Update app.py

app.py

@@ -1,64 +1,75 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from langchain import LLMChain, ConversationChain
+from langchain.memory import ConversationBufferMemory
+from langchain.llms import HuggingFacePipeline
+from transformers import pipeline
+from langchain.document_loaders import PyPDFLoader, CSVLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.chains import ConversationalRetrievalChain
+from datasets import load_dataset
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Modelo Hugging Face
+pipe = pipeline("text-generation", model="mistralai/Mistral-7B-Instruct", max_new_tokens=200, temperature=0.7)
+llm = HuggingFacePipeline(pipeline=pipe)
 
+memory = ConversationBufferMemory()
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# PDF loader e vetores
+def init_pdf(path):
+    loader = PyPDFLoader(path)
+    docs = loader.load_and_split()
+    embeddings = HuggingFaceEmbeddings()
+    vectorstore = FAISS.from_documents(docs, embeddings)
+    return vectorstore
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# CSV loader e vetores (ou pandas agent)
+def init_csv(path):
+    loader = CSVLoader(path)
+    docs = loader.load_and_split()
+    embeddings = HuggingFaceEmbeddings()
+    vectorstore = FAISS.from_documents(docs, embeddings)
+    return vectorstore
 
-    messages.append({"role": "user", "content": message})
+# Fluxo de retriever
+def make_rag_chain(vectorstore):
+    retriever = vectorstore.as_retriever()
+    return ConversationalRetrievalChain(llm=llm, retriever=retriever, memory=memory)
 
-    response = ""
+pdf_chain = None
+csv_chain = None
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+def upload_pdf(uploaded_pdf):
+    global pdf_chain
+    vectorstore = init_pdf(uploaded_pdf.name)
+    pdf_chain = make_rag_chain(vectorstore)
+    return "PDF carregado com sucesso."
 
-        response += token
-        yield response
+def upload_csv(uploaded_csv):
+    global csv_chain
+    vectorstore = init_csv(uploaded_csv.name)
+    csv_chain = make_rag_chain(vectorstore)
+    return "CSV carregado com sucesso."
 
+def chat(message):
+    if pdf_chain:
+        resp = pdf_chain.run({"question":message})
+    elif csv_chain:
+        resp = csv_chain.run({"question":message})
+    else:
+        chain = ConversationChain(llm=llm, memory=memory)
+        resp = chain.predict(input=message)
+    return resp
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+with gr.Blocks() as demo:
+    gr.Markdown("## Chatbot com memória — PDF, CSV ou conversa simples")
+    pdf_file = gr.File(label="Upload PDF", file_types=['.pdf'])
+    csv_file = gr.File(label="Upload CSV", file_types=['.csv'])
+    pdf_file.upload(upload_pdf, inputs=pdf_file, outputs="text")
+    csv_file.upload(upload_csv, inputs=csv_file, outputs="text")
+    chatbot = gr.Chatbot()
+    msg = gr.Textbox(placeholder="Digite aqui sua pergunta…")
+    msg.submit(lambda m: (msg.clear(), chatbot.append((m, chat(m)))), outputs=chatbot)
 
-
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()