Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
root 12d91d88a5 | 4 months ago | |
---|---|---|
deepsparkhub | 4 months ago | |
README.md | 4 months ago |
cd /usr
git clone -b openi-task12 https://gitee.com/deep-spark/deepsparkhub
在/usr目录下进行git操作!!/code目录下无法git
cd /usr/deepsparkhub/cv/classification/resnet50/pytorch/
vim ./scripts/train_resnet50_imagenet.sh
# 拷贝到/usr目录
cp /dataset/imagenet.zip /usr
# 解压
unzip /usr/imagenet.zip
bash scripts/fp32_1card.sh --data-path /usr/imagenet
修改train.py
执行训练
python3 train.py --data-path /usr/imagenet --model googlenet --batch-size 512
pip3 install -r requirements.txt
vim train_vgg16_amp_dist.sh
# 修改数据集路径
DATA_PATH=${DATA_PATH:-"/usr/imagenet"}
# 修改 epochs
--epochs 5
bash train_vgg16_amp_dist.sh
vim train.py
python3 train.py --data-path /usr/imagenet --model lenet
vim train.py
cd start_scripts
bash train_alexnet_torch.sh --data-path /usr/imagenet
pip install timm==0.4.12
pip install yacs
vim config.py
python3 -m torch.distributed.launch --nproc_per_node 1 --master_port 12345 main.py \
--cfg configs/swin/swin_tiny_patch4_window7_224.yaml --data-path /usr/imagenet --batch-size 128
vim train_mobilenet_v3_large_amp_dist.sh
# 修改数据集路径
DATA_PATH=${DATA_PATH:-"/usr/imagenet"}
# 修改 epoch
--epochs 5
bash train_mobilenet_v3_large_amp_dist.sh
官方文档有误(数据集路径,文件名)
vim train_inception_v3_amp_dist.sh
# 修改数据集路径
DATA_PATH=${DATA_PATH:-"/usr/imagenet"}
# 修改 epoch
--epochs 5
bash train_inception_v3_amp_dist.sh
官方文档有误(数据集路径)
修改train.py
执行训练
python3 train.py --data-path /usr/imagenet --model densenet201 --batch-size 128
python3 train.py --data-path /usr/imagenet --model squeezenet1_0 --lr 0.001 --epochs 5
No Description
Python Unity3D Asset Text Jupyter Notebook C++ other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》