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
PaddlePaddle
models
提交
e23964cf
M
models
项目概览
PaddlePaddle
/
models
大约 1 年 前同步成功
通知
222
Star
6828
Fork
2962
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
602
列表
看板
标记
里程碑
合并请求
255
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
602
Issue
602
列表
看板
标记
里程碑
合并请求
255
合并请求
255
Pages
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
提交
Issue看板
提交
e23964cf
编写于
4月 11, 2018
作者:
D
Dang Qingqing
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Enable ParallelExecutor in SSD-MobileNet and Refine code.
上级
6fa8a94b
变更
9
隐藏空白更改
内联
并排
Showing
9 changed file
with
211 addition
and
137 deletion
+211
-137
fluid/object_detection/data/pascalvoc/create_list.py
fluid/object_detection/data/pascalvoc/create_list.py
+2
-1
fluid/object_detection/data/pascalvoc/download.sh
fluid/object_detection/data/pascalvoc/download.sh
+16
-0
fluid/object_detection/data/pascalvoc/label_list
fluid/object_detection/data/pascalvoc/label_list
+0
-0
fluid/object_detection/load_model.py
fluid/object_detection/load_model.py
+0
-67
fluid/object_detection/mobilenet_ssd.py
fluid/object_detection/mobilenet_ssd.py
+4
-6
fluid/object_detection/pretrained/download_coco.sh
fluid/object_detection/pretrained/download_coco.sh
+8
-0
fluid/object_detection/pretrained/download_imagenet.sh
fluid/object_detection/pretrained/download_imagenet.sh
+8
-0
fluid/object_detection/reader.py
fluid/object_detection/reader.py
+6
-6
fluid/object_detection/train.py
fluid/object_detection/train.py
+167
-57
未找到文件。
fluid/object_detection/data/p
repare_voc_data
.py
→
fluid/object_detection/data/p
ascalvoc/create_list
.py
浏览文件 @
e23964cf
...
...
@@ -60,4 +60,5 @@ def prepare_filelist(devkit_dir, years, output_dir):
ftest
.
write
(
item
[
0
]
+
' '
+
item
[
1
]
+
'
\n
'
)
prepare_filelist
(
devkit_dir
,
years
,
'.'
)
if
__name__
==
'__main__'
:
prepare_filelist
(
devkit_dir
,
years
,
'.'
)
fluid/object_detection/data/pascalvoc/download.sh
0 → 100644
浏览文件 @
e23964cf
DIR
=
"
$(
cd
"
$(
dirname
"
$0
"
)
"
;
pwd
-P
)
"
cd
"
$DIR
"
# Download the data.
echo
"Downloading..."
wget http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar
wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tar
wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tar
# Extract the data.
echo
"Extractint..."
tar
-xf
VOCtrainval_11-May-2012.tar
tar
-xf
VOCtrainval_06-Nov-2007.tar
tar
-xf
VOCtest_06-Nov-2007.tar
echo
"Creating data lists..."
python create_list.py
fluid/object_detection/data/label_list
→
fluid/object_detection/data/
pascalvoc/
label_list
浏览文件 @
e23964cf
文件已移动
fluid/object_detection/load_model.py
已删除
100644 → 0
浏览文件 @
6fa8a94b
import
paddle.v2
as
paddle
import
paddle.fluid
as
fluid
import
numpy
as
np
# From npy
def
load_vars
():
vars
=
{}
name_map
=
{}
with
open
(
'./ssd_mobilenet_v1_coco/names.map'
,
'r'
)
as
map_file
:
for
param
in
map_file
:
fd_name
,
tf_name
=
param
.
strip
().
split
(
'
\t
'
)
name_map
[
fd_name
]
=
tf_name
tf_vars
=
np
.
load
(
'./ssd_mobilenet_v1_coco/ssd_mobilenet_v1_coco_2017_11_17.npy'
).
item
()
for
fd_name
in
name_map
:
tf_name
=
name_map
[
fd_name
]
tf_var
=
tf_vars
[
tf_name
]
if
len
(
tf_var
.
shape
)
==
4
and
'depthwise'
in
tf_name
:
vars
[
fd_name
]
=
np
.
