lightning/tests/conftest.py

193 lines
6.4 KiB
Python
Raw Normal View History

2020-12-15 17:59:13 +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.
PoC: Accelerator refactor (#5743) * restoring the result from subprocess * fix queue.get() order for results * add missing "block_backward_sync" context manager * add missing "block_backward_sync" context manager * fix sync_batchnorm * fix supported gpu-ids for tuple * fix clip gradients and inf recursion * accelerator selection: added cluster_environment plugin * fix torchelastic test * fix reduce early stopping decision for DDP * fix tests: callbacks, conversion to lightning optimizer * fix lightning optimizer does not pickle * fix setting benchmark and deterministic option * fix slurm amp test * fix prepare_data test and determine node_rank * fix retrieving last path when testing * remove obsolete plugin argument * fix test: test_trainer_config * fix torchscript tests * fix trainer.model access * move properties * fix test_transfer_batch_hook * fix auto_select_gpus * fix omegaconf test * fix test that needs to simulate slurm ddp * add horovod plugin * fix test with named arguments * clean up whitespace * fix datamodules test * remove old accelerators * fix naming * move old plugins * move to plugins * create precision subpackage * create training_type subpackage * fix all new import errors * fix wrong arguments order passed to test * fix LR finder * Added sharded training type and amp plugin * Move clip grad to precision plugin * Added sharded spawn, select accelerators based on distributed_backend + enable custom fp16 plugin automatically * Fix import issue, attempting to fix tests * Fix initial test * Reflect hook logic from master, should wrap model after move to device * Optional state consolidation, since master has optimizers not wrapped * change attribute for instance test * reset optimizers optimizers are not used in main process, so state would be wrong. * legacy * imports in accel * legacy2 * trainer imports * fix import errors after rebase * move hook to new setup location * provide unwrapping logic * fix trainer callback system * added ddp2 implementation * fix imports .legacy * move plugins * restore legacy * drop test.py from root * add tpu accelerator and plugins * fixes * fix lightning optimizer merge * reset bugreportmodel * unwrapping * step routing forward * model access * unwrap * opt * integrate distrib_type * sync changes * sync * fixes * add forgotten generators * add missing logic * update * import * missed imports * import fixes * isort * mv f * changelog * format * move helper to parallel plugin * d * add world size * clean up * duplicate * activate ddp_sharded and tpu * set nvidia flags * remove unused colab var * use_tpu <-> on_tpu attrs * make some ddp_cpu and clusterplugin tests pass * Ref/accelerator connector (#5742) * final cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * connector cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * trainer cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * accelerator cleanup + missing logic in accelerator connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add missing changes to callbacks Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * reflect accelerator changes to lightning module Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * clean cluster envs Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * cleanup plugins Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add broadcasting Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * yapf * remove plugin connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * plugins * manual optimization * update optimizer routing * add rank to torchelastic * fix memory mixed precision * setstate on trainer for pickling in ddp spawn * add predict method * add back commented accelerator code * adapt test for sync_batch_norm to new plugin * fix deprecated tests * fix ddp cpu choice when no num_processes are given * yapf format * skip a memory test that cannot pass anymore * fix pickle error in spawn plugin * x * avoid * x * fix cyclic import in docs build * add support for sharded * update typing * add sharded and sharded_spawn to distributed types * make unwrap model default * refactor LightningShardedDataParallel similar to LightningDistributedDataParallel * update sharded spawn to reflect changes * update sharded to reflect changes * Merge 1.1.5 changes * fix merge * fix merge * yapf isort * fix merge * yapf isort * fix indentation in test * copy over reinit scheduler implementation from dev1.2 * fix apex tracking calls with dev_debugger * reduce diff to dev1.2, clean up * fix trainer config test when gpus>0 and num_processes >0 and ddp_cpu * sort plugin tests legacy/new * fix error handling for amp on cpu * fix merge fix merge fix merge * [Feat] Resolve manual_backward (#5837) * resolve manual_backward * resolve flake8 * update * resolve for ddp_spawn * resolve flake8 * resolve flake8 * resolve flake8 Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * fix tests/accelerator tests on cpu * [BugFix] Resolve manual optimization (#5852) * resolve manual_optimization * update * update Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * Remove copy trainer parameters to happen earlier within the loop and add safe guard to get ref model (#5856) * resovle a bug * Accelerator refactor sharded rpc (#5854) * rpc branch * merge * update handling of rpc * make devices etc. Optional in RPC * set devices etc. later if necessary * remove devices from sequential * make devices optional in rpc * fix import * uncomment everything * fix cluster selection Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * resolve bug * fix assert in rpc test * resolve a test * fix docs compilation * accelerator refactor - fix for sharded parity test (#5866) * fix memory issue with ddp_spawn * x x x x x x x x x * x * Remove DDP2 as this does not apply * Add missing pre optimizer hook to ensure lambda closure is called * fix apex docstring * [accelerator][BugFix] Resolve some test for 1 gpu (#5863) * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * update * resolve flake8 * update * update * update * update * update * all_gather * update * make plugins work, add misconfig for RPC * update * update * remove breaking test * resolve some tests * resolve flake8 * revert to ddp_spawn Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de> * yapf isort * resolve flake8 * fix apex doctests * fix apex doctests 2 * resolve docs * update drone * clean env * update * update * update * update * merge * Fix RPC related tests, clean out old API, update for new accelerator API [skip ci] (#5881) * Fix RPC related tests, clean out old API, update for new accelerator API * Move tests out of legacy folder, update paths and names * Update test_remove_1-4.py * Expose properties for tpu cores/gpus/num_gpus * Add root GPU property * Move properties to properties.py * move tests that were previously in drone * Fix root GPU property (#5908) * Move root GPU to property, remove horovod set as this is handled in horovod plugin, ensure we mock correctly to set GPU accelerator * Add missing tests back * fix best model path