Update app.py

app.py

@@ -1,66 +1,15 @@
-import os
 import gradio as gr
-from langchain.document_loaders import PyMuPDFLoader
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.vectorstores import Chroma
-from langchain.embeddings import OpenAIEmbeddings
-from langchain.chat_models import ChatOpenAI
-from langchain.chains import RetrievalQA
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
 
-import os
-os.environ['CURL_CA_BUNDLE'] = ''
-# Initialize conversation history
-conversation_history = ""
+model_name = "distilbert-base-uncased-finetuned-sst-2-english"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
 
-def main(api_key, pdf_path, user_input):
-    global conversation_history  # Declare as global to update it
+def predict_sentiment(text):
+    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512)
+    outputs = model(**inputs)
+    probs = outputs.logits.softmax(dim=1).detach().numpy()[0]
+    return {"Negative": float(probs[0]), "Positive": float(probs[1])}
 
-    os.environ["OPENAI_API_KEY"] = api_key
-
-    persist_directory = "./storage"
-
-    loader = PyMuPDFLoader(pdf_path.name)
-    documents = loader.load()
-
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=512, chunk_overlap=10)
-    texts = text_splitter.split_documents(documents)
-
-    embeddings = OpenAIEmbeddings()
-    vectordb = Chroma.from_documents(documents=texts, 
-                                     embedding=embeddings,
-                                     persist_directory=persist_directory)
-    vectordb.persist()
-
-    retriever = vectordb.as_retriever(search_kwargs={"k": 3})
-    llm = ChatOpenAI(model_name='gpt-4')
-
-    qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever)
-
-    # Update conversation history with the user's latest question
-    conversation_history += f"User: {user_input}\n"
-
-    query = f"{conversation_history}###Prompt {user_input}"
-    try:
-        llm_response = qa(query)
-        response_text = llm_response["result"]
-
-        # Update conversation history with the model's latest answer
-        conversation_history += f"Model: {response_text}\n"
-
-        return conversation_history  # Return the entire conversation history
-    except Exception as err:
-        return f'Exception occurred. Please try again: {str(err)}'
-
-iface = gr.Interface(
-    fn=main, 
-    inputs=[
-        gr.inputs.Textbox(label="OpenAI API Key", type="password"),
-        gr.inputs.File(label="Upload PDF"),
-        gr.inputs.Textbox(label="Enter Query")
-    ], 
-    outputs="text",
-    live=False,
-    show_submit_button=True,
-    description="Enter your OpenAI API Key, upload a PDF, and enter a query to get a response."
-)
-iface.launch(share=True)
+iface = gr.Interface(fn=predict_sentiment, inputs="text", outputs="label")
+iface.launch()