import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from xgboost import XGBClassifier
from sklearn.metrics import accuracy_score
import joblib

# Load dataset
df = pd.read_csv("dummy_sentiment_dataset.csv")

# Split
X_train, X_test, y_train, y_test = train_test_split(
    df["text"], df["label"], test_size=0.2, random_state=42
)

# TF-IDF
tfidf = TfidfVectorizer(max_features=5000)
X_train_tfidf = tfidf.fit_transform(X_train)
X_test_tfidf = tfidf.transform(X_test)

# Model
model = XGBClassifier(
    n_estimators=300,
    max_depth=6,
    learning_rate=0.1,
    eval_metric='logloss'
)

model.fit(X_train_tfidf, y_train)

# Evaluate
y_pred = model.predict(X_test_tfidf)
print("Accuracy:", accuracy_score(y_test, y_pred))

# Save model + vectorizer
joblib.dump(model, "model.joblib")
joblib.dump(tfidf, "tfidf_vectorizer.joblib")

print("✅ Model and vectorizer saved!")