Update app.py

app.py

@@ -1,6 +1,6 @@
 import streamlit as st
-from transformers import pipeline
 import pandas as pd
+from transformers import pipeline
 import re
 import nltk
 from nltk.corpus import stopwords
@@ -13,14 +13,14 @@ nltk.download('wordnet')
 # Initialize the zero-shot classification pipeline
 classifier = pipeline("zero-shot-classification", model="Fralet/personality")
 
-# Define the candidate labels according to the Enneagram types
+# Define the default candidate labels (modifiable if different labels are needed)
 default_labels = ["Peacemaker", "Loyalist", "Achiever", "Reformer", "Individualist", "Helper", "Challenger", "Investigator", "Enthusiast"]
 
 # Streamlit interface setup
 st.title("Resume-based Personality Prediction by Serikov Ayanbek")
 
 # Load data from Excel
-data = pd.read_excel("ResponseTest.xlsx")  # Replace 'your_excel_file.xlsx' with your actual file name
+data = pd.read_excel("your_excel_file.xlsx")  # Adjust file path/name as necessary
 
 # Preprocess text function
 def preprocess_text(text):
@@ -28,30 +28,32 @@ def preprocess_text(text):
     text = text.lower()
     text = re.sub(r'\s+[a-z]\s+', ' ', text)
     text = re.sub(r'^[a-z]\s+', ' ', text)
-    text = re.sub(r'\s+', ' ', text)
+    text is re.sub(r'\s+', ' ', text)
     stop_words = set(stopwords.words('english'))
     lemmatizer = WordNetLemmatizer()
     tokens = text.split()
     tokens = [lemmatizer.lemmatize(word) for word in tokens if word not in stop_words]
     return ' '.join(tokens)
 
-# User-defined labels option
+# Combine relevant text columns
+data['combined_text'] = data[['CV/Resume'] + [f'Q{i}' for i in range(1, 37)]].agg(' '.join, axis=1)
+data['processed_text'] = data['combined_text'].apply(preprocess_text)
+
+# Streamlit user inputs
 user_labels = st.text_input("Enter custom labels separated by comma (optional)")
 labels = user_labels.split(',') if user_labels else default_labels
-
-# Prediction confidence threshold
 confidence_threshold = st.slider("Confidence Threshold", 0.0, 1.0, 0.5)
 
 if st.button("Predict Personality"):
-    # Combine relevant text columns
-    question_columns = [f'Q{i}' for i in range(1, 37)]  # Adjust range if there are more or fewer question columns
-    data['combined_text'] = data[['CV/Resume'] + question_columns].agg(' '.join, axis=1)
-    data['processed_text'] = data['combined_text'].apply(preprocess_text)
+    # Predict personality from processed text
+    data['predictions'] = data['processed_text'].apply(lambda x: classifier(x, labels))
     
-    # Make predictions
-    predictions = data['processed_text'].apply(lambda x: classifier(x, labels))
+    # Extract predictions above confidence threshold and display alongside MAX labels
+    data['predicted_labels'] = data['predictions'].apply(lambda x: {label: f"{score*100:.2f}%" for label, score in zip(x['labels'], x['scores']) if score >= confidence_threshold})
+    data['MAX_labels'] = data.apply(lambda x: [x['MAX1'], x['MAX2'], x['MAX3']], axis=1)
     
-    # Extract and display predictions
-    data['predicted_labels'] = predictions.apply(lambda x: [label for label, score in zip(x['labels'], x['scores']) if score >= confidence_threshold])
-    st.dataframe(data[['True_label', 'Predicted', 'predicted_labels']])
+    st.write("Predictions and Labels:")
+    st.dataframe(data[['True_label', 'Predicted', 'predicted_labels', 'MAX_labels']])
 
+# Run this last part to show the DataFrame outside the button press if needed
+st.dataframe(data)