Adds lr moitor to train
Browse files
train.py
CHANGED
@@ -13,12 +13,12 @@ torch.set_float32_matmul_precision("high")
|
|
13 |
|
14 |
|
15 |
# Init DataModule
|
16 |
-
dm = DRDataModule(batch_size=
|
17 |
dm.setup()
|
18 |
|
19 |
# Init model from datamodule's attributes
|
20 |
model = DRModel(
|
21 |
-
num_classes=dm.num_classes, learning_rate=3e-
|
22 |
)
|
23 |
|
24 |
# Init logger
|
@@ -32,14 +32,17 @@ checkpoint_callback = ModelCheckpoint(
|
|
32 |
dirpath="checkpoints",
|
33 |
)
|
34 |
|
|
|
|
|
|
|
35 |
# Init trainer
|
36 |
trainer = L.Trainer(
|
37 |
max_epochs=20,
|
38 |
accelerator="auto",
|
39 |
devices="auto",
|
40 |
logger=logger,
|
41 |
-
callbacks=[checkpoint_callback],
|
42 |
-
enable_checkpointing=True
|
43 |
)
|
44 |
|
45 |
# Pass the datamodule as arg to trainer.fit to override model hooks :)
|
|
|
13 |
|
14 |
|
15 |
# Init DataModule
|
16 |
+
dm = DRDataModule(batch_size=96, num_workers=8)
|
17 |
dm.setup()
|
18 |
|
19 |
# Init model from datamodule's attributes
|
20 |
model = DRModel(
|
21 |
+
num_classes=dm.num_classes, learning_rate=3e-5, class_weights=dm.class_weights
|
22 |
)
|
23 |
|
24 |
# Init logger
|
|
|
32 |
dirpath="checkpoints",
|
33 |
)
|
34 |
|
35 |
+
# Init LearningRateMonitor
|
36 |
+
lr_monitor = LearningRateMonitor(logging_interval="step")
|
37 |
+
|
38 |
# Init trainer
|
39 |
trainer = L.Trainer(
|
40 |
max_epochs=20,
|
41 |
accelerator="auto",
|
42 |
devices="auto",
|
43 |
logger=logger,
|
44 |
+
callbacks=[checkpoint_callback, lr_monitor],
|
45 |
+
enable_checkpointing=True,
|
46 |
)
|
47 |
|
48 |
# Pass the datamodule as arg to trainer.fit to override model hooks :)
|