Ma Zerun dbf3df21a3
[Refactor] Use out_type to specify ViT-like backbone output. (#1408)
* [Refactor] Use  to specify ViT-like backbone output.

* Fix ClsBatchNormNeck

* Update mmpretrain/models/necks/mae_neck.py

---------

Co-authored-by: Yixiao Fang <36138628+fangyixiao18@users.noreply.github.com>
2023-03-09 11:02:58 +08:00

31 lines
838 B
Python

model = dict(
type='ImageClassifier',
backbone=dict(
type='BEiTViT',
arch='l',
img_size=224,
patch_size=14,
layer_scale_init_value=0.0,
out_type='avg_featmap',
use_abs_pos_emb=True,
use_rel_pos_bias=False,
use_shared_rel_pos_bias=False,
layer_cfgs=dict(bias=True),
),
neck=None,
head=dict(
type='LinearClsHead',
num_classes=1000,
in_channels=1024,
loss=dict(
type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'),
),
init_cfg=[
dict(type='TruncNormal', layer='Linear', std=.02),
dict(type='Constant', layer='LayerNorm', val=1., bias=0.),
],
train_cfg=dict(augments=[
dict(type='Mixup', alpha=0.8),
dict(type='CutMix', alpha=1.0)
]))