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
7acb972a
Commit
7acb972a
authored
Dec 02, 2021
by
Yuqi Li
Committed by
A. Unique TensorFlower
Dec 02, 2021
Browse files
Internal change
PiperOrigin-RevId: 413820673
parent
78d99a22
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
50 additions
and
5 deletions
+50
-5
official/vision/beta/modeling/backbones/mobilenet.py
official/vision/beta/modeling/backbones/mobilenet.py
+37
-2
official/vision/beta/modeling/backbones/mobilenet_test.py
official/vision/beta/modeling/backbones/mobilenet_test.py
+13
-3
No files found.
official/vision/beta/modeling/backbones/mobilenet.py
View file @
7acb972a
...
@@ -420,7 +420,8 @@ MNMultiAVG_BLOCK_SPECS = {
...
@@ -420,7 +420,8 @@ MNMultiAVG_BLOCK_SPECS = {
# Similar to MobileNetMultiAVG and used for segmentation task.
# Similar to MobileNetMultiAVG and used for segmentation task.
# Reduced the filters by a factor of 2 in the last block.
# Reduced the filters by a factor of 2 in the last block.
MNMultiAVG_SEG_BLOCK_SPECS
=
{
MNMultiAVG_SEG_BLOCK_SPECS
=
{
'spec_name'
:
'MobileNetMultiAVGSeg'
,
'spec_name'
:
'MobileNetMultiAVGSeg'
,
'block_spec_schema'
:
[
'block_spec_schema'
:
[
'block_fn'
,
'kernel_size'
,
'strides'
,
'filters'
,
'activation'
,
'block_fn'
,
'kernel_size'
,
'strides'
,
'filters'
,
'activation'
,
'expand_ratio'
,
'use_normalization'
,
'use_bias'
,
'is_output'
'expand_ratio'
,
'use_normalization'
,
'use_bias'
,
'is_output'
...
@@ -443,7 +444,40 @@ MNMultiAVG_SEG_BLOCK_SPECS = {
...
@@ -443,7 +444,40 @@ MNMultiAVG_SEG_BLOCK_SPECS = {
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
2.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
2.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
4.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
4.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
4.
,
True
,
False
,
True
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
4.
,
True
,
False
,
True
),
(
'convbn'
,
1
,
1
,
480
,
'relu'
,
None
,
True
,
False
,
False
),
(
'convbn'
,
1
,
1
,
448
,
'relu'
,
None
,
True
,
False
,
True
),
(
'gpooling'
,
None
,
None
,
None
,
None
,
None
,
None
,
None
,
False
),
# Remove bias and add batch norm for the last layer to support QAT
# and achieve slightly better accuracy.
(
'convbn'
,
1
,
1
,
1280
,
'relu'
,
None
,
True
,
False
,
False
),
]
}
# Similar to MobileNetMultiMax and used for segmentation task.
# Reduced the filters by a factor of 2 in the last block.
MNMultiMAX_SEG_BLOCK_SPECS
=
{
'spec_name'
:
'MobileNetMultiMAXSeg'
,
'block_spec_schema'
:
[
'block_fn'
,
'kernel_size'
,
'strides'
,
'filters'
,
'activation'
,
'expand_ratio'
,
'use_normalization'
,
'use_bias'
,
'is_output'
],
'block_specs'
:
[
(
'convbn'
,
3
,
2
,
32
,
'relu'
,
None
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
2
,
32
,
'relu'
,
3.
,
True
,
False
,
True
),
(
'invertedbottleneck'
,
5
,
2
,
64
,
'relu'
,
6.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
64
,
'relu'
,
2.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
64
,
'relu'
,
2.
,
True
,
False
,
True
),
(
'invertedbottleneck'
,
5
,
2
,
128
,
'relu'
,
6.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
128
,
'relu'
,
4.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
128
,
'relu'
,
3.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
128
,
'relu'
,
3.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
128
,
'relu'
,
6.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
128
,
'relu'
,
3.
,
True
,
False
,
True
),
(
'invertedbottleneck'
,
3
,
2
,
160
,
'relu'
,
6.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
2.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
3
,
1
,
96
,
'relu'
,
4.
,
True
,
False
,
False
),
(
'invertedbottleneck'
,
5
,
1
,
96
,
'relu'
,
320.0
/
96
,
True
,
False
,
True
),
(
'convbn'
,
1
,
1
,
448
,
'relu'
,
None
,
True
,
False
,
True
),
(
'gpooling'
,
None
,
None
,
None
,
None
,
None
,
None
,
None
,
False
),
(
'gpooling'
,
None
,
None
,
None
,
None
,
None
,
None
,
None
,
False
),
# Remove bias and add batch norm for the last layer to support QAT
# Remove bias and add batch norm for the last layer to support QAT
# and achieve slightly better accuracy.
# and achieve slightly better accuracy.
...
@@ -460,6 +494,7 @@ SUPPORTED_SPECS_MAP = {
...
@@ -460,6 +494,7 @@ SUPPORTED_SPECS_MAP = {
'MobileNetMultiMAX'
:
MNMultiMAX_BLOCK_SPECS
,
'MobileNetMultiMAX'
:
MNMultiMAX_BLOCK_SPECS
,
'MobileNetMultiAVG'
:
MNMultiAVG_BLOCK_SPECS
,
'MobileNetMultiAVG'
:
MNMultiAVG_BLOCK_SPECS
,
'MobileNetMultiAVGSeg'
:
MNMultiAVG_SEG_BLOCK_SPECS
,
'MobileNetMultiAVGSeg'
:
MNMultiAVG_SEG_BLOCK_SPECS
,
'MobileNetMultiMAXSeg'
:
MNMultiMAX_SEG_BLOCK_SPECS
,
}
}
...
