Update app.py

app.py

@@ -3,7 +3,7 @@ import pandas as pd
 from neuralprophet import NeuralProphet
 import warnings
 import torch.optim as optim
-from torch.optim.lr_scheduler import LambdaLR
+from torch.optim.lr_scheduler import OneCycleLR
 
 warnings.filterwarnings("ignore", category=UserWarning)
 
@@ -20,11 +20,6 @@ class CustomNeuralProphet(NeuralProphet):
         super().__init__(**kwargs)
         self.optimizer = None
 
-    def lr_scheduler_step(self, epoch, batch_idx, optimizer):
-        # Custom logic for LR scheduler step
-        for lr_scheduler in optimizer.param_groups[0]['lr_scheduler']:
-            lr_scheduler.step()
-
 m = CustomNeuralProphet(
     n_forecasts=30,
     n_lags=12,
@@ -46,8 +41,15 @@ m = CustomNeuralProphet(
 m.fit(df, freq='D')  # Fit the model first before accessing the optimizer
 m.optimizer = optim.Adam(m.model.parameters(), lr=0.03)  # Example optimizer, adjust as needed
 
-lr_scheduler = LambdaLR(m.optimizer, lambda epoch: 0.95 ** epoch)  # Example LR scheduler, adjust as needed
-m.optimizer.param_groups[0]['lr_scheduler'] = [lr_scheduler]
+lr_scheduler = OneCycleLR(
+    m.optimizer,
+    max_lr=0.1,
+    total_steps=100,
+    pct_start=0.3,
+    anneal_strategy='cos',
+)  # Example LR scheduler, adjust as needed
+
+m.trainer.lr_schedulers = [lr_scheduler]  # Set the LR scheduler to the trainer
 
 future = m.make_future_dataframe(df, periods=30, n_historic_predictions=True)
 forecast = m.predict(future)
@@ -67,11 +69,3 @@ if __name__ == "__main__":
     disclaimer_output = gr.outputs.Textbox(label="Disclaimer")
     interface = gr.Interface(fn=predict_vn_index, inputs=dropdown, outputs=[image_output, disclaimer_output], title="Dự báo VN Index 30 ngày tới")
     interface.launch()
-
-
-
-
-
-
-
-