Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| import pdfplumber | |
| import re | |
| import openpyxl | |
| from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer | |
| # Authenticate Hugging Face API (ensure you're logged in already) | |
| model_name = "meta-llama/Llama-3.1-70B-Instruct" # Replace with your actual model name | |
| tokenizer = AutoTokenizer.from_pretrained(model_name) | |
| model = AutoModelForCausalLM.from_pretrained(model_name) | |
| # Function to extract text from PDF | |
| def extract_text_from_pdf(pdf_path): | |
| with pdfplumber.open(pdf_path) as pdf: | |
| text = '' | |
| for page in pdf.pages: | |
| text += page.extract_text() | |
| return text | |
| # Function to parse the resume text for name, email, phone, and skills | |
| def parse_resume(text): | |
| # Define the prompts for each type of information | |
| prompts = { | |
| "name": "Extract the name from this resume:\n", | |
| "email": "Extract the email address from this resume:\n", | |
| "phone": "Extract the phone number from this resume:\n", | |
| "skills": "Extract the technical skills from this resume:\n" | |
| } | |
| results = {} | |
| for key, prompt in prompts.items(): | |
| # Generate model response for each field | |
| inputs = tokenizer(prompt + text, return_tensors="pt") | |
| outputs = model.generate(**inputs, max_length=500) | |
| response = tokenizer.decode(outputs[0], skip_special_tokens=True) | |
| if key == 'email': | |
| # Use regex to validate email format | |
| email = re.findall(r'\S+@\S+', response) | |
| results[key] = email[0] if email else None | |
| elif key == 'phone': | |
| # Use regex to validate phone number format | |
| phone = re.findall(r'\b\d{10,15}\b', response) | |
| results[key] = phone[0] if phone else None | |
| elif key == 'skills': | |
| # Extract technical skills | |
| results[key] = response | |
| else: | |
| results[key] = response | |
| return results | |
| # Function to save parsed data to Excel file | |
| def save_to_excel(parsed_data, output_file): | |
| wb = openpyxl.Workbook() | |
| ws = wb.active | |
| ws.append(["Name", "Email", "Phone", "Skills"]) | |
| for data in parsed_data: | |
| ws.append([data["name"], data["email"], data["phone"], data["skills"]]) | |
| wb.save(output_file) | |
| # Function to process PDF files and output an Excel file | |
| def process_pdfs(pdfs): | |
| parsed_data = [] | |
| for pdf in pdfs: | |
| # Extract text from the PDF | |
| text = extract_text_from_pdf(pdf.name) | |
| # Parse the text for relevant details | |
| parsed_info = parse_resume(text) | |
| # Add parsed information to the list | |
| parsed_data.append(parsed_info) | |
| # Save the parsed data to an Excel file | |
| output_file = "parsed_resumes.xlsx" | |
| save_to_excel(parsed_data, output_file) | |
| return output_file | |
| # Gradio interface setup with blank API space (Hugging Face integration) | |
| iface = gr.Interface( | |
| fn=process_pdfs, | |
| inputs=gr.File(file_count="multiple", type="file"), | |
| outputs=gr.File(), | |
| live=True, | |
| title="AI Resume Parser", | |
| description="Upload PDF resumes, and the app will parse and extract Name, Email, Phone, and Skills from them.", | |
| examples=[["path_to_sample_resume.pdf"]] # Provide sample files if necessary | |
| ) | |
| # Launch the Gradio app | |
| iface.launch() | |