添加mmclassification-0.24.1代码，删除mmclassification-speed-benchmark

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb128_coslr-90e_in21k.py

+_base_ = [
+    '../_base_/models/resnet50.py', '../_base_/datasets/imagenet21k_bs128.py',
+    '../_base_/schedules/imagenet_bs1024_coslr.py',
+    '../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(head=dict(num_classes=21843))
+
+# runtime settings
+runner = dict(type='EpochBasedRunner', max_epochs=90)

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb256-rsb-a1-600e_in1k.py

+_base_ = [
+    '../_base_/models/resnet50.py',
+    '../_base_/datasets/imagenet_bs256_rsb_a12.py',
+    '../_base_/schedules/imagenet_bs2048_rsb.py',
+    '../_base_/default_runtime.py'
+]
+
+# Model settings
+model = dict(
+    backbone=dict(
+        norm_cfg=dict(type='SyncBN', requires_grad=True),
+        drop_path_rate=0.05,
+    ),
+    head=dict(
+        loss=dict(
+            type='LabelSmoothLoss',
+            label_smooth_val=0.1,
+            mode='original',
+        )),
+    train_cfg=dict(augments=[
+        dict(type='BatchMixup', alpha=0.2, num_classes=1000, prob=0.5),
+        dict(type='BatchCutMix', alpha=1.0, num_classes=1000, prob=0.5)
+    ]))
+
+# Dataset settings
+sampler = dict(type='RepeatAugSampler')
+
+# Schedule settings
+runner = dict(max_epochs=600)
+optimizer = dict(
+    weight_decay=0.01,
+    paramwise_cfg=dict(bias_decay_mult=0., norm_decay_mult=0.),
+)

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb256-rsb-a2-300e_in1k.py

+_base_ = [
+    '../_base_/models/resnet50.py',
+    '../_base_/datasets/imagenet_bs256_rsb_a12.py',
+    '../_base_/schedules/imagenet_bs2048_rsb.py',
+    '../_base_/default_runtime.py'
+]
+
+# Model settings
+model = dict(
+    backbone=dict(
+        norm_cfg=dict(type='SyncBN', requires_grad=True),
+        drop_path_rate=0.05,
+    ),
+    head=dict(loss=dict(use_sigmoid=True)),
+    train_cfg=dict(augments=[
+        dict(type='BatchMixup', alpha=0.1, num_classes=1000, prob=0.5),
+        dict(type='BatchCutMix', alpha=1.0, num_classes=1000, prob=0.5)
+    ]))
+
+# Dataset settings
+sampler = dict(type='RepeatAugSampler')
+
+# Schedule settings
+runner = dict(max_epochs=300)
+optimizer = dict(paramwise_cfg=dict(bias_decay_mult=0., norm_decay_mult=0.))

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb256-rsb-a3-100e_in1k.py

+_base_ = [
+    '../_base_/models/resnet50.py',
+    '../_base_/datasets/imagenet_bs256_rsb_a3.py',
+    '../_base_/schedules/imagenet_bs2048_rsb.py',
+    '../_base_/default_runtime.py'
+]
+
+# Model settings
+model = dict(
+    backbone=dict(norm_cfg=dict(type='SyncBN', requires_grad=True)),
+    head=dict(loss=dict(use_sigmoid=True)),
+    train_cfg=dict(augments=[
+        dict(type='BatchMixup', alpha=0.1, num_classes=1000, prob=0.5),
+        dict(type='BatchCutMix', alpha=1.0, num_classes=1000, prob=0.5)
+    ]))
+
+# Schedule settings
+optimizer = dict(
+    lr=0.008, paramwise_cfg=dict(bias_decay_mult=0., norm_decay_mult=0.))

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb32-coslr-preciseBN_in1k.py

+_base_ = 'resnet50_8xb32-coslr_in1k.py'
+
+# Precise BN hook will update the bn stats, so this hook should be executed
+# before CheckpointHook, which has priority of 'NORMAL'. So set the
+# priority of PreciseBNHook to 'ABOVE_NORMAL' here.
+custom_hooks = [
+    dict(
+        type='PreciseBNHook',
+        num_samples=8192,
+        interval=1,
+        priority='ABOVE_NORMAL')
+]

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb32-fp16-dynamic_in1k.py

+_base_ = ['./resnet50_8xb32_in1k.py']
+
+# fp16 settings
+fp16 = dict(loss_scale='dynamic')

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb32-fp16_in1k.py

+_base_ = ['./resnet50_8xb32_in1k.py']
+
+# fp16 settings
+fp16 = dict(loss_scale=512.)

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb32_in1k.py

+_base_ = [
+    '../_base_/models/resnet50.py', '../_base_/datasets/imagenet_bs32.py',
+    '../_base_/schedules/imagenet_bs256.py', '../_base_/default_runtime.py'
+]
+
+fp16 = dict(loss_scale=512.)

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb8_cars.py

+_base_ = [
+    '../_base_/models/resnet50.py',
+    '../_base_/datasets/stanford_cars_bs8_448.py',
+    '../_base_/schedules/stanford_cars_bs8.py', '../_base_/default_runtime.py'
+]
+
+# use pre-train weight converted from https://github.com/Alibaba-MIIL/ImageNet21K # noqa
+checkpoint = 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_3rdparty-mill_in21k_20220331-faac000b.pth'  # noqa
+
+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        init_cfg=dict(
+            type='Pretrained', checkpoint=checkpoint, prefix='backbone')),
+    head=dict(num_classes=196, ))
+
+log_config = dict(interval=50)
+checkpoint_config = dict(
+    interval=1, max_keep_ckpts=3)  # save last three checkpoints

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_8xb8_cub.py

+_base_ = [
+    '../_base_/models/resnet50.py', '../_base_/datasets/cub_bs8_448.py',
+    '../_base_/schedules/cub_bs64.py', '../_base_/default_runtime.py'
+]
+
+# use pre-train weight converted from https://github.com/Alibaba-MIIL/ImageNet21K # noqa
+checkpoint = 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_3rdparty-mill_in21k_20220331-faac000b.pth'  # noqa
+
+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        init_cfg=dict(
+            type='Pretrained', checkpoint=checkpoint, prefix='backbone')),
+    head=dict(num_classes=200, ))
+
+log_config = dict(interval=20)  # log every 20 intervals
+
+checkpoint_config = dict(
+    interval=1, max_keep_ckpts=3)  # save last three checkpoints

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_b16x8_cifar10.py

+_base_ = 'resnet50_8xb16_cifar10.py'
+
+_deprecation_ = dict(
+    expected='resnet50_8xb16_cifar10.py',
+    reference='https://github.com/open-mmlab/mmclassification/pull/508',
+)

openmmlab_test/mmclassification-0.24.1/configs/resnet/resnet50_b16x8_cifar100.py

+_base_ = 'resnet50_8xb16_cifar100.py'
+
+_deprecation_ = dict(
+    expected='resnet50_8xb16_cifar100.py',
+    reference='https://github.com/open-mmlab/mmclassification/pull/508',
+)