163 Commits (bits_and_tpu)

Author	SHA1	Message	Date
Ross Wightman	dff33730b3	Merge remote-tracking branch 'origin/master' into bits_and_tpu	3 years ago
Ross Wightman	fd360ac951	Merge pull request #1266 from kaczmarj/enh/no-star-imports ... ENH: replace star imports with imported names in train.py	3 years ago
Jakub Kaczmarzyk	ce5578bc3a	replace star imports with imported names	3 years ago
Jakub Kaczmarzyk	dcad288fd6	use argparse groups to group arguments	3 years ago
Jakub Kaczmarzyk	e1e4c9bbae	rm whitespace	3 years ago
han	a16171335b	fix: change milestones to decay-milestones ... - change argparser option `milestone` to `decay-milestone`	3 years ago
han	57a988df30	fix: multistep lr decay epoch bugs ... - add milestones arguments - change decay_epochs to milestones variable	3 years ago
Ross Wightman	749856cf25	Merge remote-tracking branch 'origin/norm_norm_norm' into bits_and_tpu	3 years ago
Ross Wightman	b049a5c5c6	Merge remote-tracking branch 'origin/master' into norm_norm_norm	3 years ago
Ross Wightman	04db5833eb	Merge pull request #986 from hankyul2/master ... fix: typo of argment parser desc in train.py	3 years ago
Ross Wightman	da2796ae82	Add webdataset (WDS) support, update TFDS to make some naming in parsers more similar. Fix workers=0 compatibility. Add ImageNet22k/12k synset defs.	3 years ago
Ross Wightman	0557c8257d	Fix bug introduced in non layer_decay weight_decay application. Remove debug print, fix arg desc.	3 years ago
Ross Wightman	a16ea1e355	Merge remote-tracking branch 'origin/norm_norm_norm' into bits_and_tpu	3 years ago
Ross Wightman	372ad5fa0d	Significant model refactor and additions: ... * All models updated with revised foward_features / forward_head interface * Vision transformer and MLP based models consistently output sequence from forward_features (pooling or token selection considered part of 'head') * WIP param grouping interface to allow consistent grouping of parameters for layer-wise decay across all model types * Add gradient checkpointing support to a significant % of models, especially popular architectures * Formatting and interface consistency improvements across models * layer-wise LR decay impl part of optimizer factory w/ scale support in scheduler * Poolformer and Volo architectures added	3 years ago
Ross Wightman	fafece230b	Allow changing base lr batch size from 256 via arg	3 years ago
Ross Wightman	7148039f9f	Tweak base lr log	3 years ago
Ross Wightman	f82fb6b608	Add base lr w/ linear and sqrt scaling to train script	3 years ago
Ross Wightman	95cfc9b3e8	Merge remote-tracking branch 'origin/master' into norm_norm_norm	3 years ago
Ross Wightman	abc9ba2544	Transitioning default_cfg -> pretrained_cfg. Improving handling of pretrained_cfg source (HF-Hub, files, timm config, etc). Checkpoint handling tweaks.	3 years ago
Ross Wightman	f0f9eccda8	Add --fuser arg to train/validate/benchmark scripts to select jit fuser type	3 years ago
Ross Wightman	5ccf682a8f	Remove deprecated bn-tf train arg and create_model handler. Add evos/evob models back into fx test filter until norm_norm_norm branch merged.	3 years ago
Ross Wightman	4c8bb295ab	Remove bn-tf arg	3 years ago
han	ab5ae32f75	fix: typo of argment parser desc in train.py ... - Remove duplicated `of`	3 years ago
Ross Wightman	4f338556d8	Fixes and improvements for metrics, tfds parser, loader / transform handling ... * add back ability to create transform with loader * change 'samples' -> 'examples' for tfds wrapper to match tfds naming * add support for specifying feature names for input and target in tfds wrapper * add class_to_idx for image classification datasets in tfds wrapper * add accumulate_type to avg meters and metrics to allow float32 or float64 accumulation control with lower prec data * minor cleanup, log output rate prev and avg	3 years ago
