Skip to content

add omnibenchmarkv2 #1623

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: dev
Choose a base branch
from
Open

add omnibenchmarkv2 #1623

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
51 changes: 51 additions & 0 deletions configs/_base_/datasets/omnibenchmarkv2_bs32_pil_resize.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,51 @@
# dataset settings
dataset_type = 'ImageNet'
data_preprocessor = dict(
num_classes=1000,
# RGB format normalization parameters
mean=[123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375],
# convert image from BGR to RGB
to_rgb=True,
)

train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='RandomResizedCrop', scale=224, backend='pillow'),
dict(type='RandomFlip', prob=0.5, direction='horizontal'),
dict(type='PackInputs'),
]

test_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='ResizeEdge', scale=256, edge='short', backend='pillow'),
dict(type='CenterCrop', crop_size=224),
dict(type='PackInputs'),
]

train_dataloader = dict(
batch_size=32,
num_workers=4,
dataset=dict(
type=dataset_type,
data_root='data/omnibenchmarkv2/data/activity',
ann_file='data/omnibenchmarkv2/meta/activity/train.txt',
pipeline=train_pipeline),
sampler=dict(type='DefaultSampler', shuffle=True),
)

val_dataloader = dict(
batch_size=32,
num_workers=4,
dataset=dict(
type=dataset_type,
data_root='data/omnibenchmarkv2/data/activity',
ann_file='data/omnibenchmarkv2/meta/activity/test.txt',
pipeline=test_pipeline),
sampler=dict(type='DefaultSampler', shuffle=False),
)
val_evaluator = dict(type='Accuracy', topk=(1, 5))

# If you want standard test, please manually configure the test dataset
test_dataloader = val_dataloader
test_evaluator = val_evaluator
83 changes: 83 additions & 0 deletions configs/mae/benchmarks/vit-base-p16_4xb2048-linear-coslr-90e_omniv2-activity.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
_base_ = [
'../../_base_/datasets/omnibenchmarkv2_bs32_pil_resize.py',
'../../_base_/schedules/imagenet_bs1024_adamw_swin.py',
'../../_base_/default_runtime.py'
]

# dataset settings
data_preprocessor = dict(
num_classes=691,
)

train_dataloader = dict(batch_size=2048,
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/activity/meta/train.txt',
data_prefix='data/activity/images/'),
drop_last=True)
val_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/activity/meta/val.txt',
data_prefix='data/activity/images/'),
drop_last=False)
test_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/activity/meta/test.txt',
data_prefix='data/activity/images/'),
drop_last=False)

# model settings
model = dict(
type='ImageClassifier',
backbone=dict(
type='VisionTransformer',
arch='base',
img_size=224,
patch_size=16,
frozen_stages=12,
out_type='cls_token',
final_norm=True,
init_cfg=dict(type='Pretrained', checkpoint='/mnt/petrelfs/zhangyuanhan/weights/mae_vit-base-p16_8xb512-fp16-coslr-1600e_in1k_20220825-f7569ca2.pth', prefix="backbone.")),
neck=dict(type='ClsBatchNormNeck', input_features=768),
head=dict(
type='VisionTransformerClsHead',
num_classes=691,
in_channels=768,
loss=dict(type='CrossEntropyLoss'),
init_cfg=[dict(type='TruncNormal', layer='Linear', std=0.01)]))

# optimizer
optim_wrapper = dict(
_delete_=True,
type='AmpOptimWrapper',
optimizer=dict(type='LARS', lr=6.4, weight_decay=0.0, momentum=0.9))

# learning rate scheduler
param_scheduler = [
dict(
type='LinearLR',
start_factor=1e-4,
by_epoch=True,
begin=0,
end=10,
convert_to_iter_based=True),
dict(
type='CosineAnnealingLR',
T_max=80,
by_epoch=True,
begin=10,
end=90,
eta_min=0.0,
convert_to_iter_based=True)
]

