Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/yuhao318/mwh

Last synced: 3 months ago
JSON representation

Host: GitHub
URL: https://github.com/yuhao318/mwh
Owner: yuhao318
License: bsd-3-clause
Created: 2021-01-09T07:38:51.000Z (almost 4 years ago)
Default Branch: main
Last Pushed: 2021-01-12T03:08:50.000Z (almost 4 years ago)
Last Synced: 2024-05-12T22:48:47.667Z (6 months ago)
Language: Python
Size: 22.5 KB
Stars: 40
Watchers: 3
Forks: 7
Open Issues: 0
Metadata Files:
- Readme: README.md
- License: LICENSE

Awesome Lists containing this project

Awesome-Mixup - [Code

README

        # Mixup Without Hesitation

This repo contains demo reimplementations of the CIFAR-100 training code in PyTorch based on the following paper:

Hao Yu, Huanyu Wang, Jianxin Wu. *Mixup Without Hesitation.* 

## How to use

```

git clone https://github.com/yuhao318/mwh.git

cd mwh

CUDA_VISIBLE_DEVICES=0 python easy_mwh.py --sess mwh_shufflenetv2_0.5 --alpha 0.5

```

## Results

The following table shows accuracy with $\alpha = 0.5$  and 100 epochs  in CIFAR-100:

| Model   | PreAct ResNet-18 | DenseNet-161 | Wide ResNet28-10|

| :------ | ---------------: | -----------: |---------------: |

| mixup   |           75.27% |           78.71% |78.86%|

| mwh     |           77.00% |          79.94% |79.94%|