Skip to content

Conversation

NHZlX
Copy link
Contributor

@NHZlX NHZlX commented Sep 28, 2017

fix #4465

@NHZlX NHZlX requested a review from Xreki September 28, 2017 06:53
@NHZlX NHZlX requested a review from hedaoyuan October 25, 2017 09:40
initMain(argc, argv);
initPython(argc, argv);
string confFile = TrainerConfigHelper::getConfigNameFromPath(FLAGS_model_dir);

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

在这里加个对参数的检查吧:

if (FLAGS_model_dir.empty() || FLAGS_config_file.empty() || FLAGS_model_file.empty()) {
  LOG(INFO) << "Usage: ./paddle_merge_model --model_dir=pass-00000 --config_file=config.py --model_file=model.paddle";
  return;
}


message OptimizationConfig {
required int32 batch_size = 3;
optional int32 batch_size = 3 [ default = 1 ];
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

使用ModelConfig之后,这里的修改是不是不需要了。

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

我试了下,还是需要给batch_size设置默认值的。

Copy link
Contributor

@Xreki Xreki left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM


message OptimizationConfig {
required int32 batch_size = 3;
optional int32 batch_size = 3 [ default = 1 ];
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

我试了下,还是需要给batch_size设置默认值的。

@NHZlX NHZlX merged commit 4273b35 into PaddlePaddle:develop Oct 26, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

关于paddle_merge_model 一些疑问

2 participants