2020-07-26 02:55:09 +00:00
|
|
|
# Copyright The PyTorch Lightning team.
|
|
|
|
#
|
|
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
# you may not use this file except in compliance with the License.
|
|
|
|
# You may obtain a copy of the License at
|
|
|
|
#
|
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
#
|
|
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
# See the License for the specific language governing permissions and
|
|
|
|
# limitations under the License.
|
2020-08-24 23:59:46 +00:00
|
|
|
import torch
|
2020-07-26 02:55:09 +00:00
|
|
|
from pytorch_lightning.utilities.exceptions import MisconfigurationException
|
2020-08-24 21:50:47 +00:00
|
|
|
from pytorch_lightning.accelerators.base_backend import Accelerator
|
2020-08-24 23:59:46 +00:00
|
|
|
from pytorch_lightning.utilities import AMPType
|
2020-07-26 02:55:09 +00:00
|
|
|
|
|
|
|
|
2020-08-24 21:50:47 +00:00
|
|
|
class CPUBackend(Accelerator):
|
2020-07-26 02:55:09 +00:00
|
|
|
|
|
|
|
def __init__(self, trainer):
|
2020-08-24 21:50:47 +00:00
|
|
|
super().__init__(trainer)
|
2020-07-26 02:55:09 +00:00
|
|
|
|
|
|
|
def setup(self, model):
|
|
|
|
# run through amp wrapper
|
2020-08-13 14:03:13 +00:00
|
|
|
if self.trainer.amp_backend:
|
2020-07-26 02:55:09 +00:00
|
|
|
raise MisconfigurationException('amp + cpu is not supported. Please use a GPU option')
|
|
|
|
|
|
|
|
# call setup after the ddp process has connected
|
2020-08-02 00:17:57 +00:00
|
|
|
self.trainer.call_setup_hook(model)
|
2020-07-26 02:55:09 +00:00
|
|
|
|
|
|
|
# CHOOSE OPTIMIZER
|
|
|
|
# allow for lr schedulers as well
|
|
|
|
optimizers, lr_schedulers, optimizer_frequencies = self.trainer.init_optimizers(model)
|
|
|
|
self.trainer.optimizers = optimizers
|
|
|
|
self.trainer.lr_schedulers = lr_schedulers
|
|
|
|
self.trainer.optimizer_frequencies = optimizer_frequencies
|
2020-08-27 00:03:09 +00:00
|
|
|
self.trainer.model = model
|
2020-07-26 02:55:09 +00:00
|
|
|
|
2020-08-27 00:03:09 +00:00
|
|
|
def train(self):
|
|
|
|
model = self.trainer.model
|
2020-08-31 22:06:11 +00:00
|
|
|
results = self.trainer.setup_training(model)
|
2020-07-26 02:55:09 +00:00
|
|
|
return results
|
2020-08-24 16:31:40 +00:00
|
|
|
|
|
|
|
def training_step(self, args):
|
2020-08-24 23:59:46 +00:00
|
|
|
if self.trainer.amp_backend == AMPType.NATIVE:
|
|
|
|
with torch.cuda.amp.autocast():
|
|
|
|
output = self.trainer.model.training_step(*args)
|
|
|
|
else:
|
|
|
|
output = self.trainer.model.training_step(*args)
|
|
|
|
return output
|
2020-08-24 16:31:40 +00:00
|
|
|
|
|
|
|
def validation_step(self, args):
|
2020-08-25 00:26:39 +00:00
|
|
|
if self.trainer.amp_backend == AMPType.NATIVE:
|
|
|
|
with torch.cuda.amp.autocast():
|
|
|
|
output = self.trainer.model.validation_step(*args)
|
|
|
|
else:
|
|
|
|
output = self.trainer.model.validation_step(*args)
|
|
|
|
return output
|
2020-08-24 16:31:40 +00:00
|
|
|
|
|
|
|
def test_step(self, args):
|
2020-08-25 00:26:39 +00:00
|
|
|
if self.trainer.amp_backend == AMPType.NATIVE:
|
|
|
|
with torch.cuda.amp.autocast():
|
|
|
|
output = self.trainer.model.test_step(*args)
|
|
|
|
else:
|
|
|
|
output = self.trainer.model.test_step(*args)
|
|
|
|
return output
|