Update app.py

app.py

@@ -1,61 +1,64 @@
-# app.py
 from fastapi import FastAPI
 from pydantic import BaseModel
 import numpy as np
 import pandas as pd
-import joblib
-import json
 import tensorflow as tf
 from tensorflow.keras.models import load_model
-from sklearn.preprocessing import MinMaxScaler
+import json
+import os
 
 app = FastAPI(
     title="Forex LSTM Prediction API",
-    description="API untuk prediksi harga EUR/USD H+1 menggunakan model LSTM terbaik",
-    version="1.0"
+    description="Prediksi harga EUR/USD H+1 dengan LSTM menggunakan scaler harian",
+    version="2.0"
 )
 
 # ==========================================================
-# LOAD MODEL, PARAMETER, DAN SCALER
+# LOAD MODEL DAN KONFIGURASI
 # ==========================================================
 MODEL_PATH = "lstm_model.h5"
-SCALER_PATH = "scaler.pkl"
 PARAMS_PATH = "best_params.json"
+SCALER_FILE = "scaler_config.json"
 
 print("📥 Loading LSTM model...")
-model = load_model(MODEL_PATH, compile = False)
-
-print("📥 Loading scaler...")
-scaler = joblib.load(SCALER_PATH)
+model = load_model(MODEL_PATH, compile=False)
 
 print("📥 Loading best parameters...")
 with open(PARAMS_PATH, "r") as f:
     best_params = json.load(f)
 
-LOOKBACK = best_params.get("lookback", 7)  # default 7 jika tidak ada
+LOOKBACK = best_params.get("lookback", 7)
 FEATURE_ORDER = best_params.get("features", [
     "mood_score", "t_pos", "t_neg", "c_pos", "c_neg",
     "norm_ema20", "norm_ema50", "norm_close"
 ])
 
-print(f"✅ Model loaded. Lookback={LOOKBACK}, Features={FEATURE_ORDER}")
+# ==========================================================
+# LOAD SCALER CONFIG
+# ==========================================================
+def load_scaler_config():
+    if not os.path.exists(SCALER_FILE):
+        print("⚠️ Scaler config not found, using default values.")
+        return {"CLOSE_MIN": 1.05, "CLOSE_MAX": 1.15}
+    with open(SCALER_FILE, "r") as f:
+        return json.load(f)
+
+scaler_cfg = load_scaler_config()
+CLOSE_MIN = scaler_cfg["CLOSE_MIN"]
+CLOSE_MAX = scaler_cfg["CLOSE_MAX"]
+
+print(f"✅ Scaler range loaded: {CLOSE_MIN:.5f} - {CLOSE_MAX:.5f}")
 
 # ==========================================================
 # INPUT SCHEMA
 # ==========================================================
 class LSTMInput(BaseModel):
-    data: list  # list of daily records (latest LOOKBACK hari)
-    # contoh format:
-    # [
-    #   {"date": "2025-10-23", "mood_score": 0.5, "t_pos": 0.3, "t_neg": 0.2, ...},
-    #   ...
-    # ]
+    data: list
 
 # ==========================================================
-# HELPER FUNCTION
+# HELPER FUNCTIONS
 # ==========================================================
 def prepare_input(data):
-    """Convert input list to numpy array sesuai urutan fitur dan lookback"""
     df = pd.DataFrame(data)
     missing_cols = [f for f in FEATURE_ORDER if f not in df.columns]
     if missing_cols:
@@ -64,10 +67,14 @@ def prepare_input(data):
     X = df[FEATURE_ORDER].values[-LOOKBACK:]
     if X.shape[0] < LOOKBACK:
         raise ValueError(f"Need at least {LOOKBACK} timesteps, got {X.shape[0]}")
-
     X = np.expand_dims(X, axis=0)
     return X, df
 
+
+def inverse_scale(norm_value):
+    """Denormalisasi nilai close dari [0,1] ke skala asli"""
+    return (norm_value * (CLOSE_MAX - CLOSE_MIN)) + CLOSE_MIN
+
 # ==========================================================
 # ENDPOINT
 # ==========================================================
@@ -76,30 +83,23 @@ def predict_price(input_data: LSTMInput):
     try:
         X, df = prepare_input(input_data.data)
         pred_norm = model.predict(X)[0][0]
-        pred_close = scaler.inverse_transform([[pred_norm]])[0][0]
+        pred_close = inverse_scale(pred_norm)
 
         last_date = pd.to_datetime(df["date"].iloc[-1])
         next_date = (last_date + pd.Timedelta(days=1)).strftime("%Y-%m-%d")
 
-        response = {
+        result = {
             "next_date": next_date,
             "predicted_norm_close": float(pred_norm),
             "predicted_close": float(pred_close),
-            "model_info": {
-                "lookback": LOOKBACK,
-                "features_used": FEATURE_ORDER,
-                "source_model": MODEL_PATH
-            }
+            "scaler_used": {"min": CLOSE_MIN, "max": CLOSE_MAX},
+            "features_used": FEATURE_ORDER
         }
-
-        return {"status": "ok", "result": response}
+        return {"status": "ok", "result": result}
 
     except Exception as e:
         return {"status": "error", "message": str(e)}
 
-# ==========================================================
-# ROOT TEST
-# ==========================================================
 @app.get("/")
 def root():
     return {"message": "Forex LSTM Prediction API is active!"}