Spaces:
Sleeping
Sleeping
File size: 3,167 Bytes
94516ce 646e4f3 9480219 94516ce 9480219 94516ce 8d505f4 c045a8c 9480219 8d505f4 94516ce 8d505f4 94516ce 8d505f4 94516ce 8d505f4 94516ce 8d505f4 4d83474 9480219 4d83474 8d505f4 9480219 8d505f4 94516ce 646e4f3 94516ce |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 |
# app.py
import os
import gradio as gr
from text_extractor import extract_text_from_file
from embedder import get_embeddings
from vector_store import create_faiss_index, search_similar_cvs
from groq_api import summarize_match
# Global storage
cv_texts = []
cv_names = []
cv_vectors = []
faiss_index = None
def upload_cvs(files):
global cv_texts, cv_names, cv_vectors, faiss_index
try:
cv_texts = [extract_text_from_file(f.name) for f in files]
cv_names = [f.name for f in files]
cv_vectors = get_embeddings(cv_texts)
import numpy as np
if cv_vectors is None or np.array(cv_vectors).size == 0:
return "❌ No valid CVs extracted or embedded."
faiss_index = create_faiss_index(cv_vectors)
return f"✅ Uploaded and indexed {len(files)} CVs."
except Exception as e:
return f"❌ Error during upload: {e}"
def match_jd(jd_text):
global faiss_index
try:
if not faiss_index:
return "❌ Please upload CVs first."
if not jd_text.strip():
return "❌ Job description is empty."
jd_vector = get_embeddings([jd_text])[0]
top_k_indices = search_similar_cvs(jd_vector, faiss_index, k=3)
import os
matched_names = [os.path.basename(cv_names[i]) for i in top_k_indices]
matched_texts = [
cv_texts[i][:500] if cv_texts[i].strip() else "[No CV content]"
for i in top_k_indices
]
summary = summarize_match(jd_text, matched_names, matched_texts)
return f"✅ Top Matches:\n{matched_names}\n\n📝 Summary:\n{summary}"
except Exception as e:
return f"❌ Error during matching: {e}"
def clear_data():
global cv_texts, cv_names, cv_vectors, faiss_index
cv_texts, cv_names, cv_vectors, faiss_index = [], [], [], None
return "🧹 All data cleared. You can now start fresh."
# 🌟 Redesigned Gradio Interface with Blocks
with gr.Blocks() as app:
gr.Markdown("## 📄 CV Filter App")
gr.Markdown("Upload candidate CVs, enter a job description, and let AI find the top matches using Groq + FAISS.")
with gr.Row():
cv_upload = gr.File(label="📤 Upload CVs (PDF/DOCX)", file_types=[".pdf", ".docx"], file_count="multiple")
upload_button = gr.Button("📁 Upload & Index CVs")
upload_status = gr.Textbox(label="Upload Status", interactive=False)
with gr.Row():
jd_input = gr.Textbox(label="📋 Paste Job Description", lines=8, placeholder="Enter the job description here...")
match_button = gr.Button("🔍 Match CVs")
match_result = gr.Textbox(label="📊 Matching Result & Summary", lines=12, interactive=False)
with gr.Row():
clear_button = gr.Button("🧹 Clear All")
clear_output = gr.Textbox(label="Status", interactive=False)
# Function bindings
upload_button.click(fn=upload_cvs, inputs=[cv_upload], outputs=[upload_status])
match_button.click(fn=match_jd, inputs=[jd_input], outputs=[match_result])
clear_button.click(fn=clear_data, inputs=[], outputs=[clear_output])
if __name__ == "__main__":
app.launch()
|