Skip to content

M
mindspore

magicwindyyd / mindspore
与 Fork 源项目一致

Fork自 MindSpore / mindspore

通知 1
Star 1
Fork 0
M
mindspore
提交 2a2c944d 编写于 作者: W wangmin
浏览文件
操作

fix thor optimizer readme for resnet50

上级 86beb6e9
隐藏空白更改
内联 并排
Showing with 19 addition and 21 deletion
model_zoo/official/cv/resnet_thor/README.md
浏览文件 @ 2a2c944d
...@@ -79,27 +79,25 @@ sh run_eval_gpu.sh [DATASET_PATH] [CHECKPOINT_PATH] ...@@ -79,27 +79,25 @@ sh run_eval_gpu.sh [DATASET_PATH] [CHECKPOINT_PATH]
### Script Code Structure ### Script Code Structure
```shell ```shell
├── model_zoo └── resnet_thor
├──README.md # descriptions about all the models ├── README.md # descriptions about resnet_thor
├── resnet_thor ├── scripts
├── README.md # descriptions about resnet_thor │ ├── run_distribute_train.sh # launch distributed training for Ascend
├── scripts │ └── run_eval.sh # launch inference for Ascend
│ ├── run_distribute_train.sh # launch distributed training for Ascend │ ├── run_distribute_train_gpu.sh # launch distributed training for GPU
│ └── run_eval.sh # launch infering for Ascend │ └── run_eval_gpu.sh # launch inference for GPU
│ ├── run_distribute_train_gpu.sh # launch distributed training for GPU ├──src
│ └── run_eval_gpu.sh # launch infering for GPU │ ├── crossentropy.py # CrossEntropy loss function
├──src │ ├── config.py # parameter configuration
│ ├── crossentropy.py # CrossEntropy loss function │ ├── dataset_helper.py # dataset help for minddata dataset
│ ├── config.py # parameter configuration │ ├── grad_reducer_thor.py # grad reducer for thor
│ ├── dataset_helper.py # dataset help for minddata dataset │ ├── model_thor.py # model for train
│ ├── grad_reducer_thor.py # grad reducer for thor │ ├── resnet_thor.py # resnet50_thor backone
│ ├── model_thor.py # model for train │ ├── thor.py # thor optimizer
│ ├── resnet_thor.py # resnet50_thor backone │ ├── thor_layer.py # thor layer
│ ├── thor.py # thor optimizer │ └── dataset.py # data preprocessing
│ ├── thor_layer.py # thor layer ├── eval.py # infer script
│ └── dataset.py # data preprocessing └── train.py # train script
├── eval.py # infer script
└── train.py # train script
``` ```
### Script Parameters ### Script Parameters
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或