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
夏有凉风,冬有雪
Paddle
提交
62da4a1c
P
Paddle
项目概览
夏有凉风,冬有雪
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
1
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1
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
Paddle
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
1
Issue
1
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
体验新版 GitCode,发现更多精彩内容 >>
提交
62da4a1c
编写于
7月 11, 2017
作者:
Y
Yu Yang
提交者:
GitHub
7月 11, 2017
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #2802 from reyoung/feature/fix_python_slow
Fix slow parsing a recursive depends topology
上级
abff52ab
313e9f55
变更
5
隐藏空白更改
内联
并排
Showing
5 changed file
with
524 addition
and
1 deletion
+524
-1
python/paddle/trainer_config_helpers/networks.py
python/paddle/trainer_config_helpers/networks.py
+7
-0
python/paddle/trainer_config_helpers/tests/configs/file_list.sh
.../paddle/trainer_config_helpers/tests/configs/file_list.sh
+2
-1
python/paddle/trainer_config_helpers/tests/configs/protostr/test_cost_layers_with_weight.protostr
...ts/configs/protostr/test_cost_layers_with_weight.protostr
+2
-0
python/paddle/trainer_config_helpers/tests/configs/protostr/test_recursive_topology.protostr
...s/tests/configs/protostr/test_recursive_topology.protostr
+497
-0
python/paddle/trainer_config_helpers/tests/configs/test_recursive_topology.py
...r_config_helpers/tests/configs/test_recursive_topology.py
+16
-0
未找到文件。
python/paddle/trainer_config_helpers/networks.py
浏览文件 @
62da4a1c
...
...
@@ -1408,6 +1408,8 @@ def outputs(layers, *args):
:return:
"""
traveled
=
set
()
def
__dfs_travel__
(
layer
,
predicate
=
lambda
x
:
x
.
layer_type
==
LayerType
.
DATA
):
"""
...
...
@@ -1419,6 +1421,11 @@ def outputs(layers, *args):
:type layer: LayerOutput
:return:
"""
if
layer
in
traveled
:
return
[]
else
:
traveled
.
add
(
layer
)
assert
isinstance
(
layer
,
LayerOutput
),
"layer is %s"
%
(
layer
)
retv
=
[]
if
layer
.
parents
is
not
None
:
...
...
python/paddle/trainer_config_helpers/tests/configs/file_list.sh
浏览文件 @
62da4a1c
...
...
@@ -6,6 +6,7 @@ img_layers img_trans_layers util_layers simple_rnn_layers unused_layers test_cos
test_rnn_group shared_fc shared_lstm shared_gru test_cost_layers_with_weight
test_spp_layer test_bilinear_interp test_maxout test_bi_grumemory math_ops
test_seq_concat_reshape test_pad test_smooth_l1 test_multiplex_layer
test_prelu_layer test_row_conv test_detection_output_layer test_multibox_loss_layer
)
test_prelu_layer test_row_conv test_detection_output_layer test_multibox_loss_layer
test_recursive_topology
)
export
whole_configs
=(
test_split_datasource
)
python/paddle/trainer_config_helpers/tests/configs/protostr/test_cost_layers_with_weight.protostr
浏览文件 @
62da4a1c
...
...
@@ -131,6 +131,7 @@ input_layer_names: "weight"
input_layer_names: "multi_class_label"
output_layer_names: "__cost_0__"
output_layer_names: "__mse_cost_0__"
output_layer_names: "__nce_layer_0__"
evaluators {
name: "classification_error_evaluator"
type: "classification_error"
...
...
@@ -154,6 +155,7 @@ sub_models {
input_layer_names: "multi_class_label"
output_layer_names: "__cost_0__"
output_layer_names: "__mse_cost_0__"
output_layer_names: "__nce_layer_0__"
evaluator_names: "classification_error_evaluator"
is_recurrent_layer_group: false
}
...
...
python/paddle/trainer_config_helpers/tests/configs/protostr/test_recursive_topology.protostr
0 → 100644
浏览文件 @
62da4a1c
type: "nn"
layers {
name: "data"
type: "data"
size: 100
active_type: ""
}
layers {
name: "__addto_0__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "data"
}
inputs {
input_layer_name: "data"
}
}
layers {
name: "__addto_1__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_0__"
}
inputs {
input_layer_name: "__addto_0__"
}
}
layers {
name: "__addto_2__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_1__"
}
inputs {
input_layer_name: "__addto_1__"
}
}
layers {
name: "__addto_3__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_2__"
}
inputs {
input_layer_name: "__addto_2__"
}
}
layers {
name: "__addto_4__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_3__"
}
inputs {
input_layer_name: "__addto_3__"
}
}
layers {
name: "__addto_5__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_4__"
}
inputs {
input_layer_name: "__addto_4__"
}
}
layers {
name: "__addto_6__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_5__"
}
inputs {
input_layer_name: "__addto_5__"
}
}
layers {
name: "__addto_7__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_6__"
}
inputs {
input_layer_name: "__addto_6__"
}
}
layers {
name: "__addto_8__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_7__"
}
inputs {
input_layer_name: "__addto_7__"
}
}
layers {
name: "__addto_9__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_8__"
}
inputs {
input_layer_name: "__addto_8__"
}
}
layers {
name: "__addto_10__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_9__"
}
inputs {
input_layer_name: "__addto_9__"
}
}
layers {
name: "__addto_11__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_10__"
}
inputs {
input_layer_name: "__addto_10__"
}
}
layers {
name: "__addto_12__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_11__"
}
inputs {
input_layer_name: "__addto_11__"
}
}
layers {
