Torch-Pruning icon indicating copy to clipboard operation
Torch-Pruning copied to clipboard

[CVPR 2023] Towards Any Structural Pruning; LLMs / SAM / Diffusion / Transformers / YOLOv8 / CNNs

Results 196 Torch-Pruning issues
Sort by recently updated
recently updated
newest added

You have two dead links in your Wiki. In Quick Start -> 3. High-level Prunes -> Pruning with High-level Pruners -> "this tutorial" and "benchmarks/main.py" #### Pruning with High-level Pruners...

我的剪枝流程是 1. 先提前测试验证集精度。 2. 稀疏训练10个epoch。 (3.剪枝。 4. 验证剪枝后的精度。 5. finetune并验证精度。重复这三步) 我有以下几个问题: 1. 我这剪枝流程是否正确 2. 在finetune的时候,是否要加入pruner.regularize(self.model) 3. pruner中的参数如iterative_steps和ch_sparsity这些会不会影响稀疏训练,因为这里的iterative_steps=1,而我总的iterative_steps=16。这些参数会对稀疏训练有影响嘛。 期待大佬回答 :> ![image](https://github.com/VainF/Torch-Pruning/assets/44052395/ab9ca7ad-4690-45f4-8315-11dfb34794dd)

how can I prune destseg by using torching pruning tool? thanks

Hi, the discord invite link is invalid or has expired. Would you mind updating the link? Thanks!

Hi. I am trying to prune MViTv2 with random pruning with the following code ```python import torch import timm from timm.models.mvitv2 import MultiScaleAttention, MultiScaleVit as MViT import torch_pruning as tp...

bug

Hello author: Thanks for your excellent contribution, I've encountered a bug and don't know how to solve it .... When I use the official yolov8.pt to prune-post-train, the whole process...

the error is as following: albumentations: Blur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8)) val: Scanning /media/hitcrt/6a071232-a52f-4f53-89ca-fdde738abfd8/assignment10_19/data_original/4kp_data/labeled/rgb/南航.cache... 704 images, 1 Plotting labels to runs/pose/step_0_finetune11/labels.jpg... optimizer: AdamW(lr=0.000476,...

Thank you for your great work! I encountered the following problem when reproducing pruning. What happened? Traceback (most recent call last): File "prune_hf_wav2vec.py", line 68, in for g in pruner.step(interactive=True):...

hello, VainF,Could you provide an example for us to help us prune our own block by ourselves?

分组后剪枝这个过程中需要降低整个组的稀疏性,那这个稀疏性是加在了什么地方。比如分组了,但是我们一开始并不知道需要对哪些地方进行剪枝,那么这个组稀疏要怎么实现。