Fix GPU memory initialization #150
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Verified that this fix plus the fix from #149 lets me train DetectNet in DIGITS on a TitanX without running out of memory. |
lukeyeager
force-pushed
the
nvidia/fix-gpu-memory-initialization
branch
from
43f1234 to
d0f1ee9
Compare
|
Thanks for the review @drnikolaev. I've updated the PR to keep arena in scope as per your suggestion. |
lukeyeager
force-pushed
the
nvidia/fix-gpu-memory-initialization
branch
from
d0f1ee9 to
5eb5d46
Compare
|
Should the memory manager be in use when caffe is invoked via python bindings? |
|
An excellent question. |
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
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Outdated suggestions cannot be applied.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.
For
caffe train, we weren't actually initializing the GPUMemoryManager at all unless you explicitly set-gpuon the command line. But we should - caffe gets the list of GPUs in other ways if not set on the command line.