You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Ross Wightman f54612f648 Merge branch 'select_kernel' into attention		5 years ago
..
layers	Fix module name mistake, start layers sub-package	5 years ago
__init__.py	Move Selective Kernel blocks/convs to their own sknet.py file	5 years ago
activations.py	Bring EfficientNet and MobileNetV3 up to date with my gen-efficientnet repo	5 years ago
adaptive_avgmax_pool.py	Remove all prints, change most to logging calls, tweak alignment of batch logs, improve setup.py	5 years ago
conv2d_layers.py	SelectKernel split_input works best when input channels split like grouped conv, but output is full width. Disable zero_init for SK nets, seems a bad combo.	5 years ago
densenet.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
dla.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
dpn.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
efficientnet.py	Add PyTorch trained EfficientNet-ES weights from Andrew Lavin	5 years ago
efficientnet_blocks.py	Bring EfficientNet and MobileNetV3 up to date with my gen-efficientnet repo	5 years ago
efficientnet_builder.py	Remove unused default_init for EfficientNets, experimenting with fanout calc for #84	5 years ago
factory.py	Update EfficientNet comments, MobileNetV3 non-TF create fns, fix factory arg checks, bump PyTorch version req to 1.2	5 years ago
feature_hooks.py	Bring EfficientNet and MobileNetV3 up to date with my gen-efficientnet repo	5 years ago
gluon_resnet.py	Add ResNet deep tiered stem and model weights for seresnext26t_32x4d and seresnext26d_32x4d	5 years ago
gluon_xception.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
helpers.py	Add checkpoint averaging script. Add headers, shebangs, exec perms to all scripts	5 years ago
hrnet.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
inception_resnet_v2.py	InceptionResNetV2 torchscript compatible	5 years ago
inception_v3.py	Update version, results csv files, and move remaining dropbox weights to github	5 years ago
inception_v4.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
median_pool.py	Big re-org, working towards making pip/module as 'timm'	5 years ago
mobilenetv3.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
nasnet.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
nn_ops.py	Just leave it float for now, will look at fp16 later. Remove unused reference code.	5 years ago
pnasnet.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
registry.py	Model updates. Add my best ResNet50 weights top-1=78.47. Add some other torchvision weights.	5 years ago
res2net.py	Add separate zero_init_last_bn function to support more block variety without a mess	5 years ago
resnet.py	Merge branch 'select_kernel' into attention	5 years ago
selecsls.py	Add selecsls60 weights	5 years ago
senet.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
sknet.py	SelectKernel split_input works best when input channels split like grouped conv, but output is full width. Disable zero_init for SK nets, seems a bad combo.	5 years ago
split_batchnorm.py	Update AugMix, JSD, etc comments and references	5 years ago
test_time_pool.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago
xception.py	Finish with HRNet, weights and models updated. Improve consistency in model classifier/global pool treatment.	5 years ago