Create app.py

app.py

@@ -0,0 +1,160 @@
+import gradio as gr
+import os
+import docx
+import pandas as pd
+from sentence_transformers import SentenceTransformer, util
+from PyPDF2 import PdfReader
+import re
+from datetime import datetime
+
+# Load pre-trained model for sentence embedding
+model = SentenceTransformer('paraphrase-MiniLM-L6-v2')
+
+# Define maximum number of resumes
+MAX_RESUMES = 10
+
+# Keywords related to managerial and leadership roles
+MANAGERIAL_KEYWORDS = ["manager", "team leader", "lead", "supervisor", "director", "head of", "leadership"]
+
+# Function to load job description from file path
+def load_job_description(job_desc_file):
+    if not os.path.exists(job_desc_file):
+        return "Job description file not found."
+    with open(job_desc_file, 'r') as file:
+        job_description = file.read()
+    if not job_description.strip():
+        return "Job description is empty."
+    return job_description
+
+# Function to check similarity between resumes and job description
+def check_similarity(job_description, resume_files):
+    results = []
+    job_emb = model.encode(job_description, convert_to_tensor=True)
+
+    for resume_file in resume_files:
+        resume_text = extract_text_from_resume(resume_file)
+        if not resume_text:
+            results.append((resume_file.name, 0, "Not Eligible", None, None))
+            continue
+        
+        resume_emb = model.encode(resume_text, convert_to_tensor=True)
+        similarity_score = util.pytorch_cos_sim(job_emb, resume_emb)[0][0].item()
+
+        # Extract leadership experience from resume
+        leadership_experience = extract_leadership_experience(resume_text)
+
+        # Increase the weight of the similarity score for candidates with managerial experience
+        if leadership_experience > 0:
+            similarity_score += 0.1  # Adjust the weight based on leadership experience
+        
+        # Set a higher similarity threshold for eligibility
+        if similarity_score >= 0.50:
+            candidate_name = extract_candidate_name(resume_text)
+            results.append((resume_file.name, similarity_score, "Eligible", candidate_name, leadership_experience))
+        else:
+            results.append((resume_file.name, similarity_score, "Not Eligible", None, None))
+
+    return results
+
+# Extract text from resume (handles .txt, .pdf, .docx)
+def extract_text_from_resume(resume_file):
+    file_extension = os.path.splitext(resume_file)[1].lower()
+    if file_extension not in ['.txt', '.pdf', '.docx']:
+        return "Unsupported file format"
+
+    if file_extension == '.txt':
+        return read_text_file(resume_file)
+    elif file_extension == '.pdf':
+        return read_pdf_file(resume_file)
+    elif file_extension == '.docx':
+        return read_docx_file(resume_file)
+
+    return "Failed to read the resume text."
+
+def read_text_file(file_path):
+    with open(file_path, 'r') as file:
+        return file.read()
+
+def read_pdf_file(file_path):
+    reader = PdfReader(file_path)
+    text = ""
+    for page in reader.pages:
+        text += page.extract_text()
+    return text
+
+def read_docx_file(file_path):
+    doc = docx.Document(file_path)
+    text = ""
+    for para in doc.paragraphs:
+        text += para.text
+    return text
+
+# Extract candidate name from resume text
+def extract_candidate_name(resume_text):
+    name_pattern = re.compile(r'\b([A-Z][a-z]+ [A-Z][a-z]+)\b')
+    matches = name_pattern.findall(resume_text)
+    if matches:
+        return matches[0]  # Returns the first match
+    return "Unknown Candidate"
+
+# Extract leadership experience (years of managerial experience)
+def extract_leadership_experience(resume_text):
+    experience = 0
+    for keyword in MANAGERIAL_KEYWORDS:
+        pattern = r"\b" + keyword + r"\b.*?(\d{4}|\d{2})[\s\-/]*\d{2,4}"
+        matches = re.findall(pattern, resume_text, re.IGNORECASE)
+        for match in matches:
+            if isinstance(match, str) and match.isdigit():
+                experience = max(experience, int(match))  # Use the highest value
+    return experience
+
+# Main processing function
+def process_files(job_desc, resumes):
+    try:
+        # Check if the number of resumes is within the allowed limit
+        if len(resumes) > MAX_RESUMES:
+            return "Please upload no more than 10 resumes."
+
+        # Check if all necessary files are provided
+        if not job_desc or not resumes:
+            return "Please provide all necessary files."
+
+        # Load the job description
+        job_desc_text = load_job_description(job_desc)
+
+        # Check similarity
+        results = check_similarity(job_desc_text, resumes)
+
+        # Prepare the results in tabular form
+        df = pd.DataFrame(results, columns=["Resume File", "Similarity Score", "Eligibility", "Candidate Name", "Leadership Experience"])
+        
+        # Output file for downloading
+        output_filename = f"/tmp/similarity_results_{datetime.now().strftime('%Y%m%d%H%M%S')}.csv"
+        df.to_csv(output_filename, index=False)
+        
+        # Return the results as a table
+        return df, output_filename
+
+    except Exception as e:
+        # Return any errors encountered during processing
+        return f"Error processing files: {str(e)}", None
+
+
+# Gradio Interface Components
+job_desc_input = gr.File(label="Upload Job Description (TXT)", type="filepath")
+resumes_input = gr.Files(label="Upload Resumes (TXT, DOCX, PDF)", type="filepath")
+
+# Gradio Outputs
+results_output = gr.Dataframe(label="Analysis Results")
+download_output = gr.File(label="Download Final Results")
+
+# Gradio Interface
+interface = gr.Interface(
+    fn=process_files,
+    inputs=[job_desc_input, resumes_input],
+    outputs=[results_output, download_output],
+    title="HR Assistant - Resume Screening",
+    description="Upload job description and resumes to screen candidates and download the results in a tabular format."
+)
+
+interface.launch()