biclab / attention-snn Goto Github PK

Hello author,
In your paper, the ImageNet dataset achieves an impressive accuracy of 75.92% with T=1 and ResNet-104. However, I am only able to achieve 70.56% accuracy using parameters consistent with your supplementary materials. Furthermore, when examining the events file you provided in TensorBoard, I noticed that the test set accuracy is 74.14% while the training set accuracy is only 64.25%. This discrepancy shows that the test set accuracy is 10% higher than the training set accuracy. I look forward to your response.

请问下这篇文章的附录可以从哪里获得？是否可以分享一下？

The value of 'dt' as 15 and 'T 'as 60 at TABLE I for the data set DVS128 Gesture, and found the testing accuracy 89.9305% in the epoch 138, where your work in the paper says it has an efficiency of 96.53%. So any modification in the code is there that needs to be done to improve the testing accuracy to the claimed value.

hello， How to change your own image dataset？

在gait数据集上，代码细节中需要加载train_data.npy和train_target.npy，而原始数据集中只有txt文件，请问这一部分是怎么进行处理划分的呢？

您好，DVS128 Gesture的一条数据平均时间是6s，但论文中的 dt*T = 15 * 60 = 900ms，是不是可以认为您只使用了一条数据的前0.9s的信息，这样做不会丢失掉信息吗？

Traceback (most recent call last):
File "/content/Attention-SNN/MA_SNN/DVSGestures/Att_SNN_CNN.py", line 45, in
main()
File "/content/Attention-SNN/MA_SNN/DVSGestures/Att_SNN_CNN.py", line 41, in main
Att_SNN.main()
File "/content/Attention-SNN/MA_SNN/DVSGestures/CNN/Att_SNN.py", line 32, in main
create_data(config=config)
File "/content/Attention-SNN/MA_SNN/DVSGestures/DVS_Gesture_utils/dataset.py", line 8, in create_data
config.train_dataset = create_datasets(
File "/content/Attention-SNN/MA_SNN/DVSGestures/DVS_gesture_data_process/DVS_Gesture_dataloders.py", line 246, in create_datasets
train_d = DVSGestureDataset(root,
File "/content/Attention-SNN/MA_SNN/DVSGestures/DVS_gesture_data_process/DVS_Gesture_dataloders.py", line 57, in init
self.n = len(self.files_train)
AttributeError: 'DVSGestureDataset' object has no attribute 'files_train'