lightning/pytorch_lightning/plugins/precision/ipu_precision.py

53 lines
2.0 KiB
Python
Raw Normal View History

IPU Integration 5/5 (#7867) * Initial changes * Add broken example for now * Fix reference * Fix format * Code runs * Fixes * Clear up files * Add tests, helpers, fixes * Small cleanups * Refactors based on review * Swap to special tests * Add special tests * Add source * Cleanups * Add logic to attach/detach model from devices * Fixes for tests * Fixes for tests * Move earlier * Cleanups * Add check for nvcc * Add tests, cleanups * Fix errors * fix * Try condition * Add missing annotation * Clearer * Clearer message * Fix variable * Cleanups * Add comment * CHANGELOG.md * Add simple selection test * Remove special=True to see what happens * Fix test * Update tests/accelerators/test_ipu.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Convert ipu_cores -> ipus * Add typing, fail earlier * simplify precision * Add test, add helper * fix accum * Update pytorch_lightning/plugins/training_type/ipu.py Co-authored-by: thomas chaton <thomas@grid.ai> * Use stages * Make sure warning message returned * thorw error * Add more tests, use fs * add comment * Clean * Address feedback, add IPU tests * Fixes * Fix signature * Add types * Remove autoround * Add docstring * ipu_cores -> ipus * Add test, remove unnecessary precision set * Add optimizer test * Add precision back with test * Address code review * Change to probs * Move some of the asserts earlier Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: thomas chaton <thomas@grid.ai>
2021-06-11 15:07:04 +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.
from typing import Any, Optional, Union
from torch.nn import Module
from torch.optim import Optimizer
import pytorch_lightning as pl
from pytorch_lightning.plugins.precision.precision_plugin import PrecisionPlugin
from pytorch_lightning.utilities import GradClipAlgorithmType
from pytorch_lightning.utilities.exceptions import MisconfigurationException
2021-07-08 14:02:09 +00:00
from pytorch_lightning.utilities.model_helpers import is_overridden
from pytorch_lightning.utilities.warnings import WarningCache
warning_cache = WarningCache()
IPU Integration 5/5 (#7867) * Initial changes * Add broken example for now * Fix reference * Fix format * Code runs * Fixes * Clear up files * Add tests, helpers, fixes * Small cleanups * Refactors based on review * Swap to special tests * Add special tests * Add source * Cleanups * Add logic to attach/detach model from devices * Fixes for tests * Fixes for tests * Move earlier * Cleanups * Add check for nvcc * Add tests, cleanups * Fix errors * fix * Try condition * Add missing annotation * Clearer * Clearer message * Fix variable * Cleanups * Add comment * CHANGELOG.md * Add simple selection test * Remove special=True to see what happens * Fix test * Update tests/accelerators/test_ipu.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Convert ipu_cores -> ipus * Add typing, fail earlier * simplify precision * Add test, add helper * fix accum * Update pytorch_lightning/plugins/training_type/ipu.py Co-authored-by: thomas chaton <thomas@grid.ai> * Use stages * Make sure warning message returned * thorw error * Add more tests, use fs * add comment * Clean * Address feedback, add IPU tests * Fixes * Fix signature * Add types * Remove autoround * Add docstring * ipu_cores -> ipus * Add test, remove unnecessary precision set * Add optimizer test * Add precision back with test * Address code review * Change to probs * Move some of the asserts earlier Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: thomas chaton <thomas@grid.ai>
2021-06-11 15:07:04 +00:00
class IPUPrecisionPlugin(PrecisionPlugin):
def __init__(self, precision: int) -> None:
super().__init__()
self.precision = precision
def backward(self, model: "pl.LightningModule", *args: Any, **kwargs: Any) -> None:
if is_overridden("backward", model):
2021-07-08 14:02:09 +00:00
warning_cache.warn(
"You have overridden the `LightningModule.backward` hook but it will be ignored since IPUs handle"
" the backward logic internally."
