minimal updates

pvnet_summation/models/base_model.py

@@ -91,6 +91,8 @@ def __init__(
             )
         else:
             self.pvnet_output_shape = (317, self.pvnet_model.forecast_len)
+
+        self.use_weighted_loss = False
 
     def predict_pvnet_batch(self, batch):
         """Use PVNet model to create predictions for batch"""
@@ -184,6 +186,11 @@ def validation_step(self, batch: dict, batch_idx):
 
         losses = self._calculate_common_losses(y, y_hat)
         losses.update(self._calculate_val_losses(y, y_hat))
+
+        # Store these to make horizon accuracy plot
+        self._horizon_maes.append(
+            {i: losses[f"MAE_horizon/step_{i:03}"].cpu().numpy() for i in range(self.forecast_len)}
+        )
 
         logged_losses = {f"{k}/val": v for k, v in losses.items()}
 

pvnet_summation/training.py

@@ -149,6 +149,8 @@ def train(config: DictConfig) -> Optional[float]:
         for callback in callbacks:
             log.info(f"{callback}")
             if isinstance(callback, ModelCheckpoint):
+                # Need to call the .experiment property to initialise the logger
+                wandb_logger.experiment
                 callback.dirpath = "/".join(
                     callback.dirpath.split("/")[:-1] + [wandb_logger.version]
                 )

requirements.txt

@@ -1,5 +1,5 @@
-ocf_datapipes>=3.3.19
-pvnet>=3.0.25
+ocf_datapipes>=3.3.33
+pvnet>=3.0.45
 numpy
 pandas
 matplotlib

scripts/checkpoint_to_huggingface.py

@@ -1,5 +1,5 @@
 """Command line tool to push locally save model checkpoints to huggingface
-
+    
 use:
 python checkpoint_to_huggingface.py "path/to/model/checkpoints" \
     --local-path="~/tmp/this_model" \
@@ -56,9 +56,9 @@ def push_to_huggingface(
         # Only one epoch (best) saved per model
         files = glob.glob(f"{checkpoint_dir_path}/epoch*.ckpt")
         assert len(files) == 1
-        checkpoint = torch.load(files[0])
+        checkpoint = torch.load(files[0], map_location="cpu")
     else:
-        checkpoint = torch.load(f"{checkpoint_dir_path}/last.ckpt")
+        checkpoint = torch.load(f"{checkpoint_dir_path}/last.ckpt", map_location="cpu")
 
     model.load_state_dict(state_dict=checkpoint["state_dict"])
 
@@ -72,7 +72,8 @@ def push_to_huggingface(
     model.save_pretrained(
         model_output_dir,
         config=model_config,
-        wandb_model_code=wandb_id,
+        data_config=None,
+        wandb_ids=wandb_id,
         push_to_hub=push_to_hub,
         repo_id="openclimatefix/pvnet_v2_summation" if push_to_hub else None,
         card_template_path=(