Yang Nie

[email protected]

Beijing University of Posts and Telecommunications Beijing

Results 18 comments of


                                            Yang Nie

trafficstars

【飞桨论文复现挑战赛（第七期）】榜单

【队名】：拼手速【序号】：4 【论文】：[MicroNet: Improving Image Recognition with Extremely Low FLOPs](https://openaccess.thecvf.com/content/ICCV2021/html/Li_MicroNet_Improving_Image_Recognition_With_Extremely_Low_FLOPs_ICCV_2021_paper.html) 【状态】：报名【repo链接】：https://github.com/flytocc/MicroNet_paddle

【飞桨论文复现挑战赛（第七期）】榜单

【队名】：拼手速【序号】：3 【论文】：[TResNet: High Performance GPU-Dedicated Architecture](https://openaccess.thecvf.com/content/WACV2021/papers/Ridnik_TResNet_High_Performance_GPU-Dedicated_Architecture_WACV_2021_paper.pdf) 【状态】：报名【repo链接】：https://github.com/flytocc/TResNet_paddle

reproduction of Micronet-M0 4GPU

How about the stability of the largest model (m3)? @liyunsheng13

move the tipc config of ConvNeXt and PeleeNet from `config` to `configs`

solved

C#部署FastDeploy的问题

@zhinangubei 我也碰到了同样的问题，请问你解决了吗

run project on raspbeery pi

I haven't tried that. I don't know if it's okay. You can try: `python3 main.py`

[Bug] `accumulative_counts` does not apply to `parameter_schedulers`

To modify `ParamSchedulerHook`?

add MobileNeXt

## 复现精度 | 模型 | epochs | top1 acc (参考精度) | (复现精度) | 权重 | 训练日志 | |:---------------:|:-------:|:--------------------:|:----------:|:----------------------:|:---------:| | MobileNeXt-1.0 | 200 | 74.022 | 74.024 | best_model.pdparams | train.log...

add MobileNeXt

@cuicheng01 我把这个分支rebase到最新的develop上了并重新开了一个PR https://github.com/PaddlePaddle/PaddleClas/pull/2734

add MobileNeXt

merged by https://github.com/PaddlePaddle/PaddleClas/pull/2734

1
2
›