mirrors
/
pytorch-image-models
mirror of https://github.com/huggingface/pytorch-image-models.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
pytorch-image-models/timm/models/layers/__init__.py

23 lines
1006 B
Python
Raw Normal View History

Layer refactoring continues, ResNet downsample rewrite for proper dilation in 3x3 and avg_pool cases * select_conv2d -> create_conv2d * added create_attn to create attention module from string/bool/module * factor padding helpers into own file, use in both conv2d_same and avg_pool2d_same * add some more test eca resnet variants * minor tweaks, naming, comments, consistency
2020-02-11 03:55:03 +08:00
from .padding import get_padding
Update EfficientNet feature extraction for EfficientDet. Add needed MaxPoolSame as well.
2020-04-09 16:41:54 +08:00
from .pool2d_same import AvgPool2dSame
Layer refactoring continues, ResNet downsample rewrite for proper dilation in 3x3 and avg_pool cases * select_conv2d -> create_conv2d * added create_attn to create attention module from string/bool/module * factor padding helpers into own file, use in both conv2d_same and avg_pool2d_same * add some more test eca resnet variants * minor tweaks, naming, comments, consistency
2020-02-11 03:55:03 +08:00
from .conv2d_same import Conv2dSame
A bunch more layer reorg, splitting many layers into own files. Improve torchscript compatibility.
2020-02-10 06:46:28 +08:00
from .conv_bn_act import ConvBnAct
from .mixed_conv2d import MixedConv2d
from .cond_conv2d import CondConv2d, get_condconv_initializer
Update EfficientNet feature extraction for EfficientDet. Add needed MaxPoolSame as well.
2020-04-09 16:41:54 +08:00
from .pool2d_same import create_pool2d
Layer refactoring continues, ResNet downsample rewrite for proper dilation in 3x3 and avg_pool cases * select_conv2d -> create_conv2d * added create_attn to create attention module from string/bool/module * factor padding helpers into own file, use in both conv2d_same and avg_pool2d_same * add some more test eca resnet variants * minor tweaks, naming, comments, consistency
2020-02-11 03:55:03 +08:00
from .create_conv2d import create_conv2d
from .create_attn import create_attn
A bunch more layer reorg, splitting many layers into own files. Improve torchscript compatibility.
2020-02-10 06:46:28 +08:00
from .selective_kernel import SelectiveKernelConv
Layer refactoring continues, ResNet downsample rewrite for proper dilation in 3x3 and avg_pool cases * select_conv2d -> create_conv2d * added create_attn to create attention module from string/bool/module * factor padding helpers into own file, use in both conv2d_same and avg_pool2d_same * add some more test eca resnet variants * minor tweaks, naming, comments, consistency
2020-02-11 03:55:03 +08:00
from .se import SEModule
Fix module name mistake, start layers sub-package
2020-02-10 04:44:26 +08:00
from .eca import EcaModule, CecaModule
Big move, layer modules and fn to timm/models/layers
2020-02-10 05:13:08 +08:00
from .activations import *
from .adaptive_avgmax_pool import \
adaptive_avgmax_pool2d, select_adaptive_pool2d, AdaptiveAvgMaxPool2d, SelectAdaptivePool2d
Update comments for Selective Kernel and DropBlock/Path impl, add skresnet34 weights
2020-02-19 05:58:30 +08:00
from .drop import DropBlock2d, DropPath, drop_block_2d, drop_path
Big move, layer modules and fn to timm/models/layers
2020-02-10 05:13:08 +08:00
from .test_time_pool import TestTimePoolHead, apply_test_time_pool
from .split_batchnorm import SplitBatchNorm2d, convert_splitbn_model
TResNet models
2020-04-12 23:44:12 +08:00
from .anti_aliasing import AntiAliasDownsampleLayer
Merge branch 'blur' of https://github.com/VRandme/pytorch-image-models into VRandme-blur
2020-05-02 06:19:57 +08:00
from .space_to_depth import SpaceToDepthModule
Final blurpool2d cleanup and add resnetblur50 weights, match tresnet Downsample arg order to BlurPool2d for interop
2020-05-02 08:00:21 +08:00
from .blur_pool import BlurPool2d
Remove annoying InceptionV3 dependency on scipy and insanely slow trunc_norm init. Bring InceptionV3 code into this codebase and use upcoming torch trunch_norm_ init.
2020-05-13 12:59:34 +08:00
from .weight_init import trunc_normal_