transfer when no checkpoint callback available * Fix setup hook order [wip] (#5858) * Call trainer setup hook before accelerator setup * Add test case * add new test * typo * fix callback order in test Co-authored-by: tchaton <thomas@grid.ai> Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * rename ddp sequential -> rpc sequential for special test * revert * fix stupid merge problem * Use property in connector for sampler (#5913) * merge the import conflicts * fix spawning of processes in slurm * [wip] Fix some bugs for TPU [skip ci] (#5878) * fixed for single tpu * fixed spawn * fixed spawn * update * update * wip * resolve bugs * resolve bug * update on comment * removed decorator * resolve comments * set to 4 * update * update * need cleaning * update * update * update * resolve flake8 * resolve bugs * exclude broadcast * resolve bugs * change test * update * update * skip if meet fails * properly raise trace * update * add catch * wrap test * resolve typo * update * typo Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> * resolve some tests * update * fix imports * update * resolve flake8 * update azure pipeline * skip a sharded test on cpu that requires a gpu * resolve tpus * resolve bug * resolve flake8 * update * updat utils * revert permission change on files * suggestions from carlos Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting changes * remove incomplete comment * Update pytorch_lightning/accelerators/__init__.py Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting change * add types * warn 1.7 ddp manual backward only if ddp kwarg unset * yapf + isort * pep8 unused imports * fix cyclic import in docs * Apply suggestions from code review * typer in accelerator.py * typo * Apply suggestions from code review * formatting * update on comments * update typo * Update pytorch_lightning/trainer/properties.py Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * update * suggestion from code review * suggestion from code review Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> Co-authored-by: SeanNaren <sean@grid.ai> Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz> Co-authored-by: chaton <thomas@grid.ai> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Sean Naren <sean.narenthiran@gmail.com> Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-12 20:48:56 +00:00
import os
import signal
import threading
2021-09-29 17:08:26 +00:00
from functools import partial
from http.server import SimpleHTTPRequestHandler
from pathlib import Path
import pytest
Support `DDPPlugin` to be used on CPU (#6208) * Skip test due to 'Python bus error' * Debug NCCL * Remove NCCL_DEBUG statement * Revert "Skip test due to 'Python bus error'" This reverts commit e0a3e8785d2fecd63667da433a648f958d60ef89. * fix * add test * changelog * yapf * patch os environ * make a special test * destroy pg * debug * revert * revert * problematic test * skip * try the fixture * test * update sensitive test * update changelog * remove comment * update wrong test * update test name * parameterization * Revert "parameterization" This reverts commit b0542f43f59c5ce66800883b5e2f0c66a97408cc. * remove conftest * ignore test * teardown * fix merge * deep speed parameterization * uncomment test * update chlog * update changelog * split tests * update test update test update test update test * update test comments * unroll test * unroll test * unroll test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * increase shm * sudo * unroll ipu * Revert "sudo" This reverts commit 6cc68c1478c151caee86b8a0f8ded16b62a9c8ea. * Revert "increase shm" This reverts commit 8c2716348330dbc3c6b3685647f435127b870d79. * x * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * find guilty test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * POPTORCH_WAIT_FOR_IPU=1 * move test * redo parameterize for ipu * de-comment test * move chlog * Update tests/accelerators/test_accelerator_connector.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Update tests/accelerators/test_accelerator_connector.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
2021-07-02 11:00:24 +00:00
import torch.distributed
from pytorch_lightning.plugins.environments.lightning_environment import find_free_network_port
from pytorch_lightning.trainer.connectors.signal_connector import SignalConnector
from pytorch_lightning.utilities.imports import _IS_WINDOWS, _TORCH_GREATER_EQUAL_1_8
from tests import _PATH_DATASETS
@pytest.fixture(scope="session")
def datadir():
return Path(_PATH_DATASETS)
@pytest.fixture(scope="function", autouse=True)
def preserve_global_rank_variable():
"""Ensures that the rank_zero_only.rank global variable gets reset in each test."""
from pytorch_lightning.utilities.rank_zero import rank_zero_only
rank = getattr(rank_zero_only, "rank", None)
yield
if rank is not None:
setattr(rank_zero_only, "rank", rank)
PoC: Accelerator refactor (#5743) * restoring the result from subprocess * fix queue.get() order for results * add missing "block_backward_sync" context manager * add missing "block_backward_sync" context manager * fix sync_batchnorm * fix supported gpu-ids for tuple * fix clip gradients and inf recursion * accelerator selection: added cluster_environment plugin * fix torchelastic test * fix reduce early stopping decision for DDP * fix tests: callbacks, conversion to lightning optimizer * fix lightning optimizer does not pickle * fix setting benchmark and deterministic option * fix slurm amp test * fix prepare_data test and determine node_rank * fix retrieving last path when testing * remove obsolete plugin argument * fix test: test_trainer_config * fix torchscript tests * fix trainer.model access * move properties * fix test_transfer_batch_hook * fix auto_select_gpus * fix omegaconf test * fix test that needs to simulate slurm ddp * add horovod plugin * fix test with named arguments * clean up whitespace * fix datamodules test * remove old accelerators * fix naming * move old plugins * move to plugins * create precision subpackage * create training_type subpackage * fix all new import errors * fix wrong arguments order passed to test * fix LR finder * Added sharded training type and amp plugin * Move clip grad to precision plugin * Added sharded spawn, select accelerators based on distributed_backend + enable custom fp16 plugin automatically * Fix import issue, attempting to fix tests * Fix initial test * Reflect hook logic from master, should wrap model after move to device * Optional state consolidation, since master has optimizers not wrapped * change attribute for instance test * reset optimizers optimizers are not used in main process, so state would be wrong. * legacy * imports in accel * legacy2 * trainer imports * fix import errors after rebase * move hook to new setup location * provide unwrapping logic * fix trainer callback system * added ddp2 implementation * fix imports .legacy * move plugins * restore legacy * drop test.py from root * add tpu accelerator and plugins * fixes * fix lightning optimizer merge * reset bugreportmodel * unwrapping * step routing forward * model access * unwrap * opt * integrate distrib_type * sync changes * sync * fixes * add forgotten generators * add missing