torrados, sua funçao é tirar duvidas de forma simpatica e natural sobre os graos disponiveis.'''
+            )
+
+            return "PDF carregado com sucesso! Agora você pode conversar com o bot."
+
+    except Exception as e:
+        return f"Erro ao processar PDF: {e}"
+
+# Chat com histórico estilo "messages"
+def converse_com_bot(message, chat_history):
+    global chat_engine
+
+    if chat_engine is None:
+        return "Por favor, envie um PDF primeiro.", chat_history
+
+    response = chat_engine.chat(message)
+
+    if chat_history is None:
+        chat_history = []
+
+    chat_history.append({"role": "user", "content": message})
+    chat_history.append({"role": "assistant", "content": response.response})
+
+    return "", chat_history
+
+# Resetar conversa
+def resetar_chat():
+    global chat_engine
+    if chat_engine:
+        chat_engine.reset()
+    return []
+
+# Interface Gradio com upload de PDF
+with gr.Blocks() as app:
+    gr.Markdown("# Chatbot da Serenatto - Especialista em Cafés")
+
+    with gr.Row():
+        upload = gr.File(label="📄 Envie seu PDF")
+        upload_button = gr.Button("Carregar PDF")
+
+    output_status = gr.Textbox(label="Status", interactive=False)
+
+    chatbot = gr.Chatbot(label="Conversa", type="messages")
+    msg = gr.Textbox(label='Digite a sua mensagem')
+    limpar = gr.Button('Limpar')
+
+    upload_button.click(process_pdf, inputs=upload, outputs=output_status).then(
+        resetar_chat, None, chatbot
+    )
+    msg.submit(converse_com_bot, [msg, chatbot], [msg, chatbot])
+    limpar.click(resetar_chat, None, chatbot, queue=False)
+
+    app.launch(debug=True)