Updated For An ResumeAnalysis

app.py

@@ -1,10 +1,130 @@
 import gradio as gr
+from transformers import pipeline
+import PyPDF2
+import io
+import re
+from collections import Counter
 
-with gr.Blocks(fill_height=True) as demo:
-    with gr.Sidebar():
-        gr.Markdown("# Inference Provider")
-        gr.Markdown("This Space showcases the mistralai/Mistral-7B-Instruct-v0.3 model, served by the together API. Sign in with your Hugging Face account to use this API.")
-        button = gr.LoginButton("Sign in")
-    gr.load("models/mistralai/Mistral-7B-Instruct-v0.3", accept_token=button, provider="together")
+# Initialize Mistral-7B for analysis
+analyzer = pipeline("text-generation", model="mistralai/Mistral-7B-Instruct-v0.3")
+
+# Predefined valuable skills (general mode)
+GENERAL_SKILLS = {
+    'communication': 2, 'problem solving': 2, 'project management': 3,
+    'Python': 3, 'SQL': 2, 'Excel': 1, 'teamwork': 1
+}
+
+def extract_text_from_pdf(pdf_file):
+    reader = PyPDF2.PdfReader(io.BytesIO(pdf_file))
+    return "\n".join([page.extract_text() for page in reader.pages])
+
+def calculate_scores(resume_text, job_desc=None):
+    scores = {
+        "Relevance to Job": 0,
+        "Experience Quality": 0,
+        "Skills Match": 0,
+        "Education": 0,
+        "Achievements": 0,
+        "Clarity": 10,  # Start with full points, deduct for errors
+        "Customization": 0
+    }
+    
+    # 1. Relevance to Job (0-20)
+    if job_desc:
+        resume_words = set(re.findall(r'\w+', resume_text.lower()))
+        job_words = set(re.findall(r'\w+', job_desc.lower()))
+        overlap = len(resume_words & job_words) / len(job_words)
+        scores["Relevance to Job"] = min(20, int(overlap * 20))
+    else:
+        # General employability heuristic
+        scores["Relevance to Job"] = min(10, sum(
+            1 for skill in GENERAL_SKILLS if skill in resume_text.lower()
+        ))
+    
+    # 2. Experience Quality (0-20)
+    exp_years = len(re.findall(r'\d{4}\s*[-–]\s*(?:Present|\d{4})', resume_text))
+    scores["Experience Quality"] = min(10, exp_years)  # 1 point/year
+    
+    # Quantifiable achievements
+    metrics = re.findall(r'(increased|reduced|saved|improved)\s+by\s+(\d+%|\$\d+)', resume_text, re.I)
+    scores["Experience Quality"] += min(10, len(metrics) * 2)
+    
+    # 3. Skills Match (0-15)
+    if job_desc:
+        required_skills = re.findall(r'\b(?:proficient|skilled|experienced)\s+in\s+([\w\s]+)', job_desc, re.I)
+        found_skills = sum(1 for skill in required_skills if skill.lower() in resume_text.lower())
+        scores["Skills Match"] = min(15, found_skills * 3)  # 3 points per skill
+    else:
+        scores["Skills Match"] = sum(
+            points for skill, points in GENERAL_SKILLS.items() 
+            if skill in resume_text.lower()
+        )
+    
+    # 4. Education (0-10)
+    degrees = {
+        'phd': 8, 'doctorate': 8,
+        'master': 6, 'msc': 6, 'mba': 6,
+        'bachelor': 4, 'bs ': 4, 'ba ': 4,
+        'high school': 2
+    }
+    for degree, points in degrees.items():
+        if degree in resume_text.lower():
+            scores["Education"] = max(scores["Education"], points)
+    
+    # 5. Achievements (0-15)
+    quantifiers = re.findall(r'(?:achieved|reached|attained)\s+\d+%|\$\d+', resume_text, re.I)
+    scores["Achievements"] = min(15, len(quantifiers) * 3)
+    
+    # 6. Clarity (0-10)
+    # Deduct for errors
+    typos = len(re.findall(r'\b(?:responsibilities|accomplishment|experiance)\b', resume_text, re.I))
+    scores["Clarity"] -= min(8, typos)
     
+    # 7. Customization (0-10)
+    if job_desc and 'summary' in resume_text.lower():
+        job_title = re.search(r'\b(?:senior|junior)?\s*\w+\s*(?:developer|engineer|manager)', job_desc, re.I)
+        if job_title and job_title.group().lower() in resume_text.lower():
+            scores["Customization"] += 5
+    
+    total = sum(scores.values())
+    return scores, min(100, total)  # Cap at 100
+
+def analyze_resume(pdf_file, job_desc=None):
+    resume_text = extract_text_from_pdf(pdf_file)
+    scores, total = calculate_scores(resume_text, job_desc)
+    
+    # Generate AI feedback
+    prompt = f"""
+    Given these resume scores: {scores}
+    Provide concise feedback in this format:
+    
+    🔍 Overall Score: {total}/100
+    ✅ Strengths: [Top 2 scoring categories]
+    📌 Improvements: [2 weakest categories with specific advice]
+    """
+    analysis = analyzer(prompt, max_new_tokens=300)[0]["generated_text"]
+    
+    # Format results
+    breakdown = "\n".join([f"{k}: {v}/{(20 if k=='Experience Quality' else 15 if k=='Skills Match' else 10)}" 
+                          for k, v in scores.items()])
+    
+    return f"{analysis}\n\n📊 Detailed Breakdown:\n{breakdown}"
+
+# Gradio Interface
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    gr.Markdown("# 🎯 Smart Resume Analyzer")
+    with gr.Row():
+        with gr.Column():
+            resume = gr.File(label="Upload Resume (PDF)")
+            job_desc = gr.Textbox(label="Optional: Paste Job Description", lines=3)
+            analyze_btn = gr.Button("Analyze", variant="primary")
+        with gr.Column():
+            output = gr.Textbox(label="Analysis Results", interactive=False)
+    
+    analyze_btn.click(
+        fn=analyze_resume,
+        inputs=[resume, job_desc],
+        outputs=output
+    )
+
 demo.launch()