Rooba
/
lightning
mirror of https://github.com/Lightning-AI/lightning.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix outdated docs follow up (#1266) 

* save_top_k

* num_gpus

* print_nan_grads fix leftovers

* undo doctest removal
This commit is contained in:
Adrian Wälchli 2020-03-28 16:30:57 +01:00 committed by GitHub
parent 12b39a74b4
commit 0b0180406b
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 6 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/weights_loading.rst
View File

@ -35,7 +35,7 @@ To modify the behavior of checkpointing pass in your own callback.
# DEFAULTS used by the Trainer # DEFAULTS used by the Trainer
checkpoint_callback = ModelCheckpoint( checkpoint_callback = ModelCheckpoint(
filepath=os.getcwd(), filepath=os.getcwd(),
save_best_only=True, save_top_k=True,
verbose=True, verbose=True,
monitor='val_loss', monitor='val_loss',
mode='min', mode='min',

2
pytorch_lightning/core/lightning.py
View File

@ -799,7 +799,7 @@ class LightningModule(ABC, GradInformation, ModelIO, ModelHooks):
Args: Args:
proc_rank: The current process rank within the node. proc_rank: The current process rank within the node.
world_size: Number of GPUs being use across all nodes. (num_nodes*nb_gpu_nodes). world_size: Number of GPUs being use across all nodes. (num_nodes * num_gpus).
Examples: Examples:
.. code-block:: python .. code-block:: python

9
pytorch_lightning/trainer/__init__.py
View File

@ -195,7 +195,7 @@ Example::
# default used by the Trainer # default used by the Trainer
checkpoint_callback = ModelCheckpoint( checkpoint_callback = ModelCheckpoint(
filepath=os.getcwd(), filepath=os.getcwd(),
save_best_only=True, save_top_k=True,
verbose=True, verbose=True,
monitor='val_loss', monitor='val_loss',
mode='min', mode='min',
@ -595,12 +595,11 @@ Example::
print_nan_grads print_nan_grads
^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^
Prints gradients with nan values .. warning:: .. deprecated:: 0.7.2.
Example:: Has no effect. When detected, NaN grads will be printed automatically.
Will remove 0.9.0.
# default used by the Trainer
trainer = Trainer(print_nan_grads=False)
process_position process_position
^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^