# runtime settings
train_cfg = dict(by_epoch=True, max_epochs=90,val_interval=10)

default_hooks = dict(
checkpoint=dict(type='CheckpointHook', interval=1, max_keep_ckpts=3),
logger=dict(type='LoggerHook', interval=10))

randomness = dict(seed=0, diff_rank_seed=True)
83 changes: 83 additions & 0 deletions configs/mae/benchmarks/vit-base-p16_4xb2048-linear-coslr-90e_omniv2-bird.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
_base_ = [
'../../_base_/datasets/omnibenchmarkv2_bs32_pil_resize.py',
'../../_base_/schedules/imagenet_bs1024_adamw_swin.py',
'../../_base_/default_runtime.py'
]

# dataset settings
data_preprocessor = dict(
num_classes=646,
)

train_dataloader = dict(batch_size=2048,
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/bird/meta/train.txt',
data_prefix='data/bird/images/'),
drop_last=True)
val_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/bird/meta/val.txt',
data_prefix='data/bird/images/'),
drop_last=False)
test_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/bird/meta/test.txt',
data_prefix='data/bird/images/'),
drop_last=False)

# model settings
model = dict(
type='ImageClassifier',
backbone=dict(
type='VisionTransformer',
arch='base',
img_size=224,
patch_size=16,
frozen_stages=12,
out_type='cls_token',
final_norm=True,
init_cfg=dict(type='Pretrained', checkpoint='/mnt/petrelfs/zhangyuanhan/weights/mae_vit-base-p16_8xb512-fp16-coslr-1600e_in1k_20220825-f7569ca2.pth', prefix="backbone.")),
neck=dict(type='ClsBatchNormNeck', input_features=768),
head=dict(
type='VisionTransformerClsHead',
num_classes=646,
in_channels=768,
loss=dict(type='CrossEntropyLoss'),
init_cfg=[dict(type='TruncNormal', layer='Linear', std=0.01)]))

# optimizer
optim_wrapper = dict(
_delete_=True,
type='AmpOptimWrapper',
optimizer=dict(type='LARS', lr=6.4, weight_decay=0.0, momentum=0.9))

# learning rate scheduler
param_scheduler = [
dict(
type='LinearLR',
start_factor=1e-4,
by_epoch=True,
begin=0,
end=10,
convert_to_iter_based=True),
dict(
type='CosineAnnealingLR',
T_max=80,
by_epoch=True,
begin=10,
end=90,
eta_min=0.0,
convert_to_iter_based=True)
]

# runtime settings
train_cfg = dict(by_epoch=True, max_epochs=90,val_interval=10)

default_hooks = dict(
checkpoint=dict(type='CheckpointHook', interval=-1),
logger=dict(type='LoggerHook', interval=10))

randomness = dict(seed=0, diff_rank_seed=True)
83 changes: 83 additions & 0 deletions configs/mae/benchmarks/vit-base-p16_4xb2048-linear-coslr-90e_omniv2-car.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
_base_ = [
'../../_base_/datasets/omnibenchmarkv2_bs32_pil_resize.py',
'../../_base_/schedules/imagenet_bs1024_adamw_swin.py',
'../../_base_/default_runtime.py'
]

# dataset settings
data_preprocessor = dict(
num_classes=767,
)

train_dataloader = dict(batch_size=2048,
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/car/meta/train.txt',
data_prefix='data/car/images/'),
drop_last=True)
val_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/car/meta/val.txt',
data_prefix='data/car/images/'),
drop_last=False)
test_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/car/meta/test.txt',
data_prefix='data/car/images/'),
drop_last=False)

