Optimize batch loading and metrics writing, replace PositionalSharding with NamedSharding#186
Conversation
…to tfrecord, batch iterator for tfrecord cached, namedsharding instead of positional sharding Signed-off-by: Kunjan <kunjanp@google.com>
Signed-off-by: Kunjan <kunjanp@google.com>
Signed-off-by: Kunjan <kunjanp@google.com>
Signed-off-by: Kunjan <kunjanp@google.com>
entrpn
previously approved these changes
Jun 17, 2025
Collaborator
entrpn
left a comment
entrpn
left a comment
There was a problem hiding this comment.
I'll approve it, but would really like to find the root cause as to why we need an executor to load next batch.
coolkp
force-pushed
the
sdxl-gpu
branch
5 times, most recently
from
bdded09 to
4b051fa
Compare
entrpn
previously approved these changes
Jun 18, 2025
…ocessed features Signed-off-by: Kunjan <kunjanp@google.com>
coolkp
force-pushed
the
sdxl-gpu
branch
5 times, most recently
from
6b6881d to
388ade1
Compare
entrpn
approved these changes
Jun 18, 2025
hx89
pushed a commit
to hx89/maxdiffusion
that referenced
this pull request
Jul 14, 2025
…g with NamedSharding (AI-Hypercomputer#186) * fix profiling * Use torch cpu, async write to tensorboard, script to convert latents to tfrecord, batch iterator for tfrecord cached, namedsharding instead of positional sharding Signed-off-by: Kunjan <kunjanp@google.com> * Replace positional sharding with named sharding Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Fallback to regular tfrecord iterator for datasets without all the processed features Signed-off-by: Kunjan <kunjanp@google.com> * README update --------- Signed-off-by: Kunjan <kunjanp@google.com>
hx89
pushed a commit
to hx89/maxdiffusion
that referenced
this pull request
Jul 15, 2025
…g with NamedSharding (AI-Hypercomputer#186) * fix profiling * Use torch cpu, async write to tensorboard, script to convert latents to tfrecord, batch iterator for tfrecord cached, namedsharding instead of positional sharding Signed-off-by: Kunjan <kunjanp@google.com> * Replace positional sharding with named sharding Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Fallback to regular tfrecord iterator for datasets without all the processed features Signed-off-by: Kunjan <kunjanp@google.com> * README update --------- Signed-off-by: Kunjan <kunjanp@google.com>
hx89
pushed a commit
to hx89/maxdiffusion
that referenced
this pull request
Jul 15, 2025
…g with NamedSharding (AI-Hypercomputer#186) * fix profiling * Use torch cpu, async write to tensorboard, script to convert latents to tfrecord, batch iterator for tfrecord cached, namedsharding instead of positional sharding Signed-off-by: Kunjan <kunjanp@google.com> * Replace positional sharding with named sharding Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Fallback to regular tfrecord iterator for datasets without all the processed features Signed-off-by: Kunjan <kunjanp@google.com> * README update --------- Signed-off-by: Kunjan <kunjanp@google.com>
hx89
pushed a commit
to hx89/maxdiffusion
that referenced
this pull request
Jul 15, 2025
…lSharding with NamedSharding (AI-Hypercomputer#186) * fix profiling * Use torch cpu, async write to tensorboard, script to convert latents to tfrecord, batch iterator for tfrecord cached, namedsharding instead of positional sharding Signed-off-by: Kunjan <kunjanp@google.com> * Replace positional sharding with named sharding Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Formatting Signed-off-by: Kunjan <kunjanp@google.com> * Fallback to regular tfrecord iterator for datasets without all the processed features Signed-off-by: Kunjan <kunjanp@google.com> * README update --------- Signed-off-by: Kunjan <kunjanp@google.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Comparison with main (Improvements in step time with using tf)