logic * update * import * missed imports * import fixes * isort * mv f * changelog * format * move helper to parallel plugin * d * add world size * clean up * duplicate * activate ddp_sharded and tpu * set nvidia flags * remove unused colab var * use_tpu <-> on_tpu attrs * make some ddp_cpu and clusterplugin tests pass * Ref/accelerator connector (#5742) * final cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * connector cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * trainer cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * accelerator cleanup + missing logic in accelerator connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add missing changes to callbacks Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * reflect accelerator changes to lightning module Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * clean cluster envs Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * cleanup plugins Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add broadcasting Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * yapf * remove plugin connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * plugins * manual optimization * update optimizer routing * add rank to torchelastic * fix memory mixed precision * setstate on trainer for pickling in ddp spawn * add predict method * add back commented accelerator code * adapt test for sync_batch_norm to new plugin * fix deprecated tests * fix ddp cpu choice when no num_processes are given * yapf format * skip a memory test that cannot pass anymore * fix pickle error in spawn plugin * x * avoid * x * fix cyclic import in docs build * add support for sharded * update typing * add sharded and sharded_spawn to distributed types * make unwrap model default * refactor LightningShardedDataParallel similar to LightningDistributedDataParallel * update sharded spawn to reflect changes * update sharded to reflect changes * Merge 1.1.5 changes * fix merge * fix merge * yapf isort * fix merge * yapf isort * fix indentation in test * copy over reinit scheduler implementation from dev1.2 * fix apex tracking calls with dev_debugger * reduce diff to dev1.2, clean up * fix trainer config test when gpus>0 and num_processes >0 and ddp_cpu * sort plugin tests legacy/new * fix error handling for amp on cpu * fix merge fix merge fix merge * [Feat] Resolve manual_backward (#5837) * resolve manual_backward * resolve flake8 * update * resolve for ddp_spawn * resolve flake8 * resolve flake8 * resolve flake8 Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * fix tests/accelerator tests on cpu * [BugFix] Resolve manual optimization (#5852) * resolve manual_optimization * update * update Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * Remove copy trainer parameters to happen earlier within the loop and add safe guard to get ref model (#5856) * resovle a bug * Accelerator refactor sharded rpc (#5854) * rpc branch * merge * update handling of rpc * make devices etc. Optional in RPC * set devices etc. later if necessary * remove devices from sequential * make devices optional in rpc * fix import * uncomment everything * fix cluster selection Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * resolve bug * fix assert in rpc test * resolve a test * fix docs compilation * accelerator refactor - fix for sharded parity test (#5866) * fix memory issue with ddp_spawn * x x x x x x x x x * x * Remove DDP2 as this does not apply * Add missing pre optimizer hook to ensure lambda closure is called * fix apex docstring * [accelerator][BugFix] Resolve some test for 1 gpu (#5863) * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * update * resolve flake8 * update * update * update * update * update * all_gather * update * make plugins work, add misconfig for RPC * update * update * remove breaking test * resolve some tests * resolve flake8 * revert to ddp_spawn Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de> * yapf isort * resolve flake8 * fix apex doctests * fix apex doctests 2 * resolve docs * update drone * clean env * update * update * update * update * merge * Fix RPC related tests, clean out old API, update for new accelerator API [skip ci] (#5881) * Fix RPC related tests, clean out old API, update for new accelerator API * Move tests out of legacy folder, update paths and names * Update test_remove_1-4.py * Expose properties for tpu cores/gpus/num_gpus * Add root GPU property * Move properties to properties.py * move tests that were previously in drone * Fix root GPU property (#5908) * Move root GPU to property, remove horovod set as this is handled in horovod plugin, ensure we mock correctly to set GPU accelerator * Add missing tests back * fix best model path transfer when no checkpoint callback available * Fix setup hook order [wip] (#5858) * Call trainer setup hook before accelerator setup * Add test case * add new test * typo * fix callback order in test Co-authored-by: tchaton <thomas@grid.ai> Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * rename ddp sequential -> rpc sequential for special test * revert * fix stupid merge problem * Use property in connector for sampler (#5913) * merge the import conflicts * fix spawning of processes in slurm * [wip] Fix some bugs for TPU [skip ci] (#5878) * fixed for single tpu * fixed spawn * fixed spawn * update * update * wip * resolve bugs * resolve bug * update on comment * removed decorator * resolve comments * set to 4 * update * update * need cleaning * update * update * update * resolve flake8 * resolve bugs * exclude broadcast * resolve bugs * change test * update * update * skip if meet fails * properly raise trace * update * add catch * wrap test * resolve typo * update * typo Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> * resolve some tests * update * fix imports * update * resolve flake8 * update azure pipeline * skip a sharded test on cpu that requires a gpu * resolve tpus * resolve bug * resolve flake8 * update * updat utils * revert permission change on files * suggestions from carlos Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting changes * remove incomplete comment * Update pytorch_lightning/accelerators/__init__.py Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting change * add types * warn 1.7 ddp manual backward only if ddp kwarg unset * yapf + isort * pep8 unused imports * fix cyclic import in docs * Apply suggestions from code review * typer in accelerator.py * typo * Apply suggestions from code review * formatting * update on comments * update typo * Update pytorch_lightning/trainer/properties.py Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * update * suggestion from code review * suggestion from code review Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> Co-authored-by: SeanNaren <sean@grid.ai> Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz> Co-authored-by: chaton <thomas@grid.ai> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Sean Naren <sean.narenthiran@gmail.com> Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-12 20:48:56 +00:00
@pytest.fixture(scope="function", autouse=True)
def restore_env_variables():
"""Ensures that environment variables set during the test do not leak out."""