# model settings
model = dict(
type='ImageClassifier',
backbone=dict(
type='VisionTransformer',
arch='base',
img_size=224,
patch_size=16,
frozen_stages=12,
out_type='cls_token',
final_norm=True,
init_cfg=dict(type='Pretrained', checkpoint='/mnt/petrelfs/zhangyuanhan/weights/mae_vit-base-p16_8xb512-fp16-coslr-1600e_in1k_20220825-f7569ca2.pth', prefix="backbone.")),
neck=dict(type='ClsBatchNormNeck', input_features=768),
head=dict(
type='VisionTransformerClsHead',
num_classes=767,
in_channels=768,
loss=dict(type='CrossEntropyLoss'),
init_cfg=[dict(type='TruncNormal', layer='Linear', std=0.01)]))

# optimizer
optim_wrapper = dict(
_delete_=True,
type='AmpOptimWrapper',
optimizer=dict(type='LARS', lr=6.4, weight_decay=0.0, momentum=0.9))

# learning rate scheduler
param_scheduler = [
dict(
type='LinearLR',
start_factor=1e-4,
by_epoch=True,
begin=0,
end=10,
convert_to_iter_based=True),
dict(
type='CosineAnnealingLR',
T_max=80,
by_epoch=True,
begin=10,
end=90,
eta_min=0.0,
convert_to_iter_based=True)
]

# runtime settings
train_cfg = dict(by_epoch=True, max_epochs=90,val_interval=10)

default_hooks = dict(
checkpoint=dict(type='CheckpointHook', interval=-1),
logger=dict(type='LoggerHook', interval=10))

randomness = dict(seed=0, diff_rank_seed=True)
83 changes: 83 additions & 0 deletions configs/mae/benchmarks/vit-base-p16_4xb2048-linear-coslr-90e_omniv2-consumer_goods.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
_base_ = [
'../../_base_/datasets/omnibenchmarkv2_bs32_pil_resize.py',
'../../_base_/schedules/imagenet_bs1024_adamw_swin.py',
'../../_base_/default_runtime.py'
]

# dataset settings
data_preprocessor = dict(
num_classes=190,
)

train_dataloader = dict(batch_size=2048,
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/consumer_goods/meta/train.txt',
data_prefix='data/consumer_goods/images/'),
drop_last=True)
val_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/consumer_goods/meta/val.txt',
data_prefix='data/consumer_goods/images/'),
drop_last=False)
test_dataloader = dict(
dataset=dict(
data_root='data/omnibenchmarkv2/',
ann_file='annotation/consumer_goods/meta/test.txt',
data_prefix='data/consumer_goods/images/'),
drop_last=False)

# model settings
model = dict(
type='ImageClassifier',
backbone=dict(
type='VisionTransformer',
arch='base',
img_size=224,
patch_size=16,
frozen_stages=12,
out_type='cls_token',
final_norm=True,
init_cfg=dict(type='Pretrained', checkpoint='/mnt/petrelfs/zhangyuanhan/weights/mae_vit-base-p16_8xb512-fp16-coslr-1600e_in1k_20220825-f7569ca2.pth', prefix="backbone.")),
neck=dict(type='ClsBatchNormNeck', input_features=768),
head=dict(
type='VisionTransformerClsHead',
num_classes=190,
in_channels=768,
loss=dict(type='CrossEntropyLoss'),
init_cfg=[dict(type='TruncNormal', layer='Linear', std=0.01)]))

# optimizer
optim_wrapper = dict(
_delete_=True,
type='AmpOptimWrapper',
optimizer=dict(type='LARS', lr=6.4, weight_decay=0.0, momentum=0.9))

# learning rate scheduler
param_scheduler = [
dict(
type='LinearLR',
start_factor=1e-4,
by_epoch=True,
begin=0,
end=10,
convert_to_iter_based=True),
dict(
type='CosineAnnealingLR',
T_max=80,
by_epoch=True,
begin=10,
end=90,
eta_min=0.0,
convert_to_iter_based=True)
]

# runtime settings
train_cfg = dict(by_epoch=True, max_epochs=90,val_interval=10)

default_hooks = dict(
checkpoint=dict(type='CheckpointHook', interval=-1),
logger=dict(type='LoggerHook', interval=10))

randomness = dict(seed=0, diff_rank_seed=True)
Loading