)
IPU Integration 5/5 (#7867) * Initial changes * Add broken example for now * Fix reference * Fix format * Code runs * Fixes * Clear up files * Add tests, helpers, fixes * Small cleanups * Refactors based on review * Swap to special tests * Add special tests * Add source * Cleanups * Add logic to attach/detach model from devices * Fixes for tests * Fixes for tests * Move earlier * Cleanups * Add check for nvcc * Add tests, cleanups * Fix errors * fix * Try condition * Add missing annotation * Clearer * Clearer message * Fix variable * Cleanups * Add comment * CHANGELOG.md * Add simple selection test * Remove special=True to see what happens * Fix test * Update tests/accelerators/test_ipu.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Convert ipu_cores -> ipus * Add typing, fail earlier * simplify precision * Add test, add helper * fix accum * Update pytorch_lightning/plugins/training_type/ipu.py Co-authored-by: thomas chaton <thomas@grid.ai> * Use stages * Make sure warning message returned * thorw error * Add more tests, use fs * add comment * Clean * Address feedback, add IPU tests * Fixes * Fix signature * Add types * Remove autoround * Add docstring * ipu_cores -> ipus * Add test, remove unnecessary precision set * Add optimizer test * Add precision back with test * Address code review * Change to probs * Move some of the asserts earlier Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: thomas chaton <thomas@grid.ai>
2021-06-11 15:07:04 +00:00
def clip_gradients(
self,
optimizer: Optimizer,
clip_val: Union[int, float],
gradient_clip_algorithm: GradClipAlgorithmType = GradClipAlgorithmType.NORM,
model: Optional[Module] = None,
IPU Integration 5/5 (#7867) * Initial changes * Add broken example for now * Fix reference * Fix format * Code runs * Fixes * Clear up files * Add tests, helpers, fixes * Small cleanups * Refactors based on review * Swap to special tests * Add special tests * Add source * Cleanups * Add logic to attach/detach model from devices * Fixes for tests * Fixes for tests * Move earlier * Cleanups * Add check for nvcc * Add tests, cleanups * Fix errors * fix * Try condition * Add missing annotation * Clearer * Clearer message * Fix variable * Cleanups * Add comment * CHANGELOG.md * Add simple selection test * Remove special=True to see what happens * Fix test * Update tests/accelerators/test_ipu.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Convert ipu_cores -> ipus * Add typing, fail earlier * simplify precision * Add test, add helper * fix accum * Update pytorch_lightning/plugins/training_type/ipu.py Co-authored-by: thomas chaton <thomas@grid.ai> * Use stages * Make sure warning message returned * thorw error * Add more tests, use fs * add comment * Clean * Address feedback, add IPU tests * Fixes * Fix signature * Add types * Remove autoround * Add docstring * ipu_cores -> ipus * Add test, remove unnecessary precision set * Add optimizer test * Add precision back with test * Address code review * Change to probs * Move some of the asserts earlier Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: thomas chaton <thomas@grid.ai>
2021-06-11 15:07:04 +00:00
) -> None:
"""Clips the gradients."""
if clip_val is None or float(clip_val) <= 0:
IPU Integration 5/5 (#7867) * Initial changes * Add broken example for now * Fix reference * Fix format * Code runs * Fixes * Clear up files * Add tests, helpers, fixes * Small cleanups * Refactors based on review * Swap to special tests * Add special tests * Add source * Cleanups * Add logic to attach/detach model from devices * Fixes for tests * Fixes for tests * Move earlier * Cleanups * Add check for nvcc * Add tests, cleanups * Fix errors * fix * Try condition * Add missing annotation * Clearer * Clearer message * Fix variable * Cleanups * Add comment * CHANGELOG.md * Add simple selection test * Remove special=True to see what happens * Fix test * Update tests/accelerators/test_ipu.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Convert ipu_cores -> ipus * Add typing, fail earlier * simplify precision * Add test, add helper * fix accum * Update pytorch_lightning/plugins/training_type/ipu.py Co-authored-by: thomas chaton <thomas@grid.ai> * Use stages * Make sure warning message returned * thorw error * Add more tests, use fs * add comment * Clean * Address feedback, add IPU tests * Fixes * Fix signature * Add types * Remove autoround * Add docstring * ipu_cores -> ipus * Add test, remove unnecessary precision set * Add optimizer test * Add precision back with test * Address code review * Change to probs * Move some of the asserts earlier Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: thomas chaton <thomas@grid.ai>
2021-06-11 15:07:04 +00:00
return
raise MisconfigurationException("IPUs currently do not support clipping gradients.")