app.py

import pdfplumber
import docx
from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
import gradio as gr

# Function to extract text from PDF using pdfplumber
def extract_text_from_pdf(pdf_file):
    with pdfplumber.open(pdf_file) as pdf:
        text = ''
        for page in pdf.pages:
            text += page.extract_text()
    return text

# Function to extract text from DOCX
def extract_text_from_docx(docx_file):
    doc = docx.Document(docx_file)
    full_text = []
    for paragraph in doc.paragraphs:
        full_text.append(paragraph.text)
    return '\n'.join(full_text)

# Function to generate roast based on resume text
def generate_roast(resume_text):
    tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neo-1.3B")
    model = AutoModelForCausalLM.from_pretrained("EleutherAI/gpt-neo-1.3B")

    # Define the prompt
    prompt_text = "Roast this resume:\n\n"
    
    # Tokenize the prompt
    prompt_tokenized = tokenizer(prompt_text, return_tensors="pt")
    prompt_tokens = prompt_tokenized['input_ids'].shape[1]
    
    # Calculate remaining tokens for resume text
    max_resume_tokens = 2048 - prompt_tokens
    
    # Tokenize and truncate resume text
    resume_tokenized = tokenizer(resume_text, truncation=True, max_length=max_resume_tokens, return_tensors="pt")
    
    # Decode the truncated resume text back into a string
    truncated_resume_text = tokenizer.decode(resume_tokenized['input_ids'][0], skip_special_tokens=True)
    
    # Combine prompt and truncated resume text
    final_prompt = f"{prompt_text}{truncated_resume_text}\n\nRoast:"
    
    # Generate roast
    generator = pipeline('text-generation', model=model, tokenizer=tokenizer)
    roast = generator(final_prompt, max_new_tokens=50, num_return_sequences=1)
    
    return roast[0]['generated_text']

# Function to handle file uploads and extract text from resume files
def roast_resume(file=None, resume_text=None):
    if file:
        # Handle file uploads for PDF or DOCX
        if file.name.endswith('.pdf'):
            resume_text = extract_text_from_pdf(file)
        elif file.name.endswith('.docx'):
            resume_text = extract_text_from_docx(file)
        else:
            return "Unsupported file format. Please upload a PDF or DOCX file."
    elif resume_text:
        # Use pasted resume text
        pass
    else:
        return "No resume provided."
    
    # Generate the roast based on extracted or pasted resume text
    roast = generate_roast(resume_text)
    return roast

# Gradio interface with file upload or text input options
interface = gr.Interface(
    fn=roast_resume, 
    inputs=[gr.File(label="Upload Resume (PDF/DOCX)"), gr.Textbox(label="Or Paste Your Resume")], 
    outputs="text", 
    title="Resume Roaster",
    description="Upload your resume in PDF/DOCX format or paste your resume text, and let the AI roast it!"
)

# Launch Gradio app
interface.launch()