rasel-harun commited on
Commit
946fd9f
·
verified ·
1 Parent(s): abd9376

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +21 -7
app.py CHANGED
@@ -4,7 +4,8 @@ import numpy as np
4
  import matplotlib.pyplot as plt
5
  import yfinance as yf
6
  import joblib
7
- from tensorflow.keras.models import load_model
 
8
  from sklearn.metrics import mean_squared_error, mean_absolute_percentage_error
9
  import math
10
  import warnings
@@ -23,6 +24,18 @@ def download_data(ticker, start="2015-01-01", end=None):
23
  df.rename(columns={'Close': 'close'}, inplace=True)
24
  return df
25
 
 
 
 
 
 
 
 
 
 
 
 
 
26
  # Forecast using saved models
27
  def forecast_models(ticker="AAPL", test_len=180):
28
  df = download_data(ticker)
@@ -58,19 +71,20 @@ def forecast_models(ticker="AAPL", test_len=180):
58
  except Exception as e:
59
  results['Prophet'] = {"error": str(e)}
60
 
61
- # ====== LSTM ======
62
  try:
63
- lstm_model = load_model("lstm_model.h5")
64
  scaler = joblib.load("scaler.pkl")
 
 
 
65
  scaled_series = scaler.transform(series.values.reshape(-1,1))
66
 
67
  window_size = 60
68
- X, y = [], []
69
  for i in range(window_size, len(scaled_series)):
70
  X.append(scaled_series[i-window_size:i,0])
71
- y.append(scaled_series[i,0])
72
- X = np.array(X)
73
- X = X.reshape((X.shape[0], X.shape[1], 1))
74
  split_index = len(train) - window_size
75
  X_test = X[split_index:]
76
 
 
4
  import matplotlib.pyplot as plt
5
  import yfinance as yf
6
  import joblib
7
+ from tensorflow.keras.models import Sequential
8
+ from tensorflow.keras.layers import LSTM, Dense, Dropout
9
  from sklearn.metrics import mean_squared_error, mean_absolute_percentage_error
10
  import math
11
  import warnings
 
24
  df.rename(columns={'Close': 'close'}, inplace=True)
25
  return df
26
 
27
+ # Build the same architecture as training
28
+ def build_lstm(window_size=60):
29
+ model = Sequential([
30
+ LSTM(64, return_sequences=True, input_shape=(window_size,1)),
31
+ Dropout(0.2),
32
+ LSTM(32),
33
+ Dropout(0.2),
34
+ Dense(1)
35
+ ])
36
+ model.compile(optimizer="adam", loss="mse")
37
+ return model
38
+
39
  # Forecast using saved models
40
  def forecast_models(ticker="AAPL", test_len=180):
41
  df = download_data(ticker)
 
71
  except Exception as e:
72
  results['Prophet'] = {"error": str(e)}
73
 
74
+ # ====== LSTM (weights only) ======
75
  try:
 
76
  scaler = joblib.load("scaler.pkl")
77
+ lstm_model = build_lstm(window_size=60)
78
+ lstm_model.load_weights("lstm_model.weights.h5") # ✅ weights-only loading
79
+
80
  scaled_series = scaler.transform(series.values.reshape(-1,1))
81
 
82
  window_size = 60
83
+ X = []
84
  for i in range(window_size, len(scaled_series)):
85
  X.append(scaled_series[i-window_size:i,0])
86
+ X = np.array(X).reshape((-1, window_size, 1))
87
+
 
88
  split_index = len(train) - window_size
89
  X_test = X[split_index:]
90