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
weixin_41840029
PaddleOCR
提交
f9a2b26a
P
PaddleOCR
项目概览
weixin_41840029
/
PaddleOCR
与 Fork 源项目一致
Fork自
PaddlePaddle / PaddleOCR
通知
1
Star
1
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
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
PaddleOCR
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
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
未验证
提交
f9a2b26a
编写于
3月 28, 2022
作者:
littletomatodonkey
提交者:
GitHub
3月 28, 2022
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix quant logic (#5806)
* fix quant logic * fix undef * fix doc
上级
3d692957
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
9 addition
and
2 deletion
+9
-2
deploy/slim/quantization/quant.py
deploy/slim/quantization/quant.py
+9
-2
未找到文件。
deploy/slim/quantization/quant.py
浏览文件 @
f9a2b26a
...
@@ -118,6 +118,11 @@ def main(config, device, logger, vdl_writer):
...
@@ -118,6 +118,11 @@ def main(config, device, logger, vdl_writer):
config
[
'Architecture'
][
"Head"
][
'out_channels'
]
=
char_num
config
[
'Architecture'
][
"Head"
][
'out_channels'
]
=
char_num
model
=
build_model
(
config
[
'Architecture'
])
model
=
build_model
(
config
[
'Architecture'
])
pre_best_model_dict
=
dict
()
# load fp32 model to begin quantization
if
config
[
"Global"
][
"pretrained_model"
]
is
not
None
:
pre_best_model_dict
=
load_model
(
config
,
model
)
quanter
=
QAT
(
config
=
quant_config
,
act_preprocess
=
PACT
)
quanter
=
QAT
(
config
=
quant_config
,
act_preprocess
=
PACT
)
quanter
.
quantize
(
model
)
quanter
.
quantize
(
model
)
...
@@ -134,10 +139,12 @@ def main(config, device, logger, vdl_writer):
...
@@ -134,10 +139,12 @@ def main(config, device, logger, vdl_writer):
step_each_epoch
=
len
(
train_dataloader
),
step_each_epoch
=
len
(
train_dataloader
),
parameters
=
model
.
parameters
())
parameters
=
model
.
parameters
())
# resume PACT training process
if
config
[
"Global"
][
"checkpoints"
]
is
not
None
:
pre_best_model_dict
=
load_model
(
config
,
model
,
optimizer
)
# build metric
# build metric
eval_class
=
build_metric
(
config
[
'Metric'
])
eval_class
=
build_metric
(
config
[
'Metric'
])
# load pretrain model
pre_best_model_dict
=
load_model
(
config
,
model
,
optimizer
)
logger
.
info
(
'train dataloader has {} iters, valid dataloader has {} iters'
.
logger
.
info
(
'train dataloader has {} iters, valid dataloader has {} iters'
.
format
(
len
(
train_dataloader
),
len
(
valid_dataloader
)))
format
(
len
(
train_dataloader
),
len
(
valid_dataloader
)))
...
...
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