-
Hi，我也曾经是distiller的深度用户。
看你在这份代码增加了FPGM和HRank，想问一下这两个通道排序算法是否有效。
我个人也曾经尝试过BN-gamma[1] 、Taylor[2] 、L1[3]。
在我的大量实验下，排序效果：Taylor = L1 > BN-gamma。
想问一下，你有没有试过把FPGM和HRank和上述方法对比过呢？
[1] Learning Effic…
-
你好，看到你的readme介绍，你实现了规整剪枝、正常剪枝和分组卷积结构剪枝，我理解规整和正常剪枝是实现了slimming论文里的剪枝方法是吗，那分组卷积剪枝具体实现的是Rethinking the Value of Network Pruning论文里提到的哪个方法呢？
-
Hi Jiahui,
Thanks for the great work. I'm trying to reproduce AutoSlim for CIFAR-10 (Table 2).
Could you please provide a detailed hyperparameter you used for it?
I'm able to train the baseline…
-
hi everyone, i'm trying to evaluate the mithocondrial DNA of my sample (fusarium sp.);
i performed the assembly with flye tool on galaxy.eu and use the fastg file produced by the tool.
i always get …
-
你好，我基于mobilenet，利用slimming来进行剪枝，下面是运算量和参数数量的变化
flops:313.464M->310.625M, params: 2.226M->2.206M。
实验过程中的参数设置是和您给的一样的，但是剪枝后的参数数量基本变化不大，和图表展示的有很大的差异。
您可以在空闲的时间回复一下吗？我应该怎么达到图表中剪枝的效果呢？谢谢
-
请问，有人试过训练后精度如何?
-
If you just want my mod, heres the link. Its not opitimized though.
https://drive.google.com/file/d/1JcSnRI2yBknGpTJlUjvs1Ukd-Qu-pteL/view?usp=drivesdk
Anyways, this is the second iteration of my po…
-
Hi, the network, especially stackhourglass, is too fat. It cannot be used for large images input with 11G gpu memory (RTX 2080 TI). If I want to make the network more lightweight, which parts should I…
-
## 📝 Provided detailed reproduction steps
1. Setup the editor as inline with a couple of plugins in groups
2. Resize your browser window so that the floating toolbar starts hidding plugins into the …
-
**Title:** [Release] 0.9
**Description:**
0.9 is a release for the 1.7.8. This includes a number of features. These features are listed in the acceptance criteria.
**Acceptance Criteria:**
# _…