Ross Wightman	80ca078aed	Fix a few bugs and formatting/naming issues ... * Pass optimizer resume flag through to checkpoint / updater restore. Related to #961 but not clear how relates to crash. * Rename monitor step args, cleanup handling of step_end_idx vs num_steps for consistent log output in either case * Resume from proper epoch (ie next epoch relative to checkpoint)	3 years ago
Ross Wightman	406c486ba2	Merge remote-tracking branch 'origin/more_datasets' into bits_and_tpu	3 years ago
Ross Wightman	ba65dfe2c6	Dataset work ... * support some torchvision datasets * improvements to TFDS wrapper for subsplit handling (fix #942), shuffle seed * add class-map support to train (fix #957)	3 years ago
Ross Wightman	cd638d50a5	Merge pull request #880 from rwightman/fixes_bce_regnet ... A collection of fixes, model experiments, etc	3 years ago
Ross Wightman	d9abfa48df	Make broadcast_buffers disable its own flag for now (needs more testing on interaction with dist_bn)	3 years ago
Ross Wightman	80075b0b8a	Add worker_seeding arg to allow selecting old vs updated data loader worker seed for (old) experiment repeatability	3 years ago
Shoufa Chen	908563d060	fix `use_amp` ... Fix https://github.com/rwightman/pytorch-image-models/issues/881	3 years ago
Ross Wightman	25d52ea71d	Merge remote-tracking branch 'origin/fixes_bce_regnet' into bits_and_tpu	3 years ago
Ross Wightman	0387e6057e	Update binary cross ent impl to use thresholding as an option (convert soft targets from mixup/cutmix to 0, 1)	3 years ago
Ross Wightman	3581affb77	Update train.py with some flags related to scheduler tweaks, fix best checkpoint bug.	3 years ago
Ross Wightman	0639d9a591	Fix updated validation_batch_size fallback	3 years ago
Ross Wightman	5db057dca0	Fix misnamed arg, tweak other train script args for better defaults.	3 years ago
Ross Wightman	fb94350896	Update training script and loader factory to allow use of scheduler updates, repeat augment, and bce loss	3 years ago
Ross Wightman	f2e14685a8	Add force-cpu flag for train/validate, fix CPU fallback for device init, remove old force cpu flag for EMA model weights	3 years ago
Ross Wightman	0d82876132	Add comment for reference re PyTorch XLA 'race' issue	3 years ago
Ross Wightman	40457e5691	Transforms, augmentation work for bits, add RandomErasing support for XLA (pushing into transforms), revamp of transform/preproc config, etc ongoing...	3 years ago
SamuelGabriel	7c19c35d9f	Global instead of local rank.	3 years ago
Ross Wightman	c3db5f5801	Worker hack for TFDS eval, add TPU env var setting.	3 years ago
Ross Wightman	f411724de4	Fix checkpoint delete issue. Add README about bits and initial Pytorch XLA usage on TPU-VM. Add some FIXMEs and fold train_cfg into train_state by default.	3 years ago
Ross Wightman	91ab0b6ce5	Add proper TrainState checkpoint save/load. Some reorg/refactoring and other cleanup. More to go...	3 years ago
Ross Wightman	5b9c69e80a	Add basic training resume based on legacy code	4 years ago
Ross Wightman	cbd4ee737f	Fix model init for XLA, remove some prints.	4 years ago
Ross Wightman	6d90fcf282	Fix distribute_bn and model_ema	4 years ago
Ross Wightman	aa92d7b1c5	Major timm.bits update. Updater and DeviceEnv now dataclasses, after_step closure used, metrics base impl w/ distributed reduce, many tweaks/fixes.	4 years ago
Ross Wightman	76de984a5f	Fix some bugs with XLA support, logger, add hacky xla dist launch script since torch.dist.launch doesn't work	4 years ago
Ross Wightman	12d9a6d4d2	First timm.bits commit, add initial abstractions, WIP updates to train, val... some of it working	4 years ago