v1.2.4 & chnagelog

Remove import Fix runif import Increment version Update CHANGELOG.md
Lightning-AI · Mar 16, 2021 · 7d4db76 · 7d4db76
1 parent 293ff2a
commit 7d4db76
Show file tree

Hide file tree

Showing 4 changed files with 16 additions and 28 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -51,9 +51,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Changed `setup()` and `teardown()` stage argument to take any of `{fit,validate,test,predict}` ([#6386](https://github.com/PyTorchLightning/pytorch-lightning/pull/6386))
 
 
-- Changed the default of `find_unused_parameters` back to `True` in DDP and DDP Spawn ([#6438](https://github.com/PyTorchLightning/pytorch-lightning/pull/6438))
-
-
 ### Deprecated
 
 - `period` has been deprecated in favor of `every_n_val_epochs` in the `ModelCheckpoint` callback ([#6146](https://github.com/PyTorchLightning/pytorch-lightning/pull/6146))
@@ -110,43 +107,36 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Fixed `ModelCheckpoint(save_top_k=0, save_last=True)` not saving the `last` checkpoint ([#6136](https://github.com/PyTorchLightning/pytorch-lightning/pull/6136))
 
 
-- Expose DeepSpeed loss parameters to allow users to fix loss instability ([#6115](https://github.com/PyTorchLightning/pytorch-lightning/pull/6115))
-
-
 - Fixed duplicate logs appearing in console when using the python logging module ([#5509](https://github.com/PyTorchLightning/pytorch-lightning/pull/5509), [#6275](https://github.com/PyTorchLightning/pytorch-lightning/pull/6275))
 
 
-- Fixed DP reduction with collection ([#6324](https://github.com/PyTorchLightning/pytorch-lightning/pull/6324))
-
-
 - Fixed `.teardown(stage='fit')` getting called during `trainer.test` ([#6386](https://github.com/PyTorchLightning/pytorch-lightning/pull/6386))
 
 
 - Fixed `.on_fit_{start,end}()` getting called during `trainer.test` ([#6386](https://github.com/PyTorchLightning/pytorch-lightning/pull/6386))
 
 
-- Fixed an issue where the tuner would not tune the learning rate if also tuning the batch size ([#4688](https://github.com/PyTorchLightning/pytorch-lightning/pull/4688))
+- Fixed LightningModule `all_gather` on cpu tensors ([#6416](https://github.com/PyTorchLightning/pytorch-lightning/pull/6416))
 
 
-- Fixed broacast to use PyTorch `broadcast_object_list` and add `reduce_decision` ([#6410](https://github.com/PyTorchLightning/pytorch-lightning/pull/6410))
+## [1.2.4] - 2021-03-16
 
+### Changed
 
-- Fixed logger creating directory structure too early in DDP ([#6380](https://github.com/PyTorchLightning/pytorch-lightning/pull/6380))
+- Changed the default of `find_unused_parameters` back to `True` in DDP and DDP Spawn ([#6438](https://github.com/PyTorchLightning/pytorch-lightning/pull/6438))
 
+### Fixed
 
+- Expose DeepSpeed loss parameters to allow users to fix loss instability ([#6115](https://github.com/PyTorchLightning/pytorch-lightning/pull/6115))
+- Fixed DP reduction with collection ([#6324](https://github.com/PyTorchLightning/pytorch-lightning/pull/6324))
+- Fixed an issue where the tuner would not tune the learning rate if also tuning the batch size ([#4688](https://github.com/PyTorchLightning/pytorch-lightning/pull/4688))
+- Fixed broadcast to use PyTorch `broadcast_object_list` and add `reduce_decision` ([#6410](https://github.com/PyTorchLightning/pytorch-lightning/pull/6410))
+- Fixed logger creating directory structure too early in DDP ([#6380](https://github.com/PyTorchLightning/pytorch-lightning/pull/6380))
 - Fixed DeepSpeed additional memory use on rank 0 when default device not set early enough ([#6460](https://github.com/PyTorchLightning/pytorch-lightning/pull/6460))
-
-
-- Fixed LightningModule `all_gather` on cpu tensors ([#6416](https://github.com/PyTorchLightning/pytorch-lightning/pull/6416))
-
-
 - Fixed `DummyLogger.log_hyperparams` raising a `TypeError` when running with `fast_dev_run=True` ([#6398](https://github.com/PyTorchLightning/pytorch-lightning/pull/6398))
-
-
 - Fixed an issue with `Tuner.scale_batch_size` not finding the batch size attribute in the datamodule ([#5968](https://github.com/PyTorchLightning/pytorch-lightning/pull/5968))
-
-
 - Fixed an exception in the layer summary when the model contains torch.jit scripted submodules ([#6511](https://github.com/PyTorchLightning/pytorch-lightning/pull/6511))
+- Fixed when Train loop config was run during `Trainer.predict` ([#6541](https://github.com/PyTorchLightning/pytorch-lightning/pull/6541))
 
 
 ## [1.2.3] - 2021-03-09
@@ -166,9 +156,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Fixed `Trainer` not resetting `lightning_optimizers` when calling `Trainer.fit()` multiple times ([#6372](https://github.com/PyTorchLightning/pytorch-lightning/pull/6372))
 
 
-- Fixed `DummyLogger.log_hyperparams` raising a `TypeError` when running with `fast_dev_run=True` ([#6398](https://github.com/PyTorchLightning/pytorch-lightning/pull/6398))
-
-
 ## [1.2.2] - 2021-03-02
 
 ### Added

diff --git a/pytorch_lightning/__init__.py b/pytorch_lightning/__init__.py
@@ -5,7 +5,7 @@
 import time
 
 _this_year = time.strftime("%Y")
-__version__ = '1.2.3'
+__version__ = '1.2.4'
 __author__ = 'William Falcon et al.'
 __author_email__ = 'waf2107@columbia.edu'
 __license__ = 'Apache-2.0'

diff --git a/pytorch_lightning/trainer/connectors/logger_connector/epoch_result_store.py b/pytorch_lightning/trainer/connectors/logger_connector/epoch_result_store.py
@@ -13,7 +13,6 @@
 # limitations under the License.
 from collections import defaultdict
 from typing import Any, Dict, List, Optional, Tuple
-from weakref import proxy
 
 import torch
 

diff --git a/tests/checkpointing/test_checkpoint_callback_frequency.py b/tests/checkpointing/test_checkpoint_callback_frequency.py
@@ -19,7 +19,6 @@
 
 from pytorch_lightning import callbacks, seed_everything, Trainer
 from tests.helpers import BoringModel
-from tests.helpers.runif import RunIf
 
 
 @mock.patch.dict(os.environ, {"PL_DEV_DEBUG": "1"})
@@ -102,7 +101,10 @@ def training_step(self, batch, batch_idx):
 
 
 @mock.patch('torch.save')
-@RunIf(special=True, min_gpus=2)
+@pytest.mark.skipif(torch.cuda.device_count() < 2, reason="test requires multi-GPU machine")
+@pytest.mark.skipif(
+    not os.getenv("PL_RUNNING_SPECIAL_TESTS", '0') == '1', reason="test should be run outside of pytest"
+)
 @pytest.mark.parametrize(['k', 'epochs', 'val_check_interval', 'expected'], [(1, 1, 1.0, 1), (2, 2, 0.3, 5)])
 def test_top_k_ddp(save_mock, tmpdir, k, epochs, val_check_interval, expected):