ConvNext_for_PyTorch

This implements training ConvNext of on the ImageNet dataset, mainly modified from https://github.com/facebookresearch/ConvNeXt.git

ConvNext_for_PyTorch Detail

As of the current date, Ascend-Pytorch is still inefficient for contiguous operations.

Requirements

• pip install -r requirements.txt
• pip install torch==1.8.1+ascend.rc2.20220505;torchvision==0.9.1;torch-npu 1.8.1rc2.post20220505;
• Download the ImageNet dataset from http://www.image-net.org/
  • Then, and move validation images to labeled subfolders, using the following shell script

timm

将timm_need目录下的文件替换到timm的安装目录下

cd ../ConvNeXt
/bin/cp -f timm_need/mixup.py ../timm/data/mixup.py
/bin/cp -f timm_need/model_ema.py ../timm/utils/model_ema.py

软件包

该模型为不随版本演进模型（随版本演进模型范围可在此处查看），未在最新昇腾配套软件中适配验证，您可以：

1. 根据下面提供PyTorch版本在软件版本配套表中选择匹配的CANN等软件下载使用。
2. 查看软件版本配套表后确认对该模型有新版本PyTorch和CANN中的适配需求，请在modelzoo/issues中提出您的需求。自行适配不保证精度和性能达标。

当前模型支持的历史版本软件如下所示。

• 910版本
• CANN toolkit_5.1.RC1
• torch 1.8.1+ascend.rc2.20220505
• 固件驱动 22.0.0

Training

To train a model, run main.py with the desired model architecture and the path to the ImageNet dataset:

# training 1p accuracy
bash ./test/train_full_1p.sh --data_path=real_data_path

# training 1p performance
bash ./test/train_performance_1p.sh --data_path=real_data_path

# training 8p accuracy
bash ./test/train_full_8p.sh --data_path=real_data_path

# training 8p performance
bash ./test/train_performance_8p.sh --data_path=real_data_path

#  eval 
bash test/train_eval_8p.sh --data_path=real_data_path

# finetuning
bash test/train_finetune_1p.sh --data_path=real_data_path

ConvNext_for_PyTorch training result

Acc@1	FPS	Npu_nums	Epochs	AMP_Type
-	115.10	1	300	O1
82.049	259.85	8	300	O1

公网地址说明

代码涉及公网地址参考 public_address_statement.md