transpose
(
tf_var
,
(
2
,
3
,
0
,
1
))
elif
len
(
tf_var
.
shape
)
==
4
:
vars
[
fd_name
]
=
np
.
transpose
(
tf_var
,
(
3
,
2
,
0
,
1
))
else
:
vars
[
fd_name
]
=
tf_var
return
vars
def
load_and_set_vars
(
place
):
vars
=
load_vars
()
for
k
,
v
in
vars
.
items
():
t
=
fluid
.
global_scope
().
find_var
(
k
).
get_tensor
()
#print(np.array(t).shape, v.shape, k)
assert
np
.
array
(
t
).
shape
==
v
.
shape
t
.
set
(
v
,
place
)
# From Paddle V1
def
load_paddlev1_vars
(
place
):
vars
=
{}
name_map
=
{}
with
open
(
'./caffe2paddle/names.map'
,
'r'
)
as
map_file
:
for
param
in
map_file
:
fd_name
,
tf_name
=
param
.
strip
().
split
(
'
\t
'
)
name_map
[
fd_name
]
=
tf_name
from
operator
import
mul
def
load
(
file_name
,
shape
):
with
open
(
file_name
,
'rb'
)
as
f
:
f
.
read
(
16
)
arr
=
np
.
fromfile
(
f
,
dtype
=
np
.
float32
)
#print(arr.size, reduce(mul, shape), file_name)
assert
arr
.
size
==
reduce
(
mul
,
shape
)
return
arr
.
reshape
(
shape
)
for
fd_name
in
name_map
:
v1_name
=
name_map
[
fd_name
]
t
=
fluid
.
global_scope
().
find_var
(
fd_name
).
get_tensor
()
shape
=
np
.
array
(
t
).
shape
v1_var
=
load
(
'./caffe2paddle/'
+
v1_name
,
shape
)
t
.
set
(
v1_var
,
place
)
if
__name__
==
"__main__"
:
load_vars
()
fluid/object_detection/mobilenet_ssd.py
浏览文件 @
e23964cf
...
...
@@ -13,7 +13,7 @@ def conv_bn(input,
num_groups
=
1
,
act
=
'relu'
,
use_cudnn
=
True
):
parameter_attr
=
ParamAttr
(
initializer
=
MSRA
())
parameter_attr
=
ParamAttr
(
learning_rate
=
0.1
,
initializer
=
MSRA
())
conv
=
fluid
.
layers
.
conv2d
(
input
=
input
,
num_filters
=
num_filters
,
...
...
@@ -25,11 +25,9 @@ def conv_bn(input,
use_cudnn
=
use_cudnn
,
param_attr
=
parameter_attr
,
bias_attr
=
False
)
#parameter_attr = ParamAttr(learning_rate=0.1, initializer=MSRA())
#bias_attr = ParamAttr(learning_rate=0.2)
return
fluid
.
layers
.
batch_norm
(
input
=
conv
,
act
=
act
,
epsilon
=
0.00001
)
#param_attr=parameter_attr,
#bias_attr=bias_attr)
parameter_attr
=
ParamAttr
(
learning_rate
=
0.1
,
initializer
=
MSRA
())
bias_attr
=
ParamAttr
(
learning_rate
=
0.2
)
return
fluid
.
layers
.
batch_norm
(
input
=
conv
,
act
=
act
)
def
depthwise_separable
(
input
,
num_filters1
,
num_filters2
,
num_groups
,
stride
,
...
...
fluid/object_detection/pretrained/download_coco.sh
0 → 100644
浏览文件 @
e23964cf
DIR
=
"
$(
cd
"
$(
dirname
"
$0
"
)
"
;
pwd
-P
)
"
cd
"
$DIR
"
# Download the data.
echo
"Downloading..."
wget http://paddlemodels.bj.bcebos.com/ssd_mobilenet_coco.tar.gz
echo
"Extractint..."
tar
-xf
ssd_mobilenet_coco.tar.gz
fluid/object_detection/pretrained/download_imagenet.sh
0 → 100644
浏览文件 @
e23964cf
DIR
=
"
$(
cd
"
$(
dirname
"
$0
"
)
"
;
pwd
-P
)
"
cd
"
$DIR
"
# Download the data.
echo
"Downloading..."
wget http://paddlemodels.bj.bcebos.com/mobilenet_imagenet.tar.gz
echo
"Extractint..."
tar
-xf
ssd_mobilenet_imagenet.tar.gz
fluid/object_detection/reader.py
浏览文件 @
e23964cf
...
