This repository has been archived by the owner on Jul 1, 2024. It is now read-only.
Releases: facebookresearch/ClassyVision
Releases · facebookresearch/ClassyVision
v0.7.0
v0.6.0
New Features
- Added compatibility with pytorch version 1.9.0 (#753)
- Added RegNetZ model implementation (#713)
- Redefined RegNetY 64GF to better performing model (#748)
- Added support for Pytorch AMP + ShardedDDP (#667)
- Added support for cutmix in
MixupTransform
- Added support for a 1x1 convolution in the fully connected head (#712)
- Add support to normalize inputs in the heads (#664)
- Added support for SiLU activation in RegNets (#711)
- Added support for FP16 gradient communication in DDP (#709)
- Added reduction mode
none
insoft_target_cross_entropy_loss
(#714) - Added FLOPs computation to ViT (#746)
- Upstreamed parameter schedulers to fvcore
v0.5.0
New Features
- Release Vision Transformers model implementation, with recipes(#646)
- Implemented gradient clipping (#643)
- Implemented gradient accumulation (#644)
- Added support for AdamW (#636)
- Added Precise batch norm hook (#592)
- Added support for adaptive pooling in
fully_convolutional_linear_head
(#602) - Added support for sync batch norm group size (#534)
- Added a CSV Hook to manually inspect model predictions
- Added a ClassyModel tutorial (#485)
- Migrated to Hydra 1.0 (#536)
- Migrated off of tensorboardX (#488)
Breaking Changes
v0.4.0
New Features
- Release EfficientNet model implementation (#475)
- Add support to convert any
PyTorch
model to aClassyModel
with the ability to attach heads to it (#461) - Squeeze and Excitation support for
ResNe(X)t
andDenseNet
models (#426, #427) - Made
ClassyHook
s registrable (#401) and configurable (#402) - Migrated to
TorchElastic v0.2.0
(#464) - Add
SyncBatchNorm
support (#423) - Implement
mixup
train augmentation (#469) - Support
LARC
for SGD optimizer (#408) - Added convenience wrappers for
Iterable
datasets (#455) Tensorboard
improvements- Invalid (
NaN
/Inf
) loss detection - Revamped logging (#478)
- Add
bn_weight_decay
configuration option forResNe(X)t
models - Support specifying
update_interval
to Parameter Schedulers (#418)
Breaking changes
ClassificationTask
API improvement andtrain_step
,eval_step
simplification- Rename
lr
tovalue
in parameter schedulers (#417)
v0.3.0
V0.30 release notes:
- checkpoint_folder renamed to checkpoint_load_path: #379
- head support on DenseNet: #383
- Cleaner abstraction in ClassyTask/ClassyTrainer: eval_step, on_start, on_end, …
- Speed metrics in TB: #385
- test_phase_period in ClassificationTask #395
- support for losses with trainable parameters #394
- Added presets for some typical resNe(X)t configurations: #405