Create app.py

app.py

@@ -0,0 +1,96 @@
+import gradio as gr
+from langchain_community.document_loaders import PyPDFLoader
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+from sklearn.metrics.pairwise import cosine_similarity
+from langchain_groq import ChatGroq
+from langchain.chains import LLMChain
+from langchain.prompts import PromptTemplate
+import numpy as np
+import os
+
+# ✅ Set Groq API Key
+os.environ["GROQ_API_KEY"] = "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"  # Replace with your actual Groq API key
+
+def extract_text_from_pdf(pdf_file):
+    temp_path = f"temp_{pdf_file.name}"
+    with open(temp_path, "wb") as f:
+        f.write(pdf_file.read())
+    loader = PyPDFLoader(temp_path)
+    pages = loader.load_and_split()
+    os.remove(temp_path)
+    return " ".join([page.page_content for page in pages])
+
+def extract_skills(text):
+    skills_list = ["Python", "SQL", "Machine Learning", "Deep Learning", "NLP", "Data Visualization", "Cloud", "TensorFlow", "PyTorch", "Statistics", "Java", "C++", "HTML", "CSS", "JavaScript"]
+    return [skill for skill in skills_list if skill.lower() in text.lower()]
+
+def calculate_match(user_skills, job_skills):
+    common = set(user_skills) & set(job_skills)
+    match_percent = (len(common) / len(job_skills)) * 100 if job_skills else 0
+    missing_skills = list(set(job_skills) - set(user_skills))
+    return round(match_percent, 2), missing_skills
+
+def generate_skill_gap_report(user_skills, job_skills, missing_skills, match_percent):
+    llm = ChatGroq(model="llama3-8b-8192", temperature=0.2)
+
+    template = """
+    User Skills: {user_skills}
+    Job Requirements: {job_skills}
+    Missing Skills: {missing_skills}
+    Match Percentage: {match_percent}%
+
+    Generate a short, friendly skill gap report. Suggest next steps for the user to improve their chances.
+    """
+
+    prompt = PromptTemplate.from_template(template)
+    chain = LLMChain(llm=llm, prompt=prompt)
+
+    report = chain.run({
+        "user_skills": ", ".join(user_skills),
+        "job_skills": ", ".join(job_skills),
+        "missing_skills": ", ".join(missing_skills),
+        "match_percent": match_percent
+    })
+
+    return report
+
+def process_skill_gap(resume_pdf, jd_pdf):
+    if resume_pdf is None or jd_pdf is None:
+        return "❌ Please upload both Resume and Job Description PDFs.", "", "", ""
+
+    resume_text = extract_text_from_pdf(resume_pdf)
+    jd_text = extract_text_from_pdf(jd_pdf)
+
+    user_skills = extract_skills(resume_text)
+    job_skills = extract_skills(jd_text)
+
+    match_percent, missing_skills = calculate_match(user_skills, job_skills)
+
+    embed_model = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
+    vectors = embed_model.embed_documents([resume_text, jd_text])
+    similarity_score = cosine_similarity([vectors[0]], [vectors[1]])[0][0]
+    similarity_percent = round(similarity_score * 100, 2)
+
+    report = generate_skill_gap_report(user_skills, job_skills, missing_skills, match_percent)
+
+    return f"✅ Skill Match: {match_percent}%", f"❌ Missing Skills: {', '.join(missing_skills) if missing_skills else 'None'}", f"🔎 Similarity Score: {similarity_percent}%", report
+
+demo = gr.Interface(
+    fn=process_skill_gap,
+    inputs=[
+        gr.File(label="Upload Resume (PDF)", type="binary"),
+        gr.File(label="Upload Job Description (PDF)", type="binary")
+    ],
+    outputs=[
+        gr.Textbox(label="Skill Match Percentage"),
+        gr.Textbox(label="Missing Skills"),
+        gr.Textbox(label="Similarity Score"),
+        gr.Textbox(label="AI-Generated Skill Gap Report")
+    ],
+    title="📄 Skill Gap AI Checker (Gradio + LangChain + Groq LLaMA3)",
+    description="Upload your Resume PDF and Job Description PDF to analyze your skill match percentage, missing skills, and get an AI-generated report using Groq's LLaMA3 model."
+)
+
+if __name__ == "__main__":
+    demo.launch()