...
@@ -23,10 +23,6 @@ import os
import
time
import
copy
# cocoapi
from
pycocotools.coco
import
COCO
from
pycocotools.cocoeval
import
COCOeval
class
Settings
(
object
):
def
__init__
(
self
,
dataset
,
toy
,
data_dir
,
label_file
,
resize_h
,
resize_w
,
...
...
@@ -101,6 +97,10 @@ class Settings(object):
def
_reader_creator
(
settings
,
file_list
,
mode
,
shuffle
):
def
reader
():
if
settings
.
dataset
==
'coco'
:
# cocoapi
from
pycocotools.coco
import
COCO
from
pycocotools.cocoeval
import
COCOeval
coco
=
COCO
(
file_list
)
image_ids
=
coco
.
getImgIds
()
images
=
coco
.
loadImgs
(
image_ids
)
...
...
@@ -295,6 +295,7 @@ def draw_bounding_box_on_image(image,
def
train
(
settings
,
file_list
,
shuffle
=
True
):
file_list
=
os
.
path
.
join
(
settings
.
data_dir
,
file_list
)
if
settings
.
dataset
==
'coco'
:
train_settings
=
copy
.
copy
(
settings
)
if
'2014'
in
file_list
:
...
...
@@ -302,13 +303,13 @@ def train(settings, file_list, shuffle=True):
elif
'2017'
in
file_list
:
sub_dir
=
"train2017"
train_settings
.
data_dir
=
os
.
path
.
join
(
settings
.
data_dir
,
sub_dir
)
file_list
=
os
.
path
.
join
(
settings
.
data_dir
,
file_list
)
return
_reader_creator
(
train_settings
,
file_list
,
'train'
,
shuffle
)
elif
settings
.
dataset
==
'pascalvoc'
:
return
_reader_creator
(
settings
,
file_list
,
'train'
,
shuffle
)
def
test
(
settings
,
file_list
):
file_list
=
os
.
path
.
join
(
settings
.
data_dir
,
file_list
)
if
settings
.
dataset
==
'coco'
:
test_settings
=
copy
.
copy
(
settings
)
if
'2014'
in
file_list
:
...
...
@@ -316,7 +317,6 @@ def test(settings, file_list):
elif
'2017'
in
file_list
:
sub_dir
=
"val2017"
test_settings
.
data_dir
=
os
.
path
.
join
(
settings
.
data_dir
,
sub_dir
)
file_list
=
os
.
path
.
join
(
settings
.
data_dir
,
file_list
)
return
_reader_creator
(
test_settings
,
file_list
,
'test'
,
False
)
elif
settings
.
dataset
==
'pascalvoc'
:
return
_reader_creator
(
settings
,
file_list
,
'test'
,
False
)
...
...
fluid/object_detection/train.py
浏览文件 @
e23964cf
...
...
@@ -12,46 +12,35 @@ import functools
parser
=
argparse
.
ArgumentParser
(
description
=
__doc__
)
add_arg
=
functools
.
partial
(
add_arguments
,
argparser
=
parser
)
add_arg
(
'learning_rate'
,
float
,
0.001
,
"Learning rate."
)
add_arg
(
'batch_size'
,
int
,
32
,
"Minibatch size."
)
add_arg
(
'num_passes'
,
int
,
25
,
"Epoch number."
)
add_arg
(
'parallel'
,
bool
,
True
,
"Whether use parallel training."
)
add_arg
(
'use_gpu'
,
bool
,
True
,
"Whether use GPU."
