Update model.py

model.py

@@ -6,7 +6,6 @@ from transformers import pipeline
 import pandas as pd
 from datetime import datetime
 import zipfile
-import shutil
 import re
 import imaplib
 import email
@@ -21,6 +20,7 @@ MODELS = {
 loaded_models = {}
 skills_classifier = None
 
+# ================== LOAD MODEL ==================
 def get_model(name):
     if name not in loaded_models:
         loaded_models[name] = SentenceTransformer(MODELS[name])
@@ -29,17 +29,20 @@ def get_model(name):
 def get_classifier():
     global skills_classifier
     if skills_classifier is None:
-        skills_classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+        skills_classifier = pipeline(
+            "zero-shot-classification",
+            model="facebook/bart-large-mnli"
+        )
     return skills_classifier
 
-# ================== FILE READER ==================
+# ================== TEXT EXTRACTION ==================
 def extract_text(file_path):
     ext = os.path.splitext(file_path)[1].lower()
 
     try:
         if ext == ".pdf":
             doc = fitz.open(file_path)
-            text = "\n".join([p.get_text() for p in doc])
+            text = "\n".join([page.get_text() for page in doc])
             doc.close()
             return text.strip()
 
@@ -52,7 +55,7 @@ def extract_text(file_path):
 
     return ""
 
-# ================== GMAIL FETCH ==================
+# ================== GMAIL FETCH (OPTIONAL) ==================
 def fetch_from_gmail(email_user, app_password):
     mail = imaplib.IMAP4_SSL("imap.gmail.com")
     mail.login(email_user, app_password)
@@ -81,12 +84,18 @@ def fetch_from_gmail(email_user, app_password):
 
 # ================== AI FEATURES ==================
 def extract_skills(text):
-    labels = ["Python", "Machine Learning", "Deep Learning", "SQL", "AWS", "Docker", "Communication"]
+    labels = [
+        "Python", "Machine Learning", "Deep Learning",
+        "SQL", "AWS", "Docker", "Communication"
+    ]
 
     try:
         clf = get_classifier()
         res = clf(text[:2000], labels, multi_label=True)
-        return ", ".join([l for l, s in zip(res["labels"], res["scores"]) if s > 0.4])
+
+        return ", ".join([
+            l for l, s in zip(res["labels"], res["scores"]) if s > 0.4
+        ])
     except:
         return "N/A"
 
@@ -95,11 +104,11 @@ def extract_qualifications(text):
     found = re.findall(pattern, text.lower())
     return ", ".join(set(found)).upper() if found else "Not mentioned"
 
-# ================== MAIN FUNCTION ==================
-def screen_resumes(job_desc, files, model_name="Fast (MiniLM)", threshold=0.65,
-                   gmail=None, password=None):
+# ================== MAIN FUNCTION (FIXED NAME) ==================
+def screen_resumes_backend(job_desc, files, model_name="Fast (MiniLM)", threshold=0.65,
+                           gmail=None, password=None):
 
-    # Gmail integration
+    # Gmail integration (optional)
     if gmail and password:
         files = fetch_from_gmail(gmail, password)
 
@@ -107,11 +116,11 @@ def screen_resumes(job_desc, files, model_name="Fast (MiniLM)", threshold=0.65,
     job_emb = model.encode(job_desc, convert_to_tensor=True)
 
     results = []
-
     os.makedirs("outputs", exist_ok=True)
 
     for f in files:
-        # safe file handling
+
+        # handle uploaded files (HF / Gradio)
         if hasattr(f, "read"):
             name = f.name
             path = f"temp_{name}"
@@ -134,16 +143,18 @@ def screen_resumes(job_desc, files, model_name="Fast (MiniLM)", threshold=0.65,
 
         results.append({
             "Candidate": name,
-            "Score": round(score * 100, 2),
+            "Score (%)": round(score * 100, 2),
             "Skills": extract_skills(text),
             "Qualification": extract_qualifications(text),
             "Status": status
         })
 
+    # save report
     df = pd.DataFrame(results)
     report_path = f"outputs/report_{datetime.now().strftime('%Y%m%d_%H%M')}.csv"
     df.to_csv(report_path, index=False)
 
+    # zip shortlisted
     zip_path = None
     shortlisted = df[df["Status"] == "Shortlisted"]