iantangc / contrastivelearninghar Goto Github PK

Hi,
Thank you so much for sharing the excellent work! I'm now also trying to see the effectiveness of SimCLR on HAR data, and want to discuss it with you.

Since we can output KL divergence of t-sne result to see the network feature space, I try to output the t-sne results at each pretraining epoch (the SimCLR head) using your plot function. However, the feature space doesn't change a lot.

If I understood correctly, the KL divergence is supposed to be smaller after many epochs, because the SimCLR is supposed to learn good features for different activities. However, as the SimCLR loss got smaller, the latent feature didn't get better. Did you have some idea about this problem?

Thank you very much for your help!
Best
Cassie

I've observed a bug in this function - if n_windows = 1 due to a large window size, then np.squeeze will make user_dataset_windowed[user_id][0] a number with dimension 0 rather than a numpy array, and that will break the preprocessing script.

Hello, I noticed that in your paper you mentioned the reproduction of multi-task self-supervised Learning for Human Activity Detection.
Could you share the code of this part?

作者你好，请问一下你的对比学习模型的初始参数是在哪个数据集上训练的。如 SimCLR中的预训练参数是使用ResNet在ImageNet上训练的。