Update app.py

app.py

@@ -1,64 +1,98 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+import PyPDF2
+from langchain.prompts import PromptTemplate
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Initialize the Hugging Face client with gemma2-9b-it
+client = InferenceClient("HuggingFaceH4/gemma2-9b-it")
 
+# Function to read text from a PDF file
+def pdf_to_text(pdf_path):
+    with open(pdf_path, 'rb') as file:
+        pdf_reader = PyPDF2.PdfReader(file)
+        text = ''
+        for page in range(len(pdf_reader.pages)):
+            text += pdf_reader.pages[page].extract_text()
+    return text
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Function to analyze CV content
+def analyze_cv(cv_text):
+    if not cv_text or not isinstance(cv_text, str):
+        raise ValueError("The CV text must be a non-empty string.")
+    prompt_template = PromptTemplate.from_template('''
+    You are an AI designed to extract structured information from unstructured text. Your task is to analyze the content of a candidate's resume or CV and extract the following details:
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    **CV**
+    {cv_text}
 
-    messages.append({"role": "user", "content": message})
+    **Information Extraction and Output Format**
+    For the given resume, extract and present the following details in the specified format:
 
-    response = ""
+    1. Candidate Information
+    - Full Name
+    - Contact Information (Phone, Email, Address, etc.)
+    - Date of Birth (if available)
+    - Habitat (if specified, e.g., location, region, or country of residence)
+    
+    2. Education
+    - Degree Name (e.g., Bachelor's, Master's, Ph.D.)
+    - Field of Study (e.g., Computer Science, Business Administration)
+    - Institution Name
+    - Year(s) of Graduation
+    
+    3. Professional Experience
+    - For each job extract:
+    - Job Title
+    - Company Name
+    - Duration (start and end dates, or years of experience)
+    - Summary of Key Responsibilities and Achievements
+    
+    4. Skills
+    - List of Skills (include technical, soft, and industry-specific skills mentioned in the resume)
+    
+    5. Certifications
+    - Certification Name
+    - Issuing Organization
+    - Year of Issuance
+    
+    6. Language Proficiency
+    - Languages Mentioned (include proficiency levels if specified in the resume)
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    Do not explain, comment or make up any more information that is not relative to the list of Information extraction. Respond in Vietnamese. Let's work this out in a step by step way to ensure the correct answer. [END].
+    ''')
+    prompt = prompt_template.format(cv_text=cv_text)
+    response = client.text_generation(prompt, max_tokens=2048, temperature=0.0)
+    return response
 
-        response += token
-        yield response
+# Chatbot with PDF and CV analysis
+def chatbot_with_pdf(pdf_file, user_message, history, system_message, max_tokens, temperature, top_p):
+    if pdf_file is not None:
+        pdf_text = pdf_to_text(pdf_file.name)
+        cv_analysis = analyze_cv(pdf_text)  # Call analyze_cv with the extracted PDF text
+        user_message = f"CV Analysis:\n{cv_analysis}\n\nUser Message:\n{user_message}"
+    response_gen = respond(
+        user_message, history, system_message, max_tokens, temperature, top_p
+    )
+    return list(response_gen)[-1], history + [(user_message, "")]
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+# Define Gradio interface
+interface = gr.Interface(
+    fn=chatbot_with_pdf,
+    inputs=[
+        gr.File(label="Upload a PDF File"),
+        gr.Textbox(label="Your Message"),
+        gr.State(label="Chat History"),
+        gr.Textbox(label="System Message", value="You are an AI assistant."),
+        gr.Slider(label="Max Tokens", minimum=1, maximum=1000, value=200),
+        gr.Slider(label="Temperature", minimum=0.0, maximum=1.0, value=0.7, step=0.1),
+        gr.Slider(label="Top P", minimum=0.0, maximum=1.0, value=0.9, step=0.1),
+    ],
+    outputs=[
+        gr.Textbox(label="Response"),
+        gr.State(label="Chat History"),
     ],
+    title="Chatbot with CV Analysis and PDF Integration",
 )
 
-
-if __name__ == "__main__":
-    demo.launch()
+# Launch Gradio app
+interface.launch()