)
add_arg
(
'data_dir'
,
str
,
'./data/COCO17'
,
"Root path of data"
)
add_arg
(
'train_file_list'
,
str
,
'annotations/instances_train2017.json'
,
"train file list"
)
add_arg
(
'val_file_list'
,
str
,
'annotations/instances_val2017.json'
,
"vaild file list"
)
add_arg
(
'model_save_dir'
,
str
,
'model_COCO17'
,
"where to save model"
)
add_arg
(
'dataset'
,
str
,
'coco'
,
"coco or pascalvoc"
)
add_arg
(
'is_toy'
,
int
,
0
,
"Is Toy for quick debug, 0 means using all data, while n means using only n sample"
)
add_arg
(
'label_file'
,
str
,
'label_list'
,
"Lable file which lists all label name"
)
add_arg
(
'apply_distort'
,
bool
,
True
,
"Whether apply distort"
)
add_arg
(
'apply_expand'
,
bool
,
False
,
"Whether appley expand"
)
add_arg
(
'resize_h'
,
int
,
300
,
"resize image size"
)
add_arg
(
'resize_w'
,
int
,
300
,
"resize image size"
)
add_arg
(
'mean_value_B'
,
float
,
127.5
,
"mean value which will be subtracted"
)
#123.68
add_arg
(
'mean_value_G'
,
float
,
127.5
,
"mean value which will be subtracted"
)
#116.78
add_arg
(
'mean_value_R'
,
float
,
127.5
,
"mean value which will be subtracted"
)
#103.94
def
train
(
args
,
train_file_list
,
val_file_list
,
data_args
,
learning_rate
,
batch_size
,
num_passes
,
model_save_dir
=
'model'
,
init_model_path
=
None
):
# yapf: disable
add_arg
(
'learning_rate'
,
float
,
0.001
,
"Learning rate."
)
add_arg
(
'batch_size'
,
int
,
32
,
"Minibatch size."
)
add_arg
(
'num_passes'
,
int
,
25
,
"Epoch number."
)
add_arg
(
'parallel'
,
bool
,
True
,
"Whether use parallel training."
)
add_arg
(
'use_gpu'
,
bool
,
True
,
"Whether use GPU."
)
add_arg
(
'dataset'
,
str
,
'pascalvoc'
,
"coco or pascalvoc."
)
add_arg
(
'model_save_dir'
,
str
,
'model'
,
"The path to save model."
)
add_arg
(
'pretrained_model'
,
str
,
'pretrained/ssd_mobilenet_coco/'
,
"The init model path."
)
add_arg
(
'apply_distort'
,
bool
,
True
,
"Whether apply distort"
)
add_arg
(
'apply_expand'
,
bool
,
False
,
"Whether appley expand"
)
add_arg
(
'resize_h'
,
int
,
300
,
"resize image size"
)
add_arg
(
'resize_w'
,
int
,
300
,
"resize image size"
)
add_arg
(
'mean_value_B'
,
float
,
127.5
,
"mean value which will be subtracted"
)
#123.68
add_arg
(
'mean_value_G'
,
float
,
127.5
,
"mean value which will be subtracted"
)
#116.78
add_arg
(
'mean_value_R'
,
float
,
127.5
,
"mean value which will be subtracted"
)
#103.94
add_arg
(
'is_toy'
,
int
,
0
,
"Toy for quick debug, 0 means using all data, while n means using only n sample"
)
# yapf: disable
def
parallel_do
(
args
,
train_file_list
,
val_file_list
,
data_args
,
learning_rate
,
batch_size
,
num_passes
,
model_save_dir
,
pretrained_model
=
None
):
image_shape
=
[
3
,
data_args
.
resize_h
,
data_args
.
resize_w
]
if
data_args
.
dataset
==
'coco'
:
num_classes
=
81
...
...
@@ -125,8 +114,11 @@ def train(args,
exe
=
fluid
.
Executor
(
place
)
exe
.
run
(
fluid
.
default_startup_program
())
#load_model.load_and_set_vars(place)
load_model
.
load_paddlev1_vars
(
place
)
if
pretrained_model
:
def
if_exist
(
var
):
return
os
.
path
.
exists
(
os
.
path
.
join
(
pretrained_model
,
var
.
name
))
fluid
.
io
.
load_vars
(
exe
,
pretrained_model
,
predicate
=
if_exist
)
train_reader
=
paddle
.
batch
(
reader
.
train
(
data_args
,
train_file_list
),
batch_size
=
batch_size
)
test_reader
=
paddle
.
batch
(
...
...
