Upload app.py

app.py

@@ -135,9 +135,33 @@ def extract_nouns(text):
     nouns = [token.lemma_ for token in doc if token.pos_ == "NOUN"]
     return nouns
 
+def load_model_and_tokenizer(sector):
+    model_path = sectors[sector]['model']
+    tokenizer_path = sectors[sector]['tokenizer']
+
+    if not os.path.isfile(model_path):
+        raise FileNotFoundError(f"Model file not found: {model_path}")
+    
+    if not os.path.isfile(tokenizer_path):
+        raise FileNotFoundError(f"Tokenizer file not found: {tokenizer_path}")
+
+    model = load_model(model_path)
+
+    with open(tokenizer_path, 'rb') as f:
+        tokenizers = pickle.load(f)
+
+    resume_tokenizer = tokenizers.get('resume_tokenizer')
+    description_tokenizer = tokenizers.get('description_tokenizer')
+    common_nouns_tokenizer = tokenizers.get('common_nouns_tokenizer')
+
+    if not (resume_tokenizer and description_tokenizer and common_nouns_tokenizer):
+        raise ValueError("Tokenizer components are missing from the file.")
+
+    return model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer
+
 # Define the sector options and their corresponding model and tokenizer paths
 sectors = {
-     'HR': {
+    'HR': {
         'model': r'modelfile\bighr2.keras',
         'tokenizer': r'tokernizer\tokenizershr.pkl'
     },
@@ -176,45 +200,31 @@ sector = st.selectbox("Select Sector:", list(sectors.keys()))
 if st.button("Calculate ATS Score"):
     if resume and job_description:
         try:
-            model_path = sectors[sector]['model']
-            tokenizer_path = sectors[sector]['tokenizer']
-
-            if not os.path.isfile(model_path):
-                st.error(f"Model file not found: {model_path}")
-            elif not os.path.isfile(tokenizer_path):
-                st.error(f"Tokenizer file not found: {tokenizer_path}")
-            else:
-                model = load_model(model_path)
-
-                with open(tokenizer_path, 'rb') as f:
-                    tokenizers = pickle.load(f)
-
-                resume_tokenizer = tokenizers.get('resume_tokenizer')
-                description_tokenizer = tokenizers.get('description_tokenizer')
-                common_nouns_tokenizer = tokenizers.get('common_nouns_tokenizer')
+            model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizer(sector)
 
-                if not (resume_tokenizer and description_tokenizer and common_nouns_tokenizer):
-                    st.error("Tokenizer components are missing from the file.")
-                else:
-                    processed_resume = clean_and_preprocess(resume)
-                    processed_description = clean_and_preprocess(job_description)
+            processed_resume = clean_and_preprocess(resume)
+            processed_description = clean_and_preprocess(job_description)
 
-                    resume_sequence = resume_tokenizer.texts_to_sequences([processed_resume])
-                    resume_data_padded = pad_sequences(resume_sequence, maxlen=1500)
+            resume_sequence = resume_tokenizer.texts_to_sequences([processed_resume])
+            resume_data_padded = pad_sequences(resume_sequence, maxlen=1500)
 
-                    description_sequence = description_tokenizer.texts_to_sequences([processed_description])
-                    description_data_padded = pad_sequences(description_sequence, maxlen=1500)
+            description_sequence = description_tokenizer.texts_to_sequences([processed_description])
+            description_data_padded = pad_sequences(description_sequence, maxlen=1500)
 
-                    common_nouns = set(extract_nouns(processed_resume))
-                    common_nouns_str = ' '.join(common_nouns)
+            common_nouns = set(extract_nouns(processed_resume))
+            common_nouns_str = ' '.join(common_nouns)
 
-                    common_nouns_sequence = common_nouns_tokenizer.texts_to_sequences([common_nouns_str])
-                    common_nouns_data = pad_sequences(common_nouns_sequence, maxlen=10)
+            common_nouns_sequence = common_nouns_tokenizer.texts_to_sequences([common_nouns_str])
+            common_nouns_data = pad_sequences(common_nouns_sequence, maxlen=10)
 
-                    prediction = model.predict([resume_data_padded, description_data_padded, common_nouns_data])
-                    st.success(f"Your predicted ATS Score is: {prediction[0][0]:.2f}")
+            prediction = model.predict([resume_data_padded, description_data_padded, common_nouns_data])
+            st.success(f"Your predicted ATS Score is: {prediction[0][0]:.2f}")
 
+        except FileNotFoundError as fnf_error:
+            st.error(str(fnf_error))
+        except ValueError as val_error:
+            st.error(str(val_error))
         except Exception as e:
             st.error(f"An error occurred: {e}")
     else:
-        st.error("Please paste both your resume and job description before analyzing.")
+        st.error("Please paste both your resume and job description before analyzing.")