import streamlit as st
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score, classification_report

# Load the dataset
@st.cache
def load_data():
    df = pd.read_csv("tweet_emotions.csv")
    return df

df = load_data()

# Train a Naive Bayes classifier
@st.cache
def train_classifier(data):
    vectorizer = CountVectorizer()
    X = vectorizer.fit_transform(data['content'])
    y = data['sentiment']

    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

    naive_bayes_model = MultinomialNB()
    naive_bayes_model.fit(X_train, y_train)

    return naive_bayes_model, vectorizer, X_test, y_test

naive_bayes_model, vectorizer, X_test, y_test = train_classifier(df)

# Streamlit UI
st.title("Emotion Detection App")

# User input
tweet = st.text_area("Enter a tweet:")

# Emotion detection
if st.button("Detect Emotion"):
    tweet_vectorized = vectorizer.transform([tweet])
    prediction = naive_bayes_model.predict(tweet_vectorized)
    st.success(f"Predicted Emotion: {prediction[0]}")

# Display the dataset
st.subheader("Dataset Preview:")
st.write(df.head())

# Model evaluation
st.subheader("Model Evaluation:")
y_pred = naive_bayes_model.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
classification_rep = classification_report(y_test, y_pred)

st.write(f"Accuracy: {accuracy:.4f}")
st.write("Classification Report:\n", classification_rep)