-
Notifications
You must be signed in to change notification settings - Fork 58
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
cuda_rasterizer-->backward.cu中的361行-->368行中的final_A, final_D,final_D2需要随着循环变化吗? #28
Comments
第367行是这一行:
|
Hi, here is the formulation for back-propagation of the depth-distortion loss: Now we specifically analyze the factorize the nested composition terms and we have got: |
Hi @hbb1 The above formulation is very helpful in understanding the implementation of the backward function. May I further ask about the formulation of the normal loss? |
Another quick question. When calculating the derivative, why not considering the case of d L_k / dw_m, (when m not equals to k)? |
Because we will instead compute something like dL / dw_{k-1} so that the algorithm can be efficiently run back-to-front. last_dL_dT = dL_dweight * alpha + (1 - alpha) * last_dL_dT; |
@hbb1 Thanks a lot for your prompt reply. I tried to understand how the backward is implemented these days but still a bit confused. I would be very appreciative if you could explain a bit more. Based on (17)
The following code calculates
My understanding is that I was wondering if I made any mistakes. If not, how to get the implementation above based on the formulations. Thanks a lot. |
Hi, I found I did not write it clearly. Now let's think of the gradients of
|
@hbb1 Really appreciate your prompt reply. I can understand most of the parts until the last sentence on Based on the code
Denoting May I know if I made any mistake? Thanks a lot. |
尤其367行的final_A和final_D要随循环的变化而变化吗?
The text was updated successfully, but these errors were encountered: