Update app.py

app.py

@@ -2,25 +2,25 @@ import openai
 import os
 from paperqa import Docs
 import gradio as gr
-from dotenv import load_dotenv
-
-# Load environment variables from .env file
-load_dotenv()
-
-# Get the OpenAI API key from the environment variable
-openai_api_key = os.getenv('OPENAI_API_KEY')
+from langchain.document_loaders import PyPDFLoader
+from langchain.vectorstores import Chroma
+from langchain.embeddings.openai import OpenAIEmbeddings
+from langchain.document_loaders import UnstructuredPDFLoader
+from langchain.llms import OpenAI
+from langchain.chains.question_answering import load_qa_chain
+from langchain.chat_models import ChatOpenAI
 
 css_style = """
 .gradio-container {
     font-family: "IBM Plex Mono";
 }
-
 .answerText p {
     font-size: 24px !important;
     color: #8dbcfe !important;
 }
 """
 
+
 def run(uploaded_files):
     all_files = []
     if uploaded_files is None:
@@ -31,53 +31,48 @@ def run(uploaded_files):
     print(all_files)
     return all_files
 
-def createAnswer(files, designation):
-    try:
-        os.environ['OPENAI_API_KEY'] = openai_api_key
-        docs = Docs(llm='gpt-3.5-turbo')
-        for d in files:
-            docs.add(d.name)
-        answer = docs.query(
-            f"Who is the best candidate to hire for {designation}. Provide a list with the candidate name. If you don't know, simply say None of the candidates are suited for the Job role.")
-        return answer.answer
-    except Exception as e:
-        return f"An error occurred: {str(e)}"
+
+def createAnswer(files, designation, openaikey):
+    os.environ['OPENAI_API_KEY'] = openaikey.strip()
+    docs = Docs(llm='gpt-3.5-turbo')
+    for d in files:
+        docs.add(d.name)
+    answer = docs.query(
+        f"Who is the best canidate to hire for {designation}. Provide a list with the candidate name. If you don't know, simply say None of the canidates are suited for the Job role.")
+    print(answer.formatted_answer)
+    print(type(answer))
+    return answer.answer
+
 
 with gr.Blocks(css=css_style) as demo:
     gr.Markdown(f"""
     # HR-GPT - Filter & Find The Best Candidate for the Job using AI
-
     *By Amin Memon ([@AminMemon](https://twitter.com/AminMemon))*
-
     This tool will enable asking questions of your uploaded text, PDF documents,.
     It uses OpenAI's ChatGPT model & OpenAI Embeddings and thus you must enter your API key below.
-
     This tool is under active development and currently uses many tokens - up to 10,000
     for a single query. That is $0.10-0.20 per query, so please be careful!
     Porting it to Llama.cpp soon for saved cost.
-
-    1. Upload your Resumes (Try a few resumes/cv to try < 5)
-    2. Provide Designation for which you are hiring
+    1. Enter API Key ([What is that?](https://platform.openai.com/account/api-keys))
+    2. Upload your Resumes (Try a few resumes/cv to try < 5)
+    3. Provide Designation for which you are hiring
     """)
 
-    position = gr.Text(label='Position/Designation for which you are hiring for', value="")
+    openaikey = gr.Text(
+        label='Your OpenAI Api Key', value="")
+    position = gr.Text(
+        label='Position/Designation for which you are hiring for', value="")
 
     with gr.Tab('File Upload'):
-        uploaded_files = gr.File(label="Resume Upload - ONLY PDF. (Doc File Support Coming Soon)", file_count="multiple", show_progress=True)
+        uploaded_files = gr.File(
+            label="Resume Upload - ONLY PDF. (Doc File Support Coming Soon)", file_count="multiple", show_progress=True)
 
-    uploaded_files.change(fn=run, inputs=[uploaded_files], outputs=[uploaded_files])
+    uploaded_files.change(
+        fn=run, inputs=[uploaded_files], outputs=[uploaded_files])
     ask = gr.Button("Find Top Candidate")
-    answer = gr.Markdown(label="Result", elem_classes='answerText')
-    loading_indicator = gr.Markdown(label="Loading", visible=False)
-
-    def on_ask_click():
-        loading_indicator.update(visible=True, value="Processing...")
-
-    ask.click(fn=on_ask_click, inputs=[], outputs=[loading_indicator]).then(
-        fn=createAnswer, inputs=[uploaded_files, position], outputs=[answer]
-    ).then(
-        fn=lambda: gr.Markdown.update(visible=False), inputs=[], outputs=[loading_indicator]
-    )
+    answer = gr.Markdown(label="Result",  elem_classes='answerText')
+    ask.click(fn=createAnswer, inputs=[
+              uploaded_files, position, openaikey], outputs=[answer])
 
 demo.queue(concurrency_count=20)
-demo.launch(show_error=True)
+demo.launch(show_error=True)