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
de607e51
Commit
de607e51
authored
Oct 23, 2021
by
Vishnu Banna
Browse files
allowing threshold disable
parent
b056d20c
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
84 additions
and
1 deletion
+84
-1
official/vision/beta/projects/yolo/configs/experiments/scaled-yolo/detection/yolo_l_p5_640_tpu.yaml
.../experiments/scaled-yolo/detection/yolo_l_p5_640_tpu.yaml
+81
-0
official/vision/beta/projects/yolo/ops/preprocessing_ops.py
official/vision/beta/projects/yolo/ops/preprocessing_ops.py
+3
-1
No files found.
official/vision/beta/projects/yolo/configs/experiments/scaled-yolo/detection/yolo_l_p5_640_tpu.yaml
0 → 100644
View file @
de607e51
# --experiment_type=large_yolo
# mAP 50.5%
runtime
:
distribution_strategy
:
'
tpu'
mixed_precision_dtype
:
'
float32'
tpu_enable_xla_dynamic_padder
:
false
runtime
:
distribution_strategy
:
'
mirrored'
mixed_precision_dtype
:
'
float16'
tpu_enable_xla_dynamic_padder
:
false
num_gpus
:
1
task
:
model
:
input_size
:
[
896
,
896
,
3
]
backbone
:
type
:
'
darknet'
darknet
:
model_id
:
'
csp-large'
max_level
:
5
min_level
:
3
width_scale
:
1.00
depth_scale
:
1.00
decoder
:
type
:
yolo_decoder
yolo_decoder
:
version
:
v4
type
:
csp_large
head
:
smart_bias
:
true
detection_generator
:
box_type
:
'
all'
:
scaled
scale_xy
:
'
all'
:
2.0
max_boxes
:
300
nms_type
:
iou
iou_thresh
:
0.001
nms_thresh
:
0.65
loss
:
use_scaled_loss
:
true
update_on_repeat
:
true
box_loss_type
:
'
all'
:
ciou
ignore_thresh
:
'
all'
:
0.0
iou_normalizer
:
'
all'
:
0.05
cls_normalizer
:
'
all'
:
0.5
object_normalizer
:
'
5'
:
0.4
'
4'
:
1.0
'
3'
:
4.0
objectness_smooth
:
'
all'
:
1.0
norm_activation
:
use_sync_bn
:
true
num_classes
:
80
anchor_boxes
:
anchors_per_scale
:
4
boxes
:
[
box
:
[
7
,
9
],
box
:
[
17
,
14
],
box
:
[
13
,
29
],
box
:
[
34
,
25
],
box
:
[
27
,
58
],
box
:
[
68
,
54
],
box
:
[
55
,
108
],
box
:
[
124
,
105
],
box
:
[
97
,
219
],
box
:
[
185
,
257
],
box
:
[
352
,
353
],
box
:
[
457
,
457
]]
train_data
:
input_path
:
'
/readahead/200M/placer/prod/home/tensorflow-performance-data/datasets/coco/train*'
shuffle_buffer_size
:
10000
parser
:
mosaic
:
mosaic_frequency
:
1.0
mixup_frequency
:
0.0
mosaic_crop_mode
:
'
scale'
mosaic_center
:
0.0
aug_scale_min
:
0.5
aug_scale_max
:
1.5
max_num_instances
:
300
letter_box
:
true
random_flip
:
true
aug_rand_translate
:
0.5
area_thresh
:
0.1
validation_data
:
input_path
:
'
/readahead/200M/placer/prod/home/tensorflow-performance-data/datasets/coco/val*'
official/vision/beta/projects/yolo/ops/preprocessing_ops.py
View file @
de607e51
...
...
@@ -761,7 +761,9 @@ def boxes_candidates(clipped_boxes,
Returns:
indices[:, 0]: A `Tensor` representing valid boxes after filtering.
"""
if
area_thr
==
0.0
:
wh_thr
=
0
ar_thr
=
np
.
inf
area_thr
=
tf
.
math
.
abs
(
area_thr
)
# Get the scaled and shifted heights of the original
...
...
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