Upload app.py

app.py

@@ -8,7 +8,6 @@ from tensorflow.keras.models import load_model
 from tensorflow.keras.preprocessing.sequence import pad_sequences
 import os
 
-# Abbreviations dictionary for job market
 abbreviations = {
     "mgr": "manager",
     "sr": "senior",
@@ -135,53 +134,90 @@ 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']
-
+def load_model_and_tokenizers_for_hr():
+    model_path = 'modelfile/bighr2.keras'
+    tokenizer_path = 'tokernizer/tokenizershr.pkl'
     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
 
+def load_model_and_tokenizers_for_it():
+    model_path = 'modelfile/bigit2.keras'
+    tokenizer_path = 'tokernizer/tokenizersit.pkl'
+    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
 
+def load_model_and_tokenizers_for_sales():
+    model_path = 'modelfile/bigrsales2.keras'
+    tokenizer_path = 'tokernizer/tokenizerssales.pkl'
+    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
 
+def load_model_and_tokenizers_for_health():
+    model_path = 'modelfile/bighealth2.keras'
+    tokenizer_path = 'tokernizer/tokenizershealth.pkl'
+    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': {
-        'model': r'modelfile\bighr2.keras',
-        'tokenizer': r'tokernizer\tokenizershr.pkl'
-    },
-    'IT': {
-        'model': r'modelfile\bigit2.keras',
-        'tokenizer': r'tokernizer\tokenizersit.pkl'
-    },
-    'Sales': {
-        'model': r'modelfile\bigrsales2.keras',
-        'tokenizer': r'tokernizer\tokenizerssales.pkl'
-    },
-    'Health': {
-        'model': r'modelfile\bighealth2.keras',
-        'tokenizer': r'tokernizer\tokenizershealth.pkl'
-    },
-    'Other': {
-        'model': r'modelfile\bigothers2.keras',
-        'tokenizer': r'tokernizer\tokenizersothers.pkl'
-    }
-}
+def load_model_and_tokenizers_for_other():
+    model_path = 'modelfile/bigothers2.keras'
+    tokenizer_path = 'tokernizer/tokenizersothers.pkl'
+    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
 
 # Streamlit UI
 st.title("ATS")
@@ -195,12 +231,21 @@ resume = st.text_area("Paste your Resume:", height=150)
 job_description = st.text_area("Paste Job Description:", height=150)
 
 # Sector selection
-sector = st.selectbox("Select Sector:", list(sectors.keys()))
+sector = st.selectbox("Select Sector:", ['HR', 'IT', 'Sales', 'Health', 'Other'])
 
 if st.button("Calculate ATS Score"):
     if resume and job_description:
         try:
-            model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizer(sector)
+            if sector == 'HR':
+                model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizers_for_hr()
+            elif sector == 'IT':
+                model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizers_for_it()
+            elif sector == 'Sales':
+                model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizers_for_sales()
+            elif sector == 'Health':
+                model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizers_for_health()
+            elif sector == 'Other':
+                model, resume_tokenizer, description_tokenizer, common_nouns_tokenizer = load_model_and_tokenizers_for_other()
 
             processed_resume = clean_and_preprocess(resume)
             processed_description = clean_and_preprocess(job_description)