2428 Enhance docs for pure PyTorch loss, optimizer, etc

monai/README.md

@@ -14,13 +14,13 @@
 
 * **inferers**: defines model inference methods.
 
-* **losses**: classes defining loss functions.
+* **losses**: classes defining loss functions, which follow the pattern of `torch.nn.modules.loss`.
 
 * **metrics**: defines metric tracking types.
 
 * **networks**: contains network definitions, component definitions, and Pytorch specific utilities.
 
-* **optimizers**: classes defining optimizers.
+* **optimizers**: classes defining optimizers, which follow the pattern of `torch.optim`.
 
 * **transforms**: defines data transforms for preprocessing and postprocessing.
 

monai/engines/evaluator.py

@@ -130,7 +130,7 @@ class SupervisedEvaluator(Evaluator):
     Args:
         device: an object representing the device on which to run.
         val_data_loader: Ignite engine use data_loader to run, must be Iterable, typically be torch.DataLoader.
-        network: use the network to run model forward.
+        network: network to evaluate in the evaluator, should be regular PyTorch `torch.nn.Module`.
         epoch_length: number of iterations for one epoch, default to `len(val_data_loader)`.
         non_blocking: if True and this copy is between CPU and GPU, the copy may occur asynchronously
             with respect to the host. For other cases, this argument has no effect.
@@ -245,7 +245,7 @@ class EnsembleEvaluator(Evaluator):
         device: an object representing the device on which to run.
         val_data_loader: Ignite engine use data_loader to run, must be Iterable, typically be torch.DataLoader.
         epoch_length: number of iterations for one epoch, default to `len(val_data_loader)`.
-        networks: use the networks to run model forward in order.
+        network: networks to evaluate in order in the evaluator, should be regular PyTorch `torch.nn.Module`.
         pred_keys: the keys to store every prediction data.
             the length must exactly match the number of networks.
         non_blocking: if True and this copy is between CPU and GPU, the copy may occur asynchronously

monai/engines/trainer.py

@@ -60,9 +60,11 @@ class SupervisedTrainer(Trainer):
         device: an object representing the device on which to run.
         max_epochs: the total epoch number for trainer to run.
         train_data_loader: Ignite engine use data_loader to run, must be Iterable or torch.DataLoader.
-        network: to train with this network.
-        optimizer: the optimizer associated to the network.
-        loss_function: the loss function associated to the optimizer.
+        network: network to train in the trainer, should be regular PyTorch `torch.nn.Module`.
+        optimizer: the optimizer associated to the network, should be regular PyTorch optimizer from `torch.optim`
+            or its subclass.
+        loss_function: the loss function associated to the optimizer, should be regular PyTorch loss,
+            which inherit from `torch.nn.modules.loss`.
         epoch_length: number of iterations for one epoch, default to `len(train_data_loader)`.
         non_blocking: if True and this copy is between CPU and GPU, the copy may occur asynchronously
             with respect to the host. For other cases, this argument has no effect.