mmclassification/configs/deit/deit-base_pt-16xb64_in1k.py

_base_ = './deit-small_pt-4xb256_in1k.py'

# model settings
model = dict(
    backbone=dict(
        type='VisionTransformer', arch='deit-base', drop_path_rate=0.1),
    head=dict(type='VisionTransformerClsHead', in_channels=768),
)

# dataset settings
train_dataloader = dict(batch_size=64)

# runtime settings
custom_hooks = [dict(type='EMAHook', momentum=4e-5, priority='ABOVE_NORMAL')]
[Feature] Add DeiT backbone and checkpoints. (#576) * Support DeiT backbone. * Use hook to automatically resize pos embed * Update ViT training setting * Add deit configs and update docs * Fix vit arch assertion * Remove useless init function * Add unit tests. * Fix resize_pos_embed for DeiT * Improve according to comments. 2021-12-15 22:44:57 +08:00			`_base_ = './deit-small_pt-4xb256_in1k.py'`

			`# model settings`
			`model = dict(`
[Reproduction] Reproduce training results of DeiT. (#711) * Update deit training settings * Update decay config * Add mixup&cutmix and drop path rate * Update training configs * Update model-zoo * Add comments 2022-03-02 14:23:10 +08:00			`backbone=dict(`
			`type='VisionTransformer', arch='deit-base', drop_path_rate=0.1),`
[Feature] Add DeiT backbone and checkpoints. (#576) * Support DeiT backbone. * Use hook to automatically resize pos embed * Update ViT training setting * Add deit configs and update docs * Fix vit arch assertion * Remove useless init function * Add unit tests. * Fix resize_pos_embed for DeiT * Improve according to comments. 2021-12-15 22:44:57 +08:00			`head=dict(type='VisionTransformerClsHead', in_channels=768),`
			`)`

[Refactor] Add `ResizeEdge` and refactor all dataset configs. 2022-06-01 06:11:53 +00:00			`# dataset settings`
			`train_dataloader = dict(batch_size=64)`
[Reproduction] Reproduce training results of DeiT. (#711) * Update deit training settings * Update decay config * Add mixup&cutmix and drop path rate * Update training configs * Update model-zoo * Add comments 2022-03-02 14:23:10 +08:00
[Refactor] Add `ResizeEdge` and refactor all dataset configs. 2022-06-01 06:11:53 +00:00			`# runtime settings`
[Reproduction] Reproduce training results of DeiT. (#711) * Update deit training settings * Update decay config * Add mixup&cutmix and drop path rate * Update training configs * Update model-zoo * Add comments 2022-03-02 14:23:10 +08:00			`custom_hooks = [dict(type='EMAHook', momentum=4e-5, priority='ABOVE_NORMAL')]`