qujialing c3003882de | 1 year ago | |
---|---|---|
.. | ||
README.md | 1 year ago | |
README_CN.md | 1 year ago |
RepVGG, a vgg-style architecture that outperforms many complex models
Its main highlights are:
The model has a normal (a.k.a. feedforward) structure like vgg, without any other branches, each layer takes the
output of its only previous layer as input, and feeds the output to its only next layer.
The body of the model uses only 3 × 3 conv and ReLU.
The specific architecture (including specific depth and layer width) is instantiated without automatic search, manual
refinement, compound scaling, and other complicated designs.
Pynative | Pynative | Graph | Graph | ||||||
---|---|---|---|---|---|---|---|---|---|
Model | Top-1 (%) | Top-5 (%) | train (s/epoch) | Infer (ms) | train(s/epoch) | Infer (ms) | Download | Config | |
GPU | RepVGG_A0 | 71.98 | 90.36 | model | config | ||||
Ascend | RepVGG_A0 | 71.87 | 90.43 |
The yaml config files that yield competitive results on ImageNet for different models are listed in
the configs
folder. To trigger training using preset yaml config.
comming soon
Here is the example for finetuning a pretrained RepVGG_A0 on CIFAR10 dataset using Adam optimizer.
python train.py --model=RepVGG_A0 --pretrained --opt=adam --lr=0.001 ataset=cifar10 --num_classes=10 --dataset_download
Detailed adjustable parameters and their default value can be seen in config.py
To validate the model, you can use validate.py
. Here is an example to verify the accuracy of pretrained weights.
python validate.py --model=RepVGG_A0 --dataset=imagenet --val_split=val --pretrained
To validate the model, you can use validate.py
. Here is an example to verify the accuracy of your training.
python validate.py --model=RepVGG_A0 --dataset=imagenet --val_split=val --ckpt_path='./ckpt/RepVGG_A0-best.ckpt'
No Description
Jupyter Notebook Python Markdown Text 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》