Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
曾经的那一瞬间
Models
提交
4b3214c0
M
Models
项目概览
曾经的那一瞬间
/
Models
大约 1 年 前同步成功
通知
1
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
DevOps
流水线
流水线任务
计划
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
M
Models
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
DevOps
DevOps
流水线
流水线任务
计划
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
流水线任务
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
4b3214c0
编写于
11月 11, 2020
作者:
A
A. Unique TensorFlower
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Internal change
PiperOrigin-RevId: 341913626
上级
cb0ed243
变更
14
隐藏空白更改
内联
并排
Showing
14 changed file
with
0 addition
and
34 deletion
+0
-34
official/vision/beta/configs/experiments/retinanet/coco_spinenet143_tpu.yaml
...a/configs/experiments/retinanet/coco_spinenet143_tpu.yaml
+0
-1
official/vision/beta/configs/experiments/retinanet/coco_spinenet190_tpu.yaml
...a/configs/experiments/retinanet/coco_spinenet190_tpu.yaml
+0
-1
official/vision/beta/configs/experiments/retinanet/coco_spinenet49_tpu.yaml
...ta/configs/experiments/retinanet/coco_spinenet49_tpu.yaml
+0
-1
official/vision/beta/configs/experiments/retinanet/coco_spinenet96_tpu.yaml
...ta/configs/experiments/retinanet/coco_spinenet96_tpu.yaml
+0
-1
official/vision/beta/configs/image_classification.py
official/vision/beta/configs/image_classification.py
+0
-1
official/vision/beta/configs/maskrcnn.py
official/vision/beta/configs/maskrcnn.py
+0
-1
official/vision/beta/configs/retinanet.py
official/vision/beta/configs/retinanet.py
+0
-1
official/vision/beta/configs/semantic_segmentation.py
official/vision/beta/configs/semantic_segmentation.py
+0
-1
official/vision/beta/configs/video_classification.py
official/vision/beta/configs/video_classification.py
+0
-1
official/vision/beta/tasks/image_classification.py
official/vision/beta/tasks/image_classification.py
+0
-5
official/vision/beta/tasks/maskrcnn.py
official/vision/beta/tasks/maskrcnn.py
+0
-5
official/vision/beta/tasks/retinanet.py
official/vision/beta/tasks/retinanet.py
+0
-5
official/vision/beta/tasks/semantic_segmentation.py
official/vision/beta/tasks/semantic_segmentation.py
+0
-5
official/vision/beta/tasks/video_classification.py
official/vision/beta/tasks/video_classification.py
+0
-5
未找到文件。
official/vision/beta/configs/experiments/retinanet/coco_spinenet143_tpu.yaml
浏览文件 @
4b3214c0
...
...
@@ -4,7 +4,6 @@ runtime:
mixed_precision_dtype
:
'
bfloat16'
task
:
annotation_file
:
'
/readahead/200M/placer/prod/home/snaggletooth/test/data/coco/instances_val2017.json'
gradient_clip_norm
:
10.0
losses
:
l2_weight_decay
:
4.0e-05
model
:
...
...
official/vision/beta/configs/experiments/retinanet/coco_spinenet190_tpu.yaml
浏览文件 @
4b3214c0
...
...
@@ -3,7 +3,6 @@ runtime:
mixed_precision_dtype
:
'
bfloat16'
task
:
annotation_file
:
'
/readahead/200M/placer/prod/home/snaggletooth/test/data/coco/instances_val2017.json'
gradient_clip_norm
:
0.0
losses
:
l2_weight_decay
:
4.0e-05
model
:
...
...
official/vision/beta/configs/experiments/retinanet/coco_spinenet49_tpu.yaml
浏览文件 @
4b3214c0
...
...
@@ -4,7 +4,6 @@ runtime:
mixed_precision_dtype
:
'
bfloat16'
task
:
annotation_file
:
'
/readahead/200M/placer/prod/home/snaggletooth/test/data/coco/instances_val2017.json'
gradient_clip_norm
:
0.0
losses
:
l2_weight_decay
:
4.0e-05
model
:
...
...
official/vision/beta/configs/experiments/retinanet/coco_spinenet96_tpu.yaml
浏览文件 @
4b3214c0
...
...
@@ -4,7 +4,6 @@ runtime:
mixed_precision_dtype
:
'
bfloat16'
task
:
annotation_file
:
'
/readahead/200M/placer/prod/home/snaggletooth/test/data/coco/instances_val2017.json'
gradient_clip_norm
:
0.0
losses
:
l2_weight_decay
:
4.0e-05
model
:
...
...
official/vision/beta/configs/image_classification.py
浏览文件 @
4b3214c0
...
...
@@ -70,7 +70,6 @@ class ImageClassificationTask(cfg.TaskConfig):
validation_data
:
DataConfig
=
DataConfig
(
is_training
=
False
)
losses
:
Losses
=
Losses
()
evaluation
:
Evaluation
=
Evaluation
()
gradient_clip_norm
:
float
=
0.0
init_checkpoint
:
Optional
[
str
]
=
None
init_checkpoint_modules
:
str
=
'all'
# all or backbone
...
...
official/vision/beta/configs/maskrcnn.py
浏览文件 @
4b3214c0
...
...
@@ -207,7 +207,6 @@ class MaskRCNNTask(cfg.TaskConfig):
init_checkpoint
:
Optional
[
str
]
=
None
init_checkpoint_modules
:
str
=
'all'
# all or backbone
annotation_file
:
Optional
[
str
]
=
None
gradient_clip_norm
:
float
=
0.0
per_category_metrics
:
bool
=
False
# If set, we only use masks for the specified class IDs.
allowed_mask_class_ids
:
Optional
[
List
[
int
]]
=
None
...
