-
你好,我注意到脉冲神经元是接在卷积层之后,相邻的两层脉冲神经元之间有一层卷积和池化,那么两层脉冲神经元之间的连接关系是卷积和池化的共同结果吗?
-
**Issue type**
- [ ] Bug Report
- [ ] Feature Request
- [x] Help wanted
- [ ] Other
**SpikingJelly version**
`0.0.0.0.14`
**Description**
您好!
主要想求助一些STDP的使用问题:
1. STDPLearner中的sn参数是只…
-
**Issue type**
- [ ] Bug Report
- [ ] Feature Request
- [x] Help wanted
- [x] Other
**SpikingJelly version**
'latest'
**Description**
Hi all,
I'm trying to fine-tune a VGG19 pretr…
-
@fangwei123456
**Issue type**
- [x] Bug Report
- [ ] Feature Request
- [ ] Help wanted
- [ ] Other
**SpikingJelly version**
`0.0.0.0.14`
**Description**
使用 pytorch DDP 单机多卡训练时,无法使用…
-
# **voltagescaler 中scale爆炸,变成nan**
您好!我想咨询下,本代码使用的scale分别1.0/s和s,导致无论是在文件ann_to_snn.rst的例子中还是实际实验中,都会出现scale越来越大的情况,我想咨询下这个情况是正常的么?我使用resnet50的孪生网络进行转换,导致在最后的几层频繁出现scale爆炸变成nan的情况,这种情况是否在程序考虑范围内?
…
-
**Issue type**
- [ ] Bug Report
- [ ] Feature Request
- [x] Help wanted
- [ ] Other
**SpikingJelly version**
`0.0.0.0.14`
**Description**
在对已有的复杂网络进行ann2snn后进行测试时,发现随着测试时间步的推进,其准确率会先…
-
`BatchNorm` is also a stateful module, and has different outputs when using in step-by-step and layer-by-layer mode. So, we need to define a `MultiStepBatchNorm` module.
```python
import torch
im…
-
I found that the multi-step ResNet18 (identical to the exemplification in the doc) with the S2NN surrogate doesn't work when trained on a simple image classification (e.g. cifar 10). Specifically, aft…
-
问题描述:
在对MNIST的分类任务中,一个两层的网络(784x10),也就是lif_fc_mnist.py的案例,我们使用OutputMonitor记录输出,out_fr是将各时间步的网络输出相加,record是将OutputMonitor记录的数据相加
在每个时间步不使用clear_recorded_data()方法,out_fr与record不相等(左图);
在每个时间步使用clear…
-