@@ -151,7 +143,6 @@ def train(args,
for
batch_id
,
data
in
enumerate
(
train_reader
()):
prev_start_time
=
start_time
start_time
=
time
.
time
()
#print("Batch {} start at {:.2f}".format(batch_id, start_time))
loss_v
=
exe
.
run
(
fluid
.
default_main_program
(),
feed
=
feeder
.
feed
(
data
),
fetch_list
=
[
loss
])
...
...
@@ -164,29 +155,148 @@ def train(args,
if
pass_id
%
10
==
0
or
pass_id
==
num_passes
-
1
:
model_path
=
os
.
path
.
join
(
model_save_dir
,
str
(
pass_id
))
print
'save models to %s'
%
(
model_path
)
fluid
.
io
.
save_inference_model
(
model_path
,
[
'image'
],
[
nmsed_out
],
exe
)
fluid
.
io
.
save_persistables
(
exe
,
model_path
)
def
parallel_exe
(
args
,
train_file_list
,
val_file_list
,
data_args
,
learning_rate
,
batch_size
,
num_passes
,
model_save_dir
=
'model'
,
pretrained_model
=
None
):
image_shape
=
[
3
,
data_args
.
resize_h
,
data_args
.
resize_w
]
if
data_args
.
dataset
==
'coco'
:
num_classes
=
81
elif
data_args
.
dataset
==
'pascalvoc'
:
num_classes
=
21
image
=
fluid
.
layers
.
data
(
name
=
'image'
,
shape
=
image_shape
,
dtype
=
'float32'
)
gt_box
=
fluid
.
layers
.
data
(
name
=
'gt_box'
,
shape
=
[
4
],
dtype
=
'float32'
,
lod_level
=
1
)
gt_label
=
fluid
.
layers
.
data
(
name
=
'gt_label'
,
shape
=
[
1
],
dtype
=
'int32'
,
lod_level
=
1
)
difficult
=
fluid
.
layers
.
data
(
name
=
'gt_difficult'
,
shape
=
[
1
],
dtype
=
'int32'
,
lod_level
=
1
)
locs
,
confs
,
box
,
box_var
=
mobile_net
(
num_classes
,
image
,
image_shape
)
nmsed_out
=
fluid
.
layers
.
detection_output
(
locs
,
confs
,
box
,
box_var
,
nms_threshold
=
0.45
)
loss
=
fluid
.
layers
.
ssd_loss
(
locs
,
confs
,
gt_box
,
gt_label
,
box
,
box_var
)
loss
=
fluid
.
layers
.
reduce_sum
(
loss
)
test_program
=
fluid
.
default_main_program
().
clone
(
for_test
=
True
)
with
fluid
.
program_guard
(
test_program
):
map_eval
=
fluid
.
evaluator
.
DetectionMAP
(
nmsed_out
,
gt_label
,
gt_box
,
difficult
,
num_classes
,
overlap_threshold
=
0.5
,
evaluate_difficult
=
False
,
ap_version
=
'integral'
)
if
data_args
.
dataset
==
'coco'
:
# learning rate decay in 12, 19 pass, respectively
if
'2014'
in
train_file_list
:
boundaries
=
[
82783
/
batch_size
*
12
,
82783
/
batch_size
*
19
]
elif
'2017'
in
train_file_list
:
boundaries
=
[
118287
/
batch_size
*
12
,
118287
/
batch_size
*
19
]
elif
data_args
.
dataset
==
'pascalvoc'
:
boundaries
=
[
40000
,
60000
]
values
=
[
learning_rate
,
learning_rate
*
0.5
,
learning_rate
*
0.25
]
optimizer
=
fluid
.
optimizer
.
RMSProp
(
learning_rate
=
fluid
.
layers
.
piecewise_decay
(
boundaries
,
values
),
regularization
=
fluid
.
regularizer
.
L2Decay
(
0.00005
),
)
optimizer
.
minimize
(
loss
)
place
=
fluid
.
CUDAPlace
(
0
)
if
args
.
use_gpu
else
fluid
.
CPUPlace
()
exe
=
fluid
.
Executor
(
place
)
exe
.
run
(
fluid
.
default_startup_program
())
if
pretrained_model
:
def
if_exist
(
var
):
return
os
.
path
.
exists
(
os
.
path
.
join
(
pretrained_model
,
var
.
name
))
fluid
.
io
.
load_vars
(
exe
,
pretrained_model
,
predicate
=
if_exist
)
train_exe
=
fluid
.
