feat: update src

src/evalute.py

@@ -1,40 +0,0 @@
-import pandas as pd
-import joblib
-
-from sklearn.model_selection import train_test_split
-from sklearn.metrics import classification_report
-
-from src.preprocess import load_stopwords, preprocess
-from src.build_features import transform_features
-
-
-DATA_PATH = "data/shopee_sentiment.csv"
-STOPWORD_PATH = "data/vietnamese-stopwords.txt"
-
-
-def main():
-    df = pd.read_csv(DATA_PATH)
-
-    df = df.dropna(subset=["text", "label"])
-    df = df.drop_duplicates(subset=["text"])
-
-    stopwords = load_stopwords(STOPWORD_PATH)
-    df["text"] = df["text"].apply(lambda x: preprocess(x, stopwords))
-    df["text"] = df["text"].apply(lambda x: " ".join(x))
-
-    X_train, X_test, y_train, y_test = train_test_split(
-        df["text"], df["label"], test_size=0.2, random_state=42
-    )
-
-    model = joblib.load("models/model.pkl")
-    vectorizer = joblib.load("models/vectorizer.pkl")
-
-    X_test_vec = transform_features(X_test, vectorizer)
-
-    y_pred = model.predict(X_test_vec)
-
-    print(classification_report(y_test, y_pred))
-
-
-if __name__ == "__main__":
-    main()

src/pipeline/inference_pipeline.py

@@ -1,29 +0,0 @@
-import joblib
-
-from src.preprocess import preprocess, load_stopwords
-from src.build_features import transform_features
-
-
-STOPWORD_PATH = "data/vietnamese-stopwords.txt"
-
-
-def predict(text):
-    # Load
-    model = joblib.load("models/model.pkl")
-    vectorizer = joblib.load("models/vectorizer.pkl")
-    stopwords = load_stopwords(STOPWORD_PATH)
-
-    # Preprocess
-    tokens = preprocess(text, stopwords)
-    text_clean = " ".join(tokens)
-
-    # Feature
-    X = transform_features([text_clean], vectorizer)
-
-    # Predict
-    return model.predict(X)[0]
-
-
-if __name__ == "__main__":
-    text = input("Enter text: ")
-    print("Prediction:", "positive" if predict(text) == 1 else "negative")

src/pipeline/train_pipeline.py

@@ -1,49 +0,0 @@
-import pandas as pd
-import joblib
-
-from sklearn.model_selection import train_test_split
-
-from src.preprocess import load_stopwords, preprocess
-from src.build_features import build_features
-from src.train_model import train_model
-
-
-# ===== CONFIG =====
-DATA_PATH = "data/shopee_sentiment.csv"
-STOPWORD_PATH = "data/vietnamese-stopwords.txt"
-
-
-def run_pipeline():
-    # 1. Load data
-    df = pd.read_csv(DATA_PATH)
-
-    # 2. Clean
-    df = df.dropna(subset=["text", "label"])
-    df = df.drop_duplicates(subset=["text"])
-
-    # 3. Preprocess
-    stopwords = load_stopwords(STOPWORD_PATH)
-    df["text"] = df["text"].apply(lambda x: preprocess(x, stopwords))
-    df["text"] = df["text"].apply(lambda x: " ".join(x))
-
-    # 4. Split
-    X_train, X_test, y_train, y_test = train_test_split(
-        df["text"], df["label"], test_size=0.2, random_state=42
-    )
-
-    # 5. Feature
-    X_train_vec, vectorizer = build_features(X_train)
-    X_test_vec = vectorizer.transform(X_test)
-
-    # 6. Train
-    model = train_model(X_train_vec, y_train)
-
-    # 7. Save
-    joblib.dump(model, "models/model.pkl")
-    joblib.dump(vectorizer, "models/vectorizer.pkl")
-
-    print("Pipeline training completed!")
-
-
-if __name__ == "__main__":
-    run_pipeline()

src/train_model.py

@@ -1,6 +0,0 @@
-from sklearn.svm import SVC
-
-def train_model(X_train, y_train):
-    model = SVC(kernel='linear')
-    model.fit(X_train, y_train)
-    return model