PoC: Accelerator refactor (#5743) * restoring the result from subprocess * fix queue.get() order for results * add missing "block_backward_sync" context manager * add missing "block_backward_sync" context manager * fix sync_batchnorm * fix supported gpu-ids for tuple * fix clip gradients and inf recursion * accelerator selection: added cluster_environment plugin * fix torchelastic test * fix reduce early stopping decision for DDP * fix tests: callbacks, conversion to lightning optimizer * fix lightning optimizer does not pickle * fix setting benchmark and deterministic option * fix slurm amp test * fix prepare_data test and determine node_rank * fix retrieving last path when testing * remove obsolete plugin argument * fix test: test_trainer_config * fix torchscript tests * fix trainer.model access * move properties * fix test_transfer_batch_hook * fix auto_select_gpus * fix omegaconf test * fix test that needs to simulate slurm ddp * add horovod plugin * fix test with named arguments * clean up whitespace * fix datamodules test * remove old accelerators * fix naming * move old plugins * move to plugins * create precision subpackage * create training_type subpackage * fix all new import errors * fix wrong arguments order passed to test * fix LR finder * Added sharded training type and amp plugin * Move clip grad to precision plugin * Added sharded spawn, select accelerators based on distributed_backend + enable custom fp16 plugin automatically * Fix import issue, attempting to fix tests * Fix initial test * Reflect hook logic from master, should wrap model after move to device * Optional state consolidation, since master has optimizers not wrapped * change attribute for instance test * reset optimizers optimizers are not used in main process, so state would be wrong. * legacy * imports in accel * legacy2 * trainer imports * fix import errors after rebase * move hook to new setup location * provide unwrapping logic * fix trainer callback system * added ddp2 implementation * fix imports .legacy * move plugins * restore legacy * drop test.py from root * add tpu accelerator and plugins * fixes * fix lightning optimizer merge * reset bugreportmodel * unwrapping * step routing forward * model access * unwrap * opt * integrate distrib_type * sync changes * sync * fixes * add forgotten generators * add missing logic * update * import * missed imports * import fixes * isort * mv f * changelog * format * move helper to parallel plugin * d * add world size * clean up * duplicate * activate ddp_sharded and tpu * set nvidia flags * remove unused colab var * use_tpu <-> on_tpu attrs * make some ddp_cpu and clusterplugin tests pass * Ref/accelerator connector (#5742) * final cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * connector cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * trainer cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * accelerator cleanup + missing logic in accelerator connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add missing changes to callbacks Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * reflect accelerator changes to lightning module Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * clean cluster envs Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * cleanup plugins Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add broadcasting Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * yapf * remove plugin connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * plugins * manual optimization * update optimizer routing * add rank to torchelastic * fix memory mixed precision * setstate on trainer for pickling in ddp spawn * add predict method * add back commented accelerator code * adapt test for sync_batch_norm to new plugin * fix deprecated tests * fix ddp cpu choice when no num_processes are given * yapf format * skip a memory test that cannot pass anymore * fix pickle error in spawn plugin * x * avoid * x * fix cyclic import in docs build * add support for sharded * update typing * add sharded and sharded_spawn to distributed types * make unwrap model default * refactor LightningShardedDataParallel similar to LightningDistributedDataParallel * update sharded spawn to reflect changes * update sharded to reflect changes * Merge 1.1.5 changes * fix merge * fix merge * yapf isort * fix merge * yapf isort * fix indentation in test * copy over reinit scheduler implementation from dev1.2 * fix apex tracking calls with dev_debugger * reduce diff to dev1.2, clean up * fix trainer config test when gpus>0 and num_processes >0 and ddp_cpu * sort plugin tests legacy/new * fix error handling for amp on cpu * fix merge fix merge fix merge * [Feat] Resolve manual_backward (#5837) * resolve manual_backward * resolve flake8 * update * resolve for ddp_spawn * resolve flake8 * resolve flake8 * resolve flake8 Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * fix tests/accelerator tests on cpu * [BugFix] Resolve manual optimization (#5852) * resolve manual_optimization * update * update Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * Remove copy trainer parameters to happen earlier within the loop and add safe guard to get ref model (#5856) * resovle a bug * Accelerator refactor sharded rpc (#5854) * rpc branch * merge * update handling of rpc * make devices etc. Optional in RPC * set devices etc. later if necessary * remove devices from sequential * make devices optional in rpc * fix import * uncomment everything * fix cluster selection Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * resolve bug * fix assert in rpc test * resolve a test * fix docs compilation * accelerator refactor - fix for sharded parity test (#5866) * fix memory issue with ddp_spawn * x x x x x x x x x * x * Remove DDP2 as this does not apply * Add missing pre optimizer hook to ensure lambda closure is called * fix apex docstring * [accelerator][BugFix] Resolve some test for 1 gpu (#5863) * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * update * resolve flake8 * update * update * update * update * update * all_gather * update * make plugins work, add misconfig for RPC * update * update * remove breaking test * resolve some tests * resolve flake8 * revert to ddp_spawn Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de> * yapf isort * resolve flake8 * fix apex doctests * fix apex doctests 2 * resolve docs * update drone * clean env * update * update * update * update * merge * Fix RPC related tests, clean out old API, update for new accelerator API [skip ci] (#5881) * Fix RPC related tests, clean out old API, update for new accelerator API * Move tests out of legacy folder, update paths and names * Update test_remove_1-4.py * Expose properties for tpu cores/gpus/num_gpus * Add root GPU property * Move properties to properties.py * move tests that were previously in drone * Fix root GPU property (#5908) * Move root GPU to property, remove horovod set as this is handled in horovod plugin, ensure we mock correctly to set GPU accelerator * Add missing tests back * fix best model path transfer when no checkpoint callback available * Fix setup hook order [wip] (#5858) * Call trainer setup hook before accelerator setup * Add test case * add new test * typo * fix callback order in test Co-authored-by: tchaton <thomas@grid.ai> Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * rename ddp sequential -> rpc sequential for special test * revert * fix