Upload utils.py

core/utils.py

@@ -0,0 +1,49 @@
+import numpy as np
+import pandas as pd
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score
+import yfinance as yf
+import os
+
+def download_yahoo_data(ticker, start_date, end_date):
+    df = yf.download(ticker, start=start_date, end=end_date)
+    df = df[['Close']].dropna()
+    df.reset_index(inplace=True)
+    return df
+
+def load_offline_csv(file):
+    df = pd.read_csv(file)
+    if 'Date' in df.columns:
+        df['Date'] = pd.to_datetime(df['Date'])
+        df.set_index('Date', inplace=True)
+    return df
+
+def prepare_sequences(data, window_size):
+    X, y = [], []
+    for i in range(len(data) - window_size):
+        X.append(data[i:i + window_size])
+        y.append(data[i + window_size])
+    return np.array(X), np.array(y)
+
+def scale_data(df, column='Close'):
+    scaler = StandardScaler()
+    scaled = scaler.fit_transform(df[[column]])
+    return scaled, scaler
+
+def inverse_transform(scaler, data):
+    return scaler.inverse_transform(data)
+
+def compute_metrics(y_true, y_pred):
+    return {
+        "MSE": mean_squared_error(y_true, y_pred),
+        "MAE": mean_absolute_error(y_true, y_pred),
+        "R²": r2_score(y_true, y_pred)
+    }
+
+def rolling_backtest(model, X, y, step=5):
+    predictions = []
+    for i in range(0, len(X) - step, step):
+        X_batch = X[i:i + step]
+        pred = model.predict(X_batch)
+        predictions.extend(pred.flatten())
+    return predictions