...
official/vision/beta/modeling/backbones/mobilenet_test.py
View file @
7acb972a
...
@@ -37,6 +37,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -37,6 +37,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiAVG'
,
'MobileNetMultiAVG'
,
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiMAXSeg'
,
)
)
def
test_serialize_deserialize
(
self
,
model_id
):
def
test_serialize_deserialize
(
self
,
model_id
):
# Create a network object that sets all of its config options.
# Create a network object that sets all of its config options.
...
@@ -82,6 +83,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -82,6 +83,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiAVG'
,
'MobileNetMultiAVG'
,
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiMAXSeg'
,
],
],
))
))
def
test_input_specs
(
self
,
input_dim
,
model_id
):
def
test_input_specs
(
self
,
input_dim
,
model_id
):
...
@@ -124,6 +126,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -124,6 +126,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiMAX'
:
[
32
,
64
,
128
,
160
],
'MobileNetMultiMAX'
:
[
32
,
64
,
128
,
160
],
'MobileNetMultiAVG'
:
[
32
,
64
,
160
,
192
],
'MobileNetMultiAVG'
:
[
32
,
64
,
160
,
192
],
'MobileNetMultiAVGSeg'
:
[
32
,
64
,
160
,
96
],
'MobileNetMultiAVGSeg'
:
[
32
,
64
,
160
,
96
],
'MobileNetMultiMAXSeg'
:
[
32
,
64
,
128
,
96
],
}
}
network
=
mobilenet
.
MobileNet
(
model_id
=
model_id
,
network
=
mobilenet
.
MobileNet
(
model_id
=
model_id
,
...
@@ -148,6 +151,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -148,6 +151,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiAVG'
,
'MobileNetMultiAVG'
,
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiMAXSeg'
,
],
],
[
32
,
224
],
[
32
,
224
],
))
))
...
@@ -167,6 +171,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -167,6 +171,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiMAX'
:
[
96
,
128
,
384
,
640
],
'MobileNetMultiMAX'
:
[
96
,
128
,
384
,
640
],
'MobileNetMultiAVG'
:
[
64
,
192
,
640
,
768
],
'MobileNetMultiAVG'
:
[
64
,
192
,
640
,
768
],
'MobileNetMultiAVGSeg'
:
[
64
,
192
,
640
,
384
],
'MobileNetMultiAVGSeg'
:
[
64
,
192
,
640
,
384
],
'MobileNetMultiMAXSeg'
:
[
96
,
128
,
384
,
320
],
}
}
network
=
mobilenet
.
MobileNet
(
model_id
=
model_id
,
network
=
mobilenet
.
MobileNet
(
model_id
=
model_id
,
filter_size_scale
=
1.0
,
filter_size_scale
=
1.0
,
...
@@ -196,6 +201,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -196,6 +201,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiMAXSeg'
,
],
],
[
1.0
,
0.75
],
[
1.0
,
0.75
],
))
))
...
@@ -217,8 +223,10 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -217,8 +223,10 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
(
'MobileNetMultiAVG'
,
0.75
):
2349704
,
(
'MobileNetMultiAVG'
,
0.75
):
2349704
,
(
'MobileNetMultiMAX'
,
1.0
):
3174560
,
(
'MobileNetMultiMAX'
,
1.0
):
3174560
,
(
'MobileNetMultiMAX'
,
0.75
):
2045816
,
(
'MobileNetMultiMAX'
,
0.75
):
2045816
,
(
'MobileNetMultiAVGSeg'
,
1.0
):
2284000
,
(
'MobileNetMultiAVGSeg'
,
1.0
):
2239840
,
(
'MobileNetMultiAVGSeg'
,
0.75
):
1427816
,
(
'MobileNetMultiAVGSeg'
,
0.75
):
1395272
,
(
'MobileNetMultiMAXSeg'
,
1.0
):
1929088
,
(
'MobileNetMultiMAXSeg'
,
0.75
):
1216544
,
}
}
input_size
=
224
input_size
=
224
...
@@ -241,6 +249,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -241,6 +249,7 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetMultiAVG'
,
'MobileNetMultiAVG'
,
'MobileNetMultiMAX'
,
'MobileNetMultiMAX'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiAVGSeg'
,
'MobileNetMultiMAXSeg'
,
],
],
[
8
,
16
,
32
],
[
8
,
16
,
32
],
))
))
...
@@ -258,7 +267,8 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
...
@@ -258,7 +267,8 @@ class MobileNetTest(parameterized.TestCase, tf.test.TestCase):
'MobileNetV3EdgeTPU'
:
192
,
'MobileNetV3EdgeTPU'
:
192
,
'MobileNetMultiMAX'
:
160
,
'MobileNetMultiMAX'
:
160
,
'MobileNetMultiAVG'
:
192
,
'MobileNetMultiAVG'
:
192
,
'MobileNetMultiAVGSeg'
:
96
,
'MobileNetMultiAVGSeg'
:
448
,
'MobileNetMultiMAXSeg'
:
448
,
}
}
network
=
mobilenet
.
MobileNet
(
network
=
mobilenet
.
MobileNet
(
...
...
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