stupid merge problem * Use property in connector for sampler (#5913) * merge the import conflicts * fix spawning of processes in slurm * [wip] Fix some bugs for TPU [skip ci] (#5878) * fixed for single tpu * fixed spawn * fixed spawn * update * update * wip * resolve bugs * resolve bug * update on comment * removed decorator * resolve comments * set to 4 * update * update * need cleaning * update * update * update * resolve flake8 * resolve bugs * exclude broadcast * resolve bugs * change test * update * update * skip if meet fails * properly raise trace * update * add catch * wrap test * resolve typo * update * typo Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> * resolve some tests * update * fix imports * update * resolve flake8 * update azure pipeline * skip a sharded test on cpu that requires a gpu * resolve tpus * resolve bug * resolve flake8 * update * updat utils * revert permission change on files * suggestions from carlos Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting changes * remove incomplete comment * Update pytorch_lightning/accelerators/__init__.py Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting change * add types * warn 1.7 ddp manual backward only if ddp kwarg unset * yapf + isort * pep8 unused imports * fix cyclic import in docs * Apply suggestions from code review * typer in accelerator.py * typo * Apply suggestions from code review * formatting * update on comments * update typo * Update pytorch_lightning/trainer/properties.py Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * update * suggestion from code review * suggestion from code review Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> Co-authored-by: SeanNaren <sean@grid.ai> Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz> Co-authored-by: chaton <thomas@grid.ai> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Sean Naren <sean.narenthiran@gmail.com> Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-12 20:48:56 +00:00
env_backup = os.environ.copy()
yield
leaked_vars = os.environ.keys() - env_backup.keys()
PoC: Accelerator refactor (#5743) * restoring the result from subprocess * fix queue.get() order for results * add missing "block_backward_sync" context manager * add missing "block_backward_sync" context manager * fix sync_batchnorm * fix supported gpu-ids for tuple * fix clip gradients and inf recursion * accelerator selection: added cluster_environment plugin * fix torchelastic test * fix reduce early stopping decision for DDP * fix tests: callbacks, conversion to lightning optimizer * fix lightning optimizer does not pickle * fix setting benchmark and deterministic option * fix slurm amp test * fix prepare_data test and determine node_rank * fix retrieving last path when testing * remove obsolete plugin argument * fix test: test_trainer_config * fix torchscript tests * fix trainer.model access * move properties * fix test_transfer_batch_hook * fix auto_select_gpus * fix omegaconf test * fix test that needs to simulate slurm ddp * add horovod plugin * fix test with named arguments * clean up whitespace * fix datamodules test * remove old accelerators * fix naming * move old plugins * move to plugins * create precision subpackage * create training_type subpackage * fix all new import errors * fix wrong arguments order passed to test * fix LR finder * Added sharded training type and amp plugin * Move clip grad to precision plugin * Added sharded spawn, select accelerators based on distributed_backend + enable custom fp16 plugin automatically * Fix import issue, attempting to fix tests * Fix initial test * Reflect hook logic from master, should wrap model after move to device * Optional state consolidation, since master has optimizers not wrapped * change attribute for instance test * reset optimizers optimizers are not used in main process, so state would be wrong. * legacy * imports in accel * legacy2 * trainer imports * fix import errors after rebase * move hook to new setup location * provide unwrapping logic * fix trainer callback system * added ddp2 implementation * fix imports .legacy * move plugins * restore legacy * drop test.py from root * add tpu accelerator and plugins * fixes * fix lightning optimizer merge * reset bugreportmodel * unwrapping * step routing forward * model access * unwrap * opt * integrate distrib_type * sync changes * sync * fixes * add forgotten generators * add missing logic * update * import * missed imports * import fixes * isort * mv f * changelog * format * move helper to parallel plugin * d * add world size * clean up * duplicate * activate ddp_sharded and tpu * set nvidia flags * remove unused colab var * use_tpu <-> on_tpu attrs * make some ddp_cpu and clusterplugin tests pass * Ref/accelerator connector (#5742) * final cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * connector cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * trainer cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * accelerator cleanup + missing logic in accelerator connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add missing changes to callbacks Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * reflect accelerator changes to lightning module Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * clean cluster envs Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * cleanup plugins Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add broadcasting Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * yapf * remove plugin connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * plugins * manual optimization * update optimizer routing * add rank to torchelastic * fix memory mixed precision * setstate on trainer for pickling in ddp spawn * add predict method * add back commented accelerator code * adapt test for sync_batch_norm to new plugin * fix deprecated tests * fix ddp cpu choice when no num_processes are given * yapf format * skip a memory test that cannot pass anymore * fix pickle error in spawn plugin * x * avoid * x * fix cyclic import in docs build * add support for sharded * update typing * add sharded and sharded_spawn to distributed types * make unwrap model default * refactor LightningShardedDataParallel similar to LightningDistributedDataParallel * update sharded spawn to reflect changes * update sharded to reflect changes * Merge 1.1.5 changes * fix merge * fix merge * yapf isort * fix merge * yapf isort * fix indentation in test * copy over reinit scheduler implementation from dev1.2 * fix apex tracking calls with dev_debugger * reduce diff to dev1.2, clean up * fix trainer config test when gpus>0 and num_processes >0 and ddp_cpu * sort plugin tests legacy/new * fix error handling for amp on cpu * fix merge fix merge fix merge * [Feat] Resolve manual_backward (#5837) * resolve manual_backward * resolve flake8 * update * resolve for ddp_spawn * resolve flake8 * resolve flake8 * resolve flake8 Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * fix tests/accelerator tests on cpu * [BugFix] Resolve manual optimization (#5852) * resolve manual_optimization * update * update Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * Remove copy trainer parameters to happen earlier within the loop and add safe guard to get ref model (#5856) * resovle a bug * Accelerator refactor sharded rpc (#5854) * rpc branch * merge * update handling of rpc * make devices etc. Optional in RPC * set devices etc. later if