ParallelExecutor
(
use_cuda
=
True
,
loss_name
=
loss
.
name
)
train_reader
=
paddle
.
batch
(
reader
.
train
(
data_args
,
train_file_list
),
batch_size
=
batch_size
)
test_reader
=
paddle
.
batch
(
reader
.
test
(
data_args
,
val_file_list
),
batch_size
=
batch_size
)
feeder
=
fluid
.
DataFeeder
(
place
=
place
,
feed_list
=
[
image
,
gt_box
,
gt_label
,
difficult
])
def
test
(
pass_id
):
_
,
accum_map
=
map_eval
.
get_map_var
()
map_eval
.
reset
(
exe
)
test_map
=
None
for
_
,
data
in
enumerate
(
test_reader
()):
test_map
=
exe
.
run
(
test_program
,
feed
=
feeder
.
feed
(
data
),
fetch_list
=
[
accum_map
])
print
(
"Test {0}, map {1}"
.
format
(
pass_id
,
test_map
[
0
]))
for
pass_id
in
range
(
num_passes
):
start_time
=
time
.
time
()
prev_start_time
=
start_time
end_time
=
0
test
(
pass_id
)
for
batch_id
,
data
in
enumerate
(
train_reader
()):
prev_start_time
=
start_time
start_time
=
time
.
time
()
loss_v
,
=
train_exe
.
run
(
fetch_list
=
[
loss
.
name
],
feed_dict
=
feeder
.
feed
(
data
))
end_time
=
time
.
time
()
loss_v
=
np
.
mean
(
np
.
array
(
loss_v
))
if
batch_id
%
20
==
0
:
print
(
"Pass {0}, batch {1}, loss {2}, time {3}"
.
format
(
pass_id
,
batch_id
,
loss_v
,
start_time
-
prev_start_time
))
if
pass_id
%
10
==
0
or
pass_id
==
num_passes
-
1
:
model_path
=
os
.
path
.
join
(
model_save_dir
,
str
(
pass_id
))
print
'save models to %s'
%
(
model_path
)
fluid
.
io
.
save_persistables
(
exe
,
model_path
)
if
__name__
==
'__main__'
:
args
=
parser
.
parse_args
()
print_arguments
(
args
)
data_dir
=
'data/pascalvoc'
train_file_list
=
'trainval.txt'
val_file_list
=
'test.txt'
label_file
=
'label_list'
model_save_dir
=
args
.
model_save_dir
if
args
.
dataset
==
'coco'
:
data_dir
=
'./data/COCO17'
train_file_list
=
'annotations/instances_train2017.json'
val_file_list
=
'annotations/instances_val2017.json'
label_file
=
'label_list'
data_args
=
reader
.
Settings
(
dataset
=
args
.
dataset
,
# coco or pascalvoc
dataset
=
args
.
dataset
,
toy
=
args
.
is_toy
,
data_dir
=
args
.
data_dir
,
label_file
=
args
.
label_file
,
data_dir
=
data_dir
,
label_file
=
label_file
,
apply_distort
=
args
.
apply_distort
,
apply_expand
=
args
.
apply_expand
,
resize_h
=
args
.
resize_h
,
resize_w
=
args
.
resize_w
,
mean_value
=
[
args
.
mean_value_B
,
args
.
mean_value_G
,
args
.
mean_value_R
])
train
(
args
,
train_file_list
=
args
.
train_file_list
,
val_file_list
=
args
.
val_file_list
,
data_args
=
data_args
,
learning_rate
=
args
.
learning_rate
,
batch_size
=
args
.
batch_size
,
num_passes
=
args
.
num_passes
,
model_save_dir
=
args
.
model_save_dir
)
#method = parallel_do
method
=
parallel_exe
method
(
args
,
train_file_list
=
train_file_list
,
val_file_list
=
val_file_list
,
data_args
=
data_args
,
learning_rate
=
args
.
learning_rate
,
batch_size
=
args
.
batch_size
,
num_passes
=
args
.
num_passes
,
model_save_dir
=
model_save_dir
,
pretrained_model
=
args
.
pretrained_model
)
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