You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The paper states that workers can perform τ local updates, accumulate the gradients before using the LBGM algorithm for calculation. But in the code, it seems that workers update without the process of accumulating gradients every batch, which is inconsistent with the paper. May I ask if the existing code just considers τ= 1? If there is the latest code, please could you upload it? Thanks!
The text was updated successfully, but these errors were encountered:
The paper states that workers can perform τ local updates, accumulate the gradients before using the LBGM algorithm for calculation. But in the code, it seems that workers update without the process of accumulating gradients every batch, which is inconsistent with the paper. May I ask if the existing code just considers τ= 1? If there is the latest code, please could you upload it? Thanks!
The text was updated successfully, but these errors were encountered: