[FIX] explicit stage num with uniform stage divided by flops #878
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The previous
UniformStageOption
uses "flops_if_cut_here
is the closest toaverage_flops_per_stage
" to divide a stage, however, this tends to divide a stage slightly smaller than the average. The shift accumulates and thus the last stage may not end at the last layer, but some layers before. This pr fixes the bug by cutting at "flops_till_now
is the closest toaverage_flops_per_stage * stage_idx
"