-
Notifications
You must be signed in to change notification settings - Fork 37
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
MNASNet and Pretrained Weight #119
Conversation
train_data_loader = OFRecordDataLoader( | ||
ofrecord_root=args.ofrecord_path, | ||
mode="train", | ||
dataset_size=9469, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
dataset_size感觉可以作为一个传参,因为数据集可选择两个
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK
val_data_loader = OFRecordDataLoader( | ||
ofrecord_root=args.ofrecord_path, | ||
mode="val", | ||
dataset_size=3925, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
同上
### 1. Prepare Traning Data And Pretrain Models | ||
#### Download Ofrecord | ||
```bash | ||
wget https://oneflow-public.oss-cn-beijing.aliyuncs.com/datasets/imagenette_ofrecord.tar.gz | ||
tar zxf imagenette_ofrecord.tar.gz |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
个人认为还可以加一个imagenet的下载地址,因为你的脚本里面可以选择imagenette和imagenet两个数据集
train_data_loader = OFRecordDataLoader( | ||
ofrecord_root=args.ofrecord_path, | ||
mode="train", | ||
dataset_size=9469, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
可当作传参
val_data_loader = OFRecordDataLoader( | ||
ofrecord_root=args.ofrecord_path, | ||
mode="val", | ||
dataset_size=3925, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
可当作传参
本次PR的内容如下:
已完成,可以merge
验证结果:
以下验证结果是在输入和权重完全一致的情况下校验是否对齐
oneflow版本预训练权重下载链接:
mnasnet0_5: link
mnasnet1_0: link