log change in loc spec params

just to be sure they are being trained

src/ml_downscaling_emulator/score_sde_pytorch/run_lib.py

@@ -177,7 +177,8 @@ def train(config, workdir):
 
     if config.training.random_crop_size > 0:
       random_crop = torchvision.transforms.RandomCrop(config.training.random_crop_size)
-
+    if config.model.loc_spec_channels > 0:
+      before = state['location_params'].module.params[0,20,20:30].clone().detach()
     for epoch in range(initial_epoch, num_train_epochs + 1):
       state['epoch'] = epoch
       train_set_loss = 0.0
@@ -226,3 +227,9 @@ def train(config, workdir):
         checkpoint_path = os.path.join(checkpoint_dir, f"epoch_{state['epoch']}.pth")
         save_checkpoint(checkpoint_path, state)
         logging.info(f"epoch: {state['epoch']}, checkpoint saved to {checkpoint_path}")
+
+    if config.model.loc_spec_channels > 0:
+      after = state['location_params'].module.params[0,20,20:30].clone().detach()
+      logging.info(f"Before: {before}")
+      logging.info(f"After: {after}")
+      logging.info(f"Diff: {after-before}")