necessary * remove devices from sequential * make devices optional in rpc * fix import * uncomment everything * fix cluster selection Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * resolve bug * fix assert in rpc test * resolve a test * fix docs compilation * accelerator refactor - fix for sharded parity test (#5866) * fix memory issue with ddp_spawn * x x x x x x x x x * x * Remove DDP2 as this does not apply * Add missing pre optimizer hook to ensure lambda closure is called * fix apex docstring * [accelerator][BugFix] Resolve some test for 1 gpu (#5863) * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * update * resolve flake8 * update * update * update * update * update * all_gather * update * make plugins work, add misconfig for RPC * update * update * remove breaking test * resolve some tests * resolve flake8 * revert to ddp_spawn Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de> * yapf isort * resolve flake8 * fix apex doctests * fix apex doctests 2 * resolve docs * update drone * clean env * update * update * update * update * merge * Fix RPC related tests, clean out old API, update for new accelerator API [skip ci] (#5881) * Fix RPC related tests, clean out old API, update for new accelerator API * Move tests out of legacy folder, update paths and names * Update test_remove_1-4.py * Expose properties for tpu cores/gpus/num_gpus * Add root GPU property * Move properties to properties.py * move tests that were previously in drone * Fix root GPU property (#5908) * Move root GPU to property, remove horovod set as this is handled in horovod plugin, ensure we mock correctly to set GPU accelerator * Add missing tests back * fix best model path transfer when no checkpoint callback available * Fix setup hook order [wip] (#5858) * Call trainer setup hook before accelerator setup * Add test case * add new test * typo * fix callback order in test Co-authored-by: tchaton <thomas@grid.ai> Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * rename ddp sequential -> rpc sequential for special test * revert * fix stupid merge problem * Use property in connector for sampler (#5913) * merge the import conflicts * fix spawning of processes in slurm * [wip] Fix some bugs for TPU [skip ci] (#5878) * fixed for single tpu * fixed spawn * fixed spawn * update * update * wip * resolve bugs * resolve bug * update on comment * removed decorator * resolve comments * set to 4 * update * update * need cleaning * update * update * update * resolve flake8 * resolve bugs * exclude broadcast * resolve bugs * change test * update * update * skip if meet fails * properly raise trace * update * add catch * wrap test * resolve typo * update * typo Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> * resolve some tests * update * fix imports * update * resolve flake8 * update azure pipeline * skip a sharded test on cpu that requires a gpu * resolve tpus * resolve bug * resolve flake8 * update * updat utils * revert permission change on files * suggestions from carlos Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting changes * remove incomplete comment * Update pytorch_lightning/accelerators/__init__.py Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting change * add types * warn 1.7 ddp manual backward only if ddp kwarg unset * yapf + isort * pep8 unused imports * fix cyclic import in docs * Apply suggestions from code review * typer in accelerator.py * typo * Apply suggestions from code review * formatting * update on comments * update typo * Update pytorch_lightning/trainer/properties.py Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * update * suggestion from code review * suggestion from code review Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> Co-authored-by: SeanNaren <sean@grid.ai> Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz> Co-authored-by: chaton <thomas@grid.ai> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Sean Naren <sean.narenthiran@gmail.com> Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-12 20:48:56 +00:00
# restore environment as it was before running the test
os.environ.clear()
os.environ.update(env_backup)
# these are currently known leakers - ideally these would not be allowed
allowlist = {
"CUBLAS_WORKSPACE_CONFIG", # enabled with deterministic flag
"CUDA_DEVICE_ORDER",
"LOCAL_RANK",
"NODE_RANK",
"WORLD_SIZE",
"MASTER_ADDR",
"MASTER_PORT",
"PL_GLOBAL_SEED",
"PL_SEED_WORKERS",
"WANDB_MODE",
"WANDB_REQUIRE_SERVICE",
"WANDB_SERVICE",
"HOROVOD_FUSION_THRESHOLD",
"RANK", # set by DeepSpeed
"POPLAR_ENGINE_OPTIONS", # set by IPUStrategy
# set by XLA
"TF2_BEHAVIOR",
"XRT_MESH_SERVICE_ADDRESS",
"XRT_TORCH_DIST_ROOT",
"XRT_MULTI_PROCESSING_DEVICE",
"XRT_SHARD_WORLD_SIZE",
"XRT_LOCAL_WORKER",
"XRT_HOST_WORLD_SIZE",
"XRT_SHARD_ORDINAL",
"XRT_SHARD_LOCAL_ORDINAL",
"TF_CPP_MIN_LOG_LEVEL",
}
leaked_vars.difference_update(allowlist)
assert not leaked_vars, f"test is leaking environment variable(s): {set(leaked_vars)}"
PoC: Accelerator refactor (#5743) * restoring the result from subprocess * fix queue.get() order for results * add missing "block_backward_sync" context manager * add missing "block_backward_sync" context manager * fix sync_batchnorm * fix supported gpu-ids for tuple * fix clip gradients and inf recursion * accelerator selection: added cluster_environment plugin * fix torchelastic test * fix reduce early stopping decision for DDP * fix tests: callbacks, conversion to lightning optimizer * fix lightning optimizer does not pickle * fix setting benchmark and deterministic option * fix slurm amp test * fix prepare_data test and determine node_rank * fix retrieving last path when testing * remove obsolete plugin argument * fix test: test_trainer_config * fix torchscript tests * fix trainer.model access * move properties * fix test_transfer_batch_hook * fix auto_select_gpus * fix omegaconf test * fix test that needs to simulate slurm ddp * add horovod plugin * fix test with named arguments * clean up whitespace * fix datamodules test * remove old accelerators * fix naming * move old plugins * move to plugins * create precision subpackage * create training_type subpackage * fix all new import errors * fix wrong arguments order passed to test * fix LR finder * Added sharded training type and amp plugin * Move clip grad to precision plugin * Added sharded spawn, select accelerators based on distributed_backend + enable custom fp16 plugin automatically * Fix import issue, attempting to fix tests * Fix initial test * Reflect hook logic from master, should wrap model after move to device * Optional state consolidation, since master has optimizers not wrapped * change attribute for instance test * reset optimizers optimizers are not used in main process, so state would be wrong. * legacy * imports in accel * legacy2 * trainer imports * fix import errors after rebase * move hook to new setup location * provide unwrapping logic * fix trainer callback system * added ddp2 implementation * fix imports .legacy * move plugins * restore legacy * drop test.py from root * add tpu accelerator and plugins * fixes * fix lightning optimizer merge * reset bugreportmodel * unwrapping * step routing forward * model access * unwrap * opt * integrate distrib_type * sync changes * sync * fixes * add forgotten generators * add missing logic * update * import * missed imports * import fixes * isort * mv f * changelog * format * move helper to parallel plugin * d * add world size * clean up * duplicate * activate ddp_sharded and tpu * set nvidia flags * remove unused colab var * use_tpu <-> on_tpu attrs * make some ddp_cpu and clusterplugin tests pass * Ref/accelerator connector (#5742) * final cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * connector cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * trainer cleanup Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * accelerator cleanup + missing logic in accelerator connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add missing changes to callbacks Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * reflect accelerator changes to lightning module Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * clean cluster envs Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * cleanup plugins Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * add broadcasting Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * yapf * remove plugin connector Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * plugins * manual optimization * update optimizer routing * add rank to torchelastic * fix memory mixed precision * setstate on trainer for pickling in ddp spawn * add predict method * add back commented accelerator code * adapt test for sync_batch_norm to new plugin * fix deprecated tests * fix ddp cpu choice when no num_processes are given * yapf format * skip a memory test that cannot pass anymore * fix pickle error in spawn plugin * x * avoid * x * fix cyclic import in docs build * add support for sharded * update typing * add sharded and sharded_spawn to distributed types * make unwrap model default * refactor LightningShardedDataParallel similar to LightningDistributedDataParallel * update sharded spawn to reflect changes * update sharded to reflect changes * Merge 1.1.5 changes * fix merge * fix merge * yapf isort * fix merge * yapf isort * fix indentation in test * copy over reinit scheduler implementation from dev1.2 * fix apex tracking calls with dev_debugger * reduce diff to dev1.2, clean up * fix trainer config test when gpus>0 and num_processes >0 and ddp_cpu * sort plugin tests legacy/new * fix error handling for amp on cpu * fix merge fix merge fix merge * [Feat] Resolve manual_backward (#5837) * resolve manual_backward * resolve flake8 * update * resolve for ddp_spawn * resolve flake8 * resolve flake8 * resolve flake8 Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * fix tests/accelerator tests on cpu * [BugFix] Resolve manual optimization (#5852) * resolve manual_optimization * update * update Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * Remove copy trainer parameters to happen earlier within the loop and add safe guard to get ref model (#5856) * resovle a bug * Accelerator refactor sharded rpc (#5854) * rpc branch * merge * update handling of rpc * make devices etc. Optional in RPC * set devices etc. later if necessary * remove devices from sequential * make devices optional in rpc * fix import * uncomment everything * fix cluster selection Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> * resolve bug * fix assert in rpc test * resolve a test * fix docs compilation * accelerator refactor - fix for sharded parity test (#5866) * fix memory issue with ddp_spawn * x x x x x x x x x * x * Remove DDP2 as this does not apply * Add missing pre optimizer hook to ensure lambda closure is called * fix apex docstring * [accelerator][BugFix] Resolve some test for 1 gpu (#5863) * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * update * update * revert init * resolve a bug * update * resolve flake8 * update * update * update * revert init * update * resolve flake8 * update * update * update * update * update * all_gather * update * make plugins work, add misconfig for RPC * update * update * remove breaking test * resolve some tests * resolve flake8 * revert to ddp_spawn Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de> * yapf isort * resolve flake8 * fix apex doctests * fix apex doctests 2 * resolve docs * update drone * clean env * update * update * update * update * merge * Fix RPC related tests, clean out old API, update for new accelerator API [skip ci] (#5881) * Fix RPC related tests, clean out old API, update for new accelerator API * Move tests out of legacy folder, update paths and names * Update test_remove_1-4.py * Expose properties for tpu cores/gpus/num_gpus * Add root GPU property * Move properties to properties.py * move tests that were previously in drone * Fix root GPU property (#5908) * Move root GPU to property, remove horovod set as this is handled in horovod plugin, ensure we mock correctly to set GPU accelerator * Add missing tests back * fix best model path transfer when no checkpoint callback available * Fix setup hook order [wip] (#5858) * Call trainer setup hook before accelerator setup * Add test case * add new test * typo * fix callback order in test Co-authored-by: tchaton <thomas@grid.ai> Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * rename ddp sequential -> rpc sequential for special test * revert * fix stupid merge problem * Use property in connector for sampler (#5913) * merge the import conflicts * fix spawning of processes in slurm * [wip] Fix some bugs for TPU [skip ci] (#5878) * fixed for single tpu * fixed spawn * fixed spawn * update * update * wip * resolve bugs * resolve bug * update on comment * removed decorator * resolve comments * set to 4 * update * update * need cleaning * update * update * update * resolve flake8 * resolve bugs * exclude broadcast * resolve bugs * change test * update * update * skip if meet fails * properly raise trace * update * add catch * wrap test * resolve typo * update * typo Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> * resolve some tests * update * fix imports * update * resolve flake8 * update azure pipeline * skip a sharded test on cpu that requires a gpu * resolve tpus * resolve bug * resolve flake8 * update * updat utils * revert permission change on files * suggestions from carlos Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting changes * remove incomplete comment * Update pytorch_lightning/accelerators/__init__.py Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> * remove unrelated formatting change * add types * warn 1.7 ddp manual backward only if ddp kwarg unset * yapf + isort * pep8 unused imports * fix cyclic import in docs * Apply suggestions from code review * typer in accelerator.py * typo * Apply suggestions from code review * formatting * update on comments * update typo * Update pytorch_lightning/trainer/properties.py Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> * update * suggestion from code review * suggestion from code review Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com> Co-authored-by: SeanNaren <sean@grid.ai> Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz> Co-authored-by: chaton <thomas@grid.ai> Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal> Co-authored-by: Sean Naren <sean.narenthiran@gmail.com> Co-authored-by: root <root@ip-172-31-88-60.ec2.internal> Co-authored-by: Lezwon Castelino <lezwon@gmail.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com> Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-12 20:48:56 +00:00
@pytest.fixture(scope="function", autouse=True)
def restore_signal_handlers():
"""Ensures that signal handlers get restored before the next test runs.
