Update app.py

app.py

@@ -104,32 +104,32 @@ def submit_message(prompt, prompt_template, temperature, max_tokens, context_len
     #vectordb = Chroma.from_documents(split_pages, embeddings, persist_directory=persist_directory)
     #vectordb.persist()
 
-    path = './files'
-    pdf_files = glob.glob(os.path.join(path, "*.pdf"))
+    #path = './files'
+    #pdf_files = glob.glob(os.path.join(path, "*.pdf"))
     
-    merger = PdfWriter()
+    #merger = PdfWriter()
     
     # add all file in the list to the merger object
-    for pdf in pdf_files:
-      merger.append(pdf)
-    merger.write("merged-pdf.pdf")
-    merger.close()
+    #for pdf in pdf_files:
+    #  merger.append(pdf)
+    #merger.write("merged-pdf.pdf")
+    #merger.close()
     
-    reader = PdfReader("merged-pdf.pdf")
-    raw_text = ''
-    for i, page in enumerate(reader.pages):
-      text = page.extract_text()
-      if text:
-          raw_text += text
-    text_splitter = CharacterTextSplitter(        
-      separator = "\n",
-      chunk_size = 1000,
-      chunk_overlap  = 200,
-      length_function = len,
-    )
-    texts = text_splitter.split_text(raw_text)
-    len(texts)
-    embeddings = OpenAIEmbeddings()
+    #reader = PdfReader("merged-pdf.pdf")
+    #raw_text = ''
+    #for i, page in enumerate(reader.pages):
+    #  text = page.extract_text()
+    #  if text:
+     #     raw_text += text
+    #text_splitter = CharacterTextSplitter(        
+    #  separator = "\n",
+    #  chunk_size = 1000,
+    #  chunk_overlap  = 200,
+    #  length_function = len,
+    #)
+    #texts = text_splitter.split_text(raw_text)
+    #len(texts)
+    #embeddings = OpenAIEmbeddings()
         
     
     history = state['messages']
@@ -168,10 +168,13 @@ def submit_message(prompt, prompt_template, temperature, max_tokens, context_len
         #with open("foo.pkl", 'rb') as f:
         #    new_docsearch = pickle.load(f)
         
-        docsearch = FAISS.from_texts(texts, embeddings)
+        #docsearch = FAISS.from_texts(texts, embeddings)
+        with open("embeddings.pkl", 'rb') as f: 
+            new_docsearch = pickle.load(f)
+            
         #query = str(system_prompt + history[-context_length*2:] +  [prompt_msg])
         query = str(system_prompt + history +  [prompt_msg])
-        docs = docsearch.similarity_search(query)
+        docs = new_docsearch.similarity_search(query)
         #print(docs[0].page_content)
 
         chain = load_qa_chain(ChatOpenAI(temperature=temperature, max_tokens=max_tokens, model_name="gpt-3.5-turbo"), chain_type="stuff")
@@ -229,11 +232,11 @@ with gr.Blocks(css=css) as demo:
 
 
     with gr.Column(elem_id="col-container"):
-        with open("embeddings.pkl", 'rb') as f: 
-            new_docsearch = pickle.load(f)
+        #with open("embeddings.pkl", 'rb') as f: 
+         #   new_docsearch = pickle.load(f)
             
-        query = str("performance")
-        docs = new_docsearch.similarity_search(query)
+        #query = str("performance")
+        #docs = new_docsearch.similarity_search(query)
            
         gr.Markdown("""# Chat with Needs Assessment Experts (Past and Present)
                     ## Ask questions of experts on needs assessments, get responses from *needs assessment* version of ChatGPT.
@@ -246,8 +249,8 @@ with gr.Blocks(css=css) as demo:
         with gr.Row():
             with gr.Column():
                 chatbot = gr.Chatbot(elem_id="chatbox")
-                input_message = gr.Textbox(show_label=False, placeholder=docs, visible=True).style(container=False)
-                #input_message = gr.Textbox(show_label=False, placeholder="Enter your needs assessment question and press enter", visible=True).style(container=False)
+                #input_message = gr.Textbox(show_label=False, placeholder=docs, visible=True).style(container=False)
+                input_message = gr.Textbox(show_label=False, placeholder="Enter your needs assessment question and press enter", visible=True).style(container=False)
 
                 btn_submit = gr.Button("Submit")
                 total_tokens_str = gr.Markdown(elem_id="total_tokens_str")