Update app.py

app.py

@@ -3,36 +3,30 @@ from pydantic import BaseModel
 import joblib
 from sentence_transformers import SentenceTransformer
 import re
-import nltk
 from nltk.corpus import stopwords
 import numpy as np
 
-# Pastikan stopwords ada
-try:
-    stopwords.words("english")
-except LookupError:
-    nltk.download("stopwords")
-
 # === Preprocessing Function ===
+stop_words = set(stopwords.words("english"))
+
 def preprocess_text(text: str) -> str:
     if not isinstance(text, str) or text.strip() == "":
         return ""
     text = text.lower()
     text = re.sub(r"\r\n", " ", text)
     text = re.sub(r"[^a-z\s]", "", text)
-    tokens = text.split()
-    stop_words = set(stopwords.words("english"))
-    tokens = [word for word in tokens if word not in stop_words]
+    tokens = [w for w in text.split() if w not in stop_words]
     return " ".join(tokens)
 
-# === Load Models ===
+# === Load SentenceTransformer ===
 print("Loading SentenceTransformer...")
-st_model = SentenceTransformer("all-mpnet-base-v2")
+st_model = SentenceTransformer("sentence-transformers/all-mpnet-base-v2")
 
+# === Load XGBoost models ===
 print("Loading XGBoost models...")
 models = joblib.load("xgb_models_all.joblib")
 
-# === FastAPI ===
+# === FastAPI app ===
 app = FastAPI(title="Essay Scoring API")
 
 class EssayInput(BaseModel):
@@ -46,11 +40,11 @@ def predict(input_data: EssayInput):
     # 2. Embedding
     vec = st_model.encode([clean_text], normalize_embeddings=True)
 
-    # 3. Tambah feature essay_length
+    # 3. Add essay_length feature
     essay_length = len(input_data.text)
     X = np.concatenate([vec, [[essay_length]]], axis=1)
 
-    # 4. Predict dengan setiap model
+    # 4. Predictions from all models
     results = {}
     for col, model in models.items():
         results[col] = float(model.predict(X)[0])