This is a safety net for tests that don't run Trainer's teardown.
"""
valid_signals = SignalConnector._valid_signals()
if not _IS_WINDOWS:
# SIGKILL and SIGSTOP are not allowed to be modified by the user
valid_signals -= {signal.SIGKILL, signal.SIGSTOP}
handlers = {signum: signal.getsignal(signum) for signum in valid_signals}
yield
for signum, handler in handlers.items():
if handler is not None:
signal.signal(signum, handler)
Support `DDPPlugin` to be used on CPU (#6208) * Skip test due to 'Python bus error' * Debug NCCL * Remove NCCL_DEBUG statement * Revert "Skip test due to 'Python bus error'" This reverts commit e0a3e8785d2fecd63667da433a648f958d60ef89. * fix * add test * changelog * yapf * patch os environ * make a special test * destroy pg * debug * revert * revert * problematic test * skip * try the fixture * test * update sensitive test * update changelog * remove comment * update wrong test * update test name * parameterization * Revert "parameterization" This reverts commit b0542f43f59c5ce66800883b5e2f0c66a97408cc. * remove conftest * ignore test * teardown * fix merge * deep speed parameterization * uncomment test * update chlog * update changelog * split tests * update test update test update test update test * update test comments * unroll test * unroll test * unroll test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * increase shm * sudo * unroll ipu * Revert "sudo" This reverts commit 6cc68c1478c151caee86b8a0f8ded16b62a9c8ea. * Revert "increase shm" This reverts commit 8c2716348330dbc3c6b3685647f435127b870d79. * x * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * find guilty test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * POPTORCH_WAIT_FOR_IPU=1 * move test * redo parameterize for ipu * de-comment test * move chlog * Update tests/accelerators/test_accelerator_connector.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Update tests/accelerators/test_accelerator_connector.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
2021-07-02 11:00:24 +00:00
@pytest.fixture(scope="function", autouse=True)
def teardown_process_group():
"""Ensures that the distributed process group gets closed before the next test runs."""
Support `DDPPlugin` to be used on CPU (#6208) * Skip test due to 'Python bus error' * Debug NCCL * Remove NCCL_DEBUG statement * Revert "Skip test due to 'Python bus error'" This reverts commit e0a3e8785d2fecd63667da433a648f958d60ef89. * fix * add test * changelog * yapf * patch os environ * make a special test * destroy pg * debug * revert * revert * problematic test * skip * try the fixture * test * update sensitive test * update changelog * remove comment * update wrong test * update test name * parameterization * Revert "parameterization" This reverts commit b0542f43f59c5ce66800883b5e2f0c66a97408cc. * remove conftest * ignore test * teardown * fix merge * deep speed parameterization * uncomment test * update chlog * update changelog * split tests * update test update test update test update test * update test comments * unroll test * unroll test * unroll test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * increase shm * sudo * unroll ipu * Revert "sudo" This reverts commit 6cc68c1478c151caee86b8a0f8ded16b62a9c8ea. * Revert "increase shm" This reverts commit 8c2716348330dbc3c6b3685647f435127b870d79. * x * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * find guilty test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * POPTORCH_WAIT_FOR_IPU=1 * move test * redo parameterize for ipu * de-comment test * move chlog * Update tests/accelerators/test_accelerator_connector.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> * Update tests/accelerators/test_accelerator_connector.py Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
2021-07-02 11:00:24 +00:00
yield
if torch.distributed.is_available() and torch.distributed.is_initialized():
torch.distributed.destroy_process_group()
@pytest.fixture(scope="function", autouse=True)
def reset_deterministic_algorithm():
"""Ensures that torch determinism settings are reset before the next test runs."""
yield
if _TORCH_GREATER_EQUAL_1_8:
torch.use_deterministic_algorithms(False)
else:
torch.set_deterministic(False)
@pytest.fixture
def caplog(caplog):
"""Workaround for https://github.com/pytest-dev/pytest/issues/3697.
Setting ``filterwarnings`` with pytest breaks ``caplog`` when ``not logger.propagate``.
"""
import logging
lightning_logger = logging.getLogger("pytorch_lightning")
propagate = lightning_logger.propagate
lightning_logger.propagate = True
yield caplog
lightning_logger.propagate = propagate
@pytest.fixture
def tmpdir_server(tmpdir):
2021-12-21 17:06:15 +00:00
Handler = partial(SimpleHTTPRequestHandler, directory=str(tmpdir))
from http.server import ThreadingHTTPServer
with ThreadingHTTPServer(("localhost", 0), Handler) as server:
server_thread = threading.Thread(target=server.serve_forever)
# Exit the server thread when the main thread terminates
server_thread.daemon = True
server_thread.start()
yield server.server_address
server.shutdown()
@pytest.fixture
def single_process_pg():
"""Initialize the default process group with only the current process for testing purposes.
The process group is destroyed when the with block is exited.
"""
if torch.distributed.is_initialized():
raise RuntimeError("Can't use `single_process_pg` when the default process group is already initialized.")
orig_environ = os.environ.copy()
os.environ["MASTER_ADDR"] = "localhost"
os.environ["MASTER_PORT"] = str(find_free_network_port())
os.environ["RANK"] = "0"
os.environ["WORLD_SIZE"] = "1"
torch.distributed.init_process_group("gloo")
try:
yield
finally:
torch.distributed.destroy_process_group()
os.environ.clear()
os.environ.update(orig_environ)
def pytest_collection_modifyitems(items):
# filter out special tests
if os.getenv("PL_RUN_STANDALONE_TESTS", "0") == "1":
items[:] = [
item
for item in items
for marker in item.own_markers
# has `@RunIf(standalone=True)`
if marker.name == "skipif" and marker.kwargs.get("standalone")
]
elif os.getenv("PL_RUN_SLOW_TESTS", "0") == "1":
items[:] = [
item
for item in items
for marker in item.own_markers
# has `@RunIf(slow=True)`
if marker.name == "skipif" and marker.kwargs.get("slow")
]