Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
ModelZoo
ResNet50_tensorflow
Commits
6c9a23e2
Commit
6c9a23e2
authored
Aug 29, 2022
by
Cristina Vasconcelos
Committed by
A. Unique TensorFlower
Aug 29, 2022
Browse files
Internal change
PiperOrigin-RevId: 470729449
parent
4a8e45d1
Changes
4
Hide whitespace changes
Inline
Side-by-side
Showing
4 changed files
with
136 additions
and
0 deletions
+136
-0
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_fpn_600epochs.yaml
...eriments/retinanet/retinanet_resnet101_fpn_600epochs.yaml
+34
-0
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_fpn_72epochs.yaml
...periments/retinanet/retinanet_resnet101_fpn_72epochs.yaml
+34
-0
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_nasfpn_600epochs.yaml
...ments/retinanet/retinanet_resnet101_nasfpn_600epochs.yaml
+34
-0
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_nasfpn_72epochs.yaml
...iments/retinanet/retinanet_resnet101_nasfpn_72epochs.yaml
+34
-0
No files found.
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_fpn_600epochs.yaml
0 → 100644
View file @
6c9a23e2
task
:
# init_checkpoint: 'a_pretrained_backbone_checkpoint'
init_checkpoint_modules
:
backbone
freeze_backbone
:
true
model
:
backbone
:
resnet
:
model_id
:
101
replace_stem_max_pool
:
true
resnetd_shortcut
:
true
scale_stem
:
true
se_ratio
:
0.25
stem_type
:
v1
type
:
resnet
decoder
:
type
:
fpn
train_data
:
global_batch_size
:
256
parser
:
aug_rand_hflip
:
true
aug_scale_max
:
2.0
aug_scale_min
:
0.1
trainer
:
optimizer_config
:
learning_rate
:
stepwise
:
boundaries
:
[
265684
,
272615
]
name
:
PiecewiseConstantDecay
offset
:
0
values
:
[
0.32
,
0.032
,
0.0032
]
type
:
stepwise
steps_per_loop
:
462
summary_interval
:
462
train_steps
:
277235
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_fpn_72epochs.yaml
0 → 100644
View file @
6c9a23e2
task
:
# init_checkpoint: 'a_pretrained_backbone_checkpoint'
init_checkpoint_modules
:
backbone
freeze_backbone
:
true
model
:
backbone
:
resnet
:
model_id
:
101
replace_stem_max_pool
:
true
resnetd_shortcut
:
true
scale_stem
:
true
se_ratio
:
0.25
stem_type
:
v1
type
:
resnet
decoder
:
type
:
fpn
train_data
:
global_batch_size
:
256
parser
:
aug_rand_hflip
:
true
aug_scale_max
:
2.0
aug_scale_min
:
0.1
trainer
:
optimizer_config
:
learning_rate
:
stepwise
:
boundaries
:
[
22176
,
31416
]
name
:
PiecewiseConstantDecay
offset
:
0
values
:
[
0.32
,
0.032
,
0.0032
]
type
:
stepwise
steps_per_loop
:
462
summary_interval
:
462
train_steps
:
33264
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_nasfpn_600epochs.yaml
0 → 100644
View file @
6c9a23e2
task
:
# init_checkpoint: 'a_pretrained_backbone_checkpoint'
init_checkpoint_modules
:
backbone
freeze_backbone
:
true
model
:
backbone
:
resnet
:
model_id
:
101
replace_stem_max_pool
:
true
resnetd_shortcut
:
true
scale_stem
:
true
se_ratio
:
0.25
stem_type
:
v1
type
:
resnet
decoder
:
type
:
nasfpn
train_data
:
global_batch_size
:
256
parser
:
aug_rand_hflip
:
true
aug_scale_max
:
2.0
aug_scale_min
:
0.1
trainer
:
optimizer_config
:
learning_rate
:
stepwise
:
boundaries
:
[
265684
,
272615
]
name
:
PiecewiseConstantDecay
offset
:
0
values
:
[
0.32
,
0.032
,
0.0032
]
type
:
stepwise
steps_per_loop
:
462
summary_interval
:
462
train_steps
:
277235
official/projects/backbone_reuse/configs/experiments/retinanet/retinanet_resnet101_nasfpn_72epochs.yaml
0 → 100644
View file @
6c9a23e2
task
:
# init_checkpoint: 'a_pretrained_backbone_checkpoint'
init_checkpoint_modules
:
backbone
freeze_backbone
:
true
model
:
backbone
:
resnet
:
model_id
:
101
replace_stem_max_pool
:
true
resnetd_shortcut
:
true
scale_stem
:
true
se_ratio
:
0.25
stem_type
:
v1
type
:
resnet
decoder
:
type
:
nasfpn
train_data
:
global_batch_size
:
256
parser
:
aug_rand_hflip
:
true
aug_scale_max
:
2.0
aug_scale_min
:
0.1
trainer
:
optimizer_config
:
learning_rate
:
stepwise
:
boundaries
:
[
22176
,
31416
]
name
:
PiecewiseConstantDecay
offset
:
0
values
:
[
0.32
,
0.032
,
0.0032
]
type
:
stepwise
steps_per_loop
:
462
summary_interval
:
462
train_steps
:
33264
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment