Update app.py

app.py

@@ -1,8 +1,8 @@
 import gradio as gr
-from langchain.vectorstores import Chroma
+from langchain_community.vectorstores import Chroma
 from langchain_community.document_loaders import PyPDFLoader
 from langchain_community.embeddings import HuggingFaceEmbeddings
-from transformers import LayoutLMv3Processor, AutoModelForSeq2SeqLM
+from transformers import LayoutLMv3Processor, AutoModelForTokenClassification
 from langchain.chains import RetrievalQA
 from langchain.prompts import PromptTemplate
 from pdf2image import convert_from_path
@@ -10,16 +10,20 @@ import os
 
 class LayoutLMv3OCR:
     def __init__(self):
+        # Lade den LayoutLMv3-Prozessor und das Modell für Token-Klassifikation
         self.processor = LayoutLMv3Processor.from_pretrained("microsoft/layoutlmv3-base")
-        self.model = AutoModelForSeq2SeqLM.from_pretrained("microsoft/layoutlmv3-base")
+        self.model = AutoModelForTokenClassification.from_pretrained("microsoft/layoutlmv3-base")
 
     def extract_text(self, pdf_path):
         images = convert_from_path(pdf_path)
         text_pages = []
         for image in images:
+            # Verarbeite die Bilddaten mit LayoutLMv3
             inputs = self.processor(images=image, return_tensors="pt")
-            outputs = self.model.generate(**inputs)
-            text = self.processor.batch_decode(outputs, skip_special_tokens=True)[0]
+            # Führe Vorhersagen durch
+            outputs = self.model(**inputs)
+            # Extrahiere den Text aus den Vorhersagen (falls dies vorgesehen ist)
+            text = self.processor.batch_decode(outputs.logits, skip_special_tokens=True)[0]
             text_pages.append(text)
         return text_pages
 
@@ -48,25 +52,4 @@ def chatbot_response(pdf, question):
     with open(pdf_path, "wb") as f:
         f.write(pdf.read())
     
-    extracted_text = ocr_tool.extract_text(pdf_path)
-    answer = process_pdf_and_query(pdf_path, question)
-    
-    # Lösche die gespeicherte PDF-Datei nach der Verarbeitung
-    os.remove(pdf_path)
-    
-    return answer
-
-pdf_input = gr.File(label="PDF-Datei hochladen")
-question_input = gr.Textbox(label="Frage eingeben")
-response_output = gr.Textbox(label="Antwort")
-
-interface = gr.Interface(
-    fn=chatbot_response,
-    inputs=[pdf_input, question_input],
-    outputs=response_output,
-    title="RAG Chatbot mit PDF-Unterstützung",
-    description="Lade eine PDF-Datei hoch und stelle Fragen zu ihrem Inhalt."
-)
-
-if __name__ == "__main__":
-    interface.launch(share=True)
+    extracted_text = ocr_tool.extr