Torch-Pruning

Torch-Pruning copied to clipboard

> 我在执行过程中遇到代码错误，指出张量不在同一设备上。@Hyunseok-Kim0你能给我解释一下吗，谢谢

3

> 我在执行过程中遇到代码错误，指出张量不在同一设备上。@Hyunseok-Kim0你能给我解释一下吗，谢谢 That's the problem. How did you solve it? _Originally posted by @aicmaodyu in https://github.com/VainF/Torch-Pruning/issues/147#issuecomment-1727343947_ 解决了吗兄弟？我也是张量不在同一个设备上

Yuanshan627

fine-tuning stage after yolov8-seg prune using torch-pruning, when use torch.load to load save model, the error occured 

luoshiyong

@JonnyKong Hello!When I run prune_hf_swin.py, I met a problem:"Could not find image processor class in the image processor config or the model config. Loading based on pattern matching with the...

IASAnthony

Question: Why is Eigendamage and FPGM not docummented nor exposed in importance.py

1

Hi @VainF , first of all, thanks for the great work! Looking at `importance.py` I noticed that there are 2 importances which are not documented in the docs, nor exposed,...

Hrayo712

  测试环境： pytorch == 1.13 or pytorch==1.8 ultralytics==6a2fddf 运行脚本： 示例中的yolov8_example 直接异常原因： x类型应当为list而不是一个tensor 应当如何修改

MapleAura

mobilefacenet剪枝效果差

8

https://github.com/deepinsight/insightface/blob/master/recognition/arcface_torch/backbones/mobilefacenet.py 模型定义 https://github.com/deepinsight/insightface/blob/master/recognition/arcface_torch/backbones/mobilefacenet.py 使用group_norm剪枝方式，参数如下： model_name = 'mobilefacev2' global_pruning = 'True' reg = 1e-4 max_ch_sparsity = 1.0 target_flops = 1.0 prune_pm = 'True' method = "group_norm" 使用insightface训练框架，训练20轮，效果比原来差很多

eeric

对resnet18的剪枝流程

2

您好，对resnet18的整个预训练，稀疏化，剪枝，在训练的流程代码是在哪里啊？我想看看。

jazz9726

我使用代码中的剪枝方法对sknet进行剪枝，程序运行到pruner.step()的时候运行很长时间都不出结果，好像是卡在这里了。请问这是什么情况？对预训练的ResNet50进行剪枝时，很快就完成了。

2

yang-ming-uc

pengge990330

tks for very creative idea of torch pruning, But my model is exported from pretrained tensorflow model, it's hard to fine tune or re-trained again. So can you give me...

trangtv57