...
official/vision/beta/configs/retinanet.py
浏览文件 @
4b3214c0
...
...
@@ -128,7 +128,6 @@ class RetinaNetTask(cfg.TaskConfig):
init_checkpoint
:
Optional
[
str
]
=
None
init_checkpoint_modules
:
str
=
'all'
# all or backbone
annotation_file
:
Optional
[
str
]
=
None
gradient_clip_norm
:
float
=
0.0
per_category_metrics
:
bool
=
False
...
...
official/vision/beta/configs/semantic_segmentation.py
浏览文件 @
4b3214c0
...
...
@@ -87,7 +87,6 @@ class SemanticSegmentationTask(cfg.TaskConfig):
train_data
:
DataConfig
=
DataConfig
(
is_training
=
True
)
validation_data
:
DataConfig
=
DataConfig
(
is_training
=
False
)
losses
:
Losses
=
Losses
()
gradient_clip_norm
:
float
=
0.0
init_checkpoint
:
Optional
[
str
]
=
None
init_checkpoint_modules
:
Union
[
str
,
List
[
str
]]
=
'all'
# all, backbone, and/or decoder
...
...
official/vision/beta/configs/video_classification.py
浏览文件 @
4b3214c0
...
...
@@ -97,7 +97,6 @@ class VideoClassificationTask(cfg.TaskConfig):
train_data
:
DataConfig
=
DataConfig
(
is_training
=
True
)
validation_data
:
DataConfig
=
DataConfig
(
is_training
=
False
)
losses
:
Losses
=
Losses
()
gradient_clip_norm
:
float
=
-
1.0
def
add_trainer
(
experiment
:
cfg
.
ExperimentConfig
,
...
...
official/vision/beta/tasks/image_classification.py
浏览文件 @
4b3214c0
...
...
@@ -180,11 +180,6 @@ class ImageClassificationTask(base_task.Task):
if
isinstance
(
optimizer
,
tf
.
keras
.
mixed_precision
.
LossScaleOptimizer
):
grads
=
optimizer
.
get_unscaled_gradients
(
grads
)
# Apply gradient clipping.
if
self
.
task_config
.
gradient_clip_norm
>
0
:
grads
,
_
=
tf
.
clip_by_global_norm
(
grads
,
self
.
task_config
.
gradient_clip_norm
)
optimizer
.
apply_gradients
(
list
(
zip
(
grads
,
tvars
)))
logs
=
{
self
.
loss
:
loss
}
...
...
official/vision/beta/tasks/maskrcnn.py
浏览文件 @
4b3214c0
...
...
@@ -280,11 +280,6 @@ class MaskRCNNTask(base_task.Task):
# Scales back gradient when LossScaleOptimizer is used.
if
isinstance
(
optimizer
,
tf
.
keras
.
mixed_precision
.
LossScaleOptimizer
):
grads
=
optimizer
.
get_unscaled_gradients
(
grads
)
# Apply gradient clipping.
if
self
.
task_config
.
gradient_clip_norm
>
0
:
grads
,
_
=
tf
.
clip_by_global_norm
(
grads
,
self
.
task_config
.
gradient_clip_norm
)
optimizer
.
apply_gradients
(
list
(
zip
(
grads
,
tvars
)))
logs
=
{
self
.
loss
:
losses
[
'total_loss'
]}
...
...
official/vision/beta/tasks/retinanet.py
浏览文件 @
4b3214c0
...
...
@@ -218,11 +218,6 @@ class RetinaNetTask(base_task.Task):
# Scales back gradient when LossScaleOptimizer is used.
if
isinstance
(
optimizer
,
tf
.
keras
.
mixed_precision
.
LossScaleOptimizer
):
grads
=
optimizer
.
get_unscaled_gradients
(
grads
)
# Apply gradient clipping.
if
self
.
task_config
.
gradient_clip_norm
>
0
:
grads
,
_
=
tf
.
clip_by_global_norm
(
grads
,
self
.
task_config
.
gradient_clip_norm
)
optimizer
.
apply_gradients
(
list
(
zip
(
grads
,
tvars
)))
logs
=
{
self
.
loss
:
loss
}
...
...
official/vision/beta/tasks/semantic_segmentation.py
浏览文件 @
4b3214c0
...
...
@@ -188,11 +188,6 @@ class SemanticSegmentationTask(base_task.Task):
# used.
if
isinstance
(
optimizer
,
tf
.
keras
.
mixed_precision
.
LossScaleOptimizer
):
grads
=
optimizer
.
get_unscaled_gradients
(
grads
)
# Apply gradient clipping.
if
self
.
task_config
.
gradient_clip_norm
>
0
:
grads
,
_
=
tf
.
clip_by_global_norm
(
grads
,
self
.
task_config
.
gradient_clip_norm
)
optimizer
.
apply_gradients
(
list
(
zip
(
grads
,
tvars
)))
logs
=
{
self
.
loss
:
loss
}
...
...
official/vision/beta/tasks/video_classification.py
浏览文件 @
4b3214c0
...
...
@@ -160,11 +160,6 @@ class VideoClassificationTask(base_task.Task):
# used.
if
isinstance
(
optimizer
,
tf
.
keras
.
mixed_precision
.
LossScaleOptimizer
):
grads
=
optimizer
.
get_unscaled_gradients
(
grads
)
# Apply gradient clipping.
if
self
.
task_config
.
gradient_clip_norm
>
0
:
grads
,
_
=
tf
.
clip_by_global_norm
(
grads
,
self
.
task_config
.
gradient_clip_norm
)
optimizer
.
apply_gradients
(
list
(
zip
(
grads
,
tvars
)))
logs
=
{
self
.
loss
:
loss
}
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录