Simple Anthropic

app.py

@@ -1,193 +1,51 @@
 import gradio as gr
 import os
-import time
-import threading
 from langchain.document_loaders import OnlinePDFLoader
 from langchain.text_splitter import CharacterTextSplitter
-from langchain.llms import OpenAI
-from langchain.embeddings import OpenAIEmbeddings
-from langchain.vectorstores import Chroma
-from langchain.chains import ConversationalRetrievalChain
-from langchain.chat_models import ChatOpenAI
-from langchain.document_loaders import WebBaseLoader
-from langchain.chains.summarize import load_summarize_chain
-from langchain.chains.llm import LLMChain
-from langchain.prompts import PromptTemplate
-from langchain.chains.combine_documents.stuff import StuffDocumentsChain
+from anthropic import LanguageModel
 
-os.environ['OPENAI_API_KEY'] = os.getenv("Your_API_Key")
+# Set API keys from environment variables
+os.environ['ANTHROPIC_API_KEY'] = os.getenv("Your_Anthropic_API_Key")
 
-# Global variable for tracking last interaction time
-last_interaction_time = 0
+# Initialize the Anthropic model
+anthropic_model = LanguageModel(api_key=os.environ['ANTHROPIC_API_KEY'], model="some_model")
 
-def loading_pdf():
-    return "Working on the upload. Also, pondering the usefulness of sporks..."
+pdf_content = ""
 
-# Inside Chroma mod
-def summary(self):
-    num_documents = len(self.documents)
-    avg_doc_length = sum(len(doc) for doc in self.documents) / num_documents
-    return f"Number of documents: {num_documents}, Average document length: {avg_doc_length}"
-
-# Gradio state
-summary_state = gr.State(initial_value="")
-
-# Initialize loader and load documents
-def load_documents(pdf_doc):
-    loader = OnlinePDFLoader(pdf_doc.name)
-    return loader.load()
-
-# Generate summary using StuffDocumentsChain
-def generate_summary(documents):
-    prompt_template = """Write a concise summary of the following:
-    "{text}"
-    CONCISE SUMMARY:"""
-    prompt = PromptTemplate.from_template(prompt_template)
-    llm = ChatOpenAI(temperature=0, model_name="gpt-3.5-turbo-16k")
-    llm_chain = LLMChain(llm=llm, prompt=prompt)
-    stuff_chain = StuffDocumentsChain(
-        llm_chain=llm_chain, document_variable_name="text"
-    )
-    return stuff_chain.run(documents)
-
-# Setup Chroma, embeddings, and retrieval
-def setup_retrieval(documents):
-    embeddings = OpenAIEmbeddings()
-    db = Chroma.from_documents(documents, embeddings)
-    retriever = db.as_retriever()
-    qa = ConversationalRetrievalChain.from_llm(
-        llm=OpenAI(temperature=0.2, model_name="gpt-3.5-turbo-16k", max_tokens=-1, n=2),
-        retriever=retriever,
-        return_source_documents=False
-    )
-    return db, qa
-
-# Main function to handle PDF changes
-def pdf_changes(pdf_doc):
+def load_pdf(pdf_doc):
+    global pdf_content
     try:
-        documents = load_documents(pdf_doc)
-        global full_summary
-        full_summary = generate_summary(documents)
-        summary_state.value = full_summary  # Update the state variable
-        # ... (rest of your code)
-        return f"Ready. Full Summary loaded."
-    except Exception as e:
-        return f"Error processing PDF: {str(e)}"
+        if pdf_doc is None:
+            return "No PDF uploaded."
 
+        # Load and split PDF content
+        loader = OnlinePDFLoader(pdf_doc.name)
+        documents = loader.load()
+        text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
+        pdf_content = ' '.join(text_splitter.split_documents(documents))
 
-def clear_data():
-    global qa, db
-    qa = None
-    db = None
-    return "Data cleared"
+        return "PDF Loaded Successfully."
 
-def add_text(history, text):
-    global last_interaction_time
-    last_interaction_time = time.time()
-    history = history + [(text, None)]
-    return history, ""
-
-def bot(history):
-    global full_summary  
-    if 'summary' in history[-1][0].lower():  # Check if the last question asks for a summary
-        response = full_summary
-        return full_summary
-    else:
-        response = infer(history[-1][0], history)
-
-    sentences = '  \n'.join(response.split('. '))
-    formatted_response = f"**Bot:**\n\n{sentences}"
-    history[-1][1] = formatted_response
-    return history
-
-
-def infer(question, history):
-    try:
-        res = []
-        for human, ai in history[:-1]:
-            pair = (human, ai)
-            res.append(pair)
-    
-        chat_history = res
-        query = question
-        result = qa({"question": query, "chat_history": chat_history, "system": "This is a world-class summarizing AI, be helpful."})
-        return result["answer"]
     except Exception as e:
-        return f"Error querying chatbot: {str(e)}"
+        return f"Error processing PDF: {e}"
 
-def auto_clear_data():
-    global qa, db, last_interaction_time
-    if time.time() - last_interaction_time > 1000:
-        qa = None
-        db = None
-        print("Data cleared successfully.")  # Logging
+def chat_with_pdf(question):
+    context = [{"role": "system", "content": pdf_content}]
+    response = anthropic_model.query(question, context=context)
+    return response['answer']
 
-def periodic_clear():
-    while True:
-        auto_clear_data()
-        time.sleep(1000)
-
-threading.Thread(target=periodic_clear).start()
-
-css = """
-#col-container {max-width: 700px; margin-left: auto; margin-right: auto;}
-"""
-
-title = """
-<div style="text-align: center;max-width: 700px;">
-    <h1>CauseWriter Chat with PDF • OpenAI</h1>
-    <p style="text-align: center;">Upload a .PDF from your computer, click the "Load PDF to LangChain" button, <br />
-    when everything is ready, you can start asking questions about the pdf. Limit ~11k words. <br />
-    This version is set to erase chat history automatically after page timeout and uses OpenAI.</p>
-</div>
-"""
-# Global variable for tracking last interaction time
-last_interaction_time = 0
-full_summary = ""  # Added global full_summary
-
-def update_summary_box():
-    global full_summary
-    return {"summary_box": full_summary}
-
-with gr.Blocks(css=css) as demo:
-    with gr.Column(elem_id="col-container"):
-        gr.HTML(title)
-        
-        with gr.Column():
-            pdf_doc = gr.File(label="Load a pdf", file_types=['.pdf'], type="file")
-            with gr.Row():
-                langchain_status = gr.Textbox(label="Status", placeholder="", interactive=False)
-                load_pdf = gr.Button("Convert PDF to Magic AI language")
-                clear_btn = gr.Button("Clear Data")
-            
-            # New Textbox to display summary
-            summary_box = gr.Textbox(
-              label="Document Summary",
-              placeholder="Summary will appear here.",
-              interactive=False,
-              rows=5,
-              elem_id="summary_box",
-              state=summary_state  # Bind the state here
-            )
-
-        
-        chatbot = gr.Chatbot([], elem_id="chatbot").style(height=450)
-        question = gr.Textbox(label="Question", placeholder="Type your question and hit Enter")
-        submit_btn = gr.Button("Send Message")
-
-        load_pdf.click(loading_pdf, None, langchain_status)
-        load_pdf.click(pdf_changes, inputs=[pdf_doc], outputs=[langchain_status], queue=False).then(
-          update_summary_box
-      )
-
-
-    # Then update the summary_box
-    clear_btn.click(clear_data, outputs=[langchain_status], queue=False)
-    question.submit(add_text, [chatbot, question], [chatbot, question]).then(
-        bot, chatbot, chatbot
-    )
-    submit_btn.click(add_text, [chatbot, question], [chatbot, question]).then(
-        bot, chatbot, chatbot
-    )
-
-demo.launch()
+# Define Gradio UI
+def gradio_interface(pdf_doc, question):
+    if not pdf_content:
+        return load_pdf(pdf_doc)
+    else:
+        return chat_with_pdf(question)
+
+gr.Interface(fn=gradio_interface,
+             inputs=[gr.File(label="Load a pdf", file_types=['.pdf'], type="file"),
+                     gr.Textbox(label="Ask a question about the PDF")],
+             outputs="text",
+             live=True,
+             title="Chat with PDF content using Anthropic",
+             description="Upload a .PDF and interactively chat about its content."
+            ).launch()