camellia120

Results 11 comments of camellia120

我在复现作者的代码出现的问题

I am very glad to hear from you.Thank you very much!

Still can not support batch size more than 1?

I am very glad to hear from you.Thank you very much!

Queries about the test output

I am very glad to hear from you.Thank you very much!

question about testing

I am very glad to hear from you.Thank you very much!

您好，我最近在复现您的代码，但是遇到一些问题，可以和您交流一下吗

I am very glad to hear from you.Thank you very much!

The expanded size of the tensor (313) must match the existing size (256) at non-singleton dimension 2.

I am very glad to hear from you.Thank you very much!

RuntimeError: Output 0 of ScatterBackward is a view and is being modified inplace.

I am very glad to hear from you.Thank you very much!

why do you say #the batchsize Need to be set to 1 in training process ?

@KumapowerLIU In my experiments, when setting the batch size to 16, the GPU memory increases to 84% (9203MiB) in RTX 2080, but it seems that the GPU utilization is still...

About the preprocess of the dataset alignment

Hi @aiff22, Thx for your reply. The mismatch caused by optical systems and camera sensors is almost random: it can be expected when it has enough data in training. Beacuse...

batch为1时大概需要训练多久呢？

I am very glad to hear from you.Thank you very much!