from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.naive_bayes import MultinomialNB

from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.metrics import classification_report, confusion_matrix, accuracy_score, ConfusionMatrixDisplay
from sklearn.preprocessing import LabelEncoder
from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer, TfidfTransformer

import joblib
import gradio as gr 





# Load the saved model and preprocessing objects
model = joblib.load('models/spam_classifier_model.joblib')
cv = joblib.load('models/count_vectorizer.joblib')
tfidf = joblib.load('models/tfidf_transformer.joblib')
le = joblib.load('models/label_encoder.joblib')

def predict_spam(message):
    X_new_counts = cv.transform([message])
    X_new_tfidf = tfidf.transform(X_new_counts)
    pred = model.predict(X_new_tfidf)[0]
    label = le.inverse_transform([pred])[0]
    return f"Prediction: {label}"

# Create Gradio interface
iface = gr.Interface(
    title=gr.Markdown('# 📱💬 SMS Spam Classifier'),
    theme=gr.themes.Soft(),
    fn=predict_spam,
    inputs=gr.Textbox(lines=10, placeholder="Enter an SMS message..."),
    outputs=gr.Textbox(lines=10),
    description="# 📱💬 SMS Spam Classifier\n\nEnter an SMS message to classify it as 'spam' or 'ham' using the best model.",
    examples=['Congratulations! You have won a $1,000 Walmart gift card. Go to http://bit.ly/123456 to claim your prize now. Reply STOP to opt out.',
              'Hi, how are you?']
)

iface.launch()