name: "__addto_13__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_12__"
}
inputs {
input_layer_name: "__addto_12__"
}
}
layers {
name: "__addto_14__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_13__"
}
inputs {
input_layer_name: "__addto_13__"
}
}
layers {
name: "__addto_15__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_14__"
}
inputs {
input_layer_name: "__addto_14__"
}
}
layers {
name: "__addto_16__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_15__"
}
inputs {
input_layer_name: "__addto_15__"
}
}
layers {
name: "__addto_17__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_16__"
}
inputs {
input_layer_name: "__addto_16__"
}
}
layers {
name: "__addto_18__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_17__"
}
inputs {
input_layer_name: "__addto_17__"
}
}
layers {
name: "__addto_19__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_18__"
}
inputs {
input_layer_name: "__addto_18__"
}
}
layers {
name: "__addto_20__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_19__"
}
inputs {
input_layer_name: "__addto_19__"
}
}
layers {
name: "__addto_21__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_20__"
}
inputs {
input_layer_name: "__addto_20__"
}
}
layers {
name: "__addto_22__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_21__"
}
inputs {
input_layer_name: "__addto_21__"
}
}
layers {
name: "__addto_23__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_22__"
}
inputs {
input_layer_name: "__addto_22__"
}
}
layers {
name: "__addto_24__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_23__"
}
inputs {
input_layer_name: "__addto_23__"
}
}
layers {
name: "__addto_25__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_24__"
}
inputs {
input_layer_name: "__addto_24__"
}
}
layers {
name: "__addto_26__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_25__"
}
inputs {
input_layer_name: "__addto_25__"
}
}
layers {
name: "__addto_27__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_26__"
}
inputs {
input_layer_name: "__addto_26__"
}
}
layers {
name: "__addto_28__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_27__"
}
inputs {
input_layer_name: "__addto_27__"
}
}
layers {
name: "__addto_29__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_28__"
}
inputs {
input_layer_name: "__addto_28__"
}
}
layers {
name: "__addto_30__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_29__"
}
inputs {
input_layer_name: "__addto_29__"
}
}
layers {
name: "__addto_31__"
type: "addto"
size: 100
active_type: ""
inputs {
input_layer_name: "__addto_30__"
}
inputs {
input_layer_name: "__addto_30__"
}
}
layers {
name: "__fc_layer_0__"
type: "fc"
size: 32
active_type: "relu"
inputs {
input_layer_name: "__addto_31__"
input_parameter_name: "___fc_layer_0__.w0"
}
bias_parameter_name: "___fc_layer_0__.wbias"
}
layers {
name: "__fc_layer_1__"
type: "fc"
size: 10
active_type: "softmax"
inputs {
input_layer_name: "__fc_layer_0__"
input_parameter_name: "___fc_layer_1__.w0"
}
bias_parameter_name: "___fc_layer_1__.wbias"
}
parameters {
name: "___fc_layer_0__.w0"
size: 3200
initial_mean: 0.0
initial_std: 0.1
dims: 100
dims: 32
initial_strategy: 0
initial_smart: true
}
parameters {
name: "___fc_layer_0__.wbias"
size: 32
initial_mean: 0.0
initial_std: 0.0
dims: 1
dims: 32
initial_strategy: 0
initial_smart: false
}
parameters {
name: "___fc_layer_1__.w0"
size: 320
initial_mean: 0.0
initial_std: 0.176776695297
dims: 32
dims: 10
initial_strategy: 0
initial_smart: true
}
parameters {
name: "___fc_layer_1__.wbias"
size: 10
initial_mean: 0.0
initial_std: 0.0
dims: 1
dims: 10
initial_strategy: 0
initial_smart: false
}
input_layer_names: "data"
output_layer_names: "__fc_layer_1__"
sub_models {
name: "root"
layer_names: "data"
layer_names: "__addto_0__"
layer_names: "__addto_1__"
layer_names: "__addto_2__"
layer_names: "__addto_3__"
layer_names: "__addto_4__"
layer_names: "__addto_5__"
layer_names: "__addto_6__"
layer_names: "__addto_7__"
layer_names: "__addto_8__"
layer_names: "__addto_9__"
layer_names: "__addto_10__"
layer_names: "__addto_11__"
layer_names: "__addto_12__"
layer_names: "__addto_13__"
layer_names: "__addto_14__"
layer_names: "__addto_15__"
layer_names: "__addto_16__"
layer_names: "__addto_17__"
layer_names: "__addto_18__"
layer_names: "__addto_19__"
layer_names: "__addto_20__"
layer_names: "__addto_21__"
layer_names: "__addto_22__"
layer_names: "__addto_23__"
layer_names: "__addto_24__"
layer_names: "__addto_25__"
layer_names: "__addto_26__"
layer_names: "__addto_27__"
layer_names: "__addto_28__"
layer_names: "__addto_29__"
layer_names: "__addto_30__"
layer_names: "__addto_31__"
layer_names: "__fc_layer_0__"
layer_names: "__fc_layer_1__"
input_layer_names: "data"
output_layer_names: "__fc_layer_1__"
is_recurrent_layer_group: false
}
python/paddle/trainer_config_helpers/tests/configs/test_recursive_topology.py
0 → 100644
浏览文件 @
62da4a1c
from
paddle.trainer_config_helpers
import
*
settings
(
batch_size
=
1000
,
learning_rate
=
1e-5
)
din
=
data_layer
(
name
=
'data'
,
size
=
100
)
enc
=
din
for
i
in
range
(
32
):
enc
=
addto_layer
([
enc
,
enc
])
pred
=
fc_layer
(
input
=
fc_layer
(
input
=
enc
,
size
=
32
,
act
=
ReluActivation
()),
size
=
10
,
act
=
SoftmaxActivation
())
outputs
(
pred
)
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