embed extended header inside .ptd flatbuffer section #7965
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7965
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Unrelated FailureAs of commit b73baeb with merge base cd51da4 ( NEW FAILURE - The following job has failed:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D68578075 |
Summary: Embed the header inside the flatbuffer. We do this for .pte and it lets us reuse a lot of flatbuffer tools natively. Differential Revision: D68578075
JacobSzwejbka
force-pushed
the
export-D68578075
branch
from
fc72e99 to
a052347
Compare
Summary: Embed the header inside the flatbuffer. We do this for .pte and it lets us reuse a lot of flatbuffer tools natively. Differential Revision: D68578075
JacobSzwejbka
force-pushed
the
export-D68578075
branch
from
a052347 to
f2987eb
Compare
|
This pull request was exported from Phabricator. Differential Revision: D68578075 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D68578075 |
lucylq
left a comment
lucylq
left a comment
There was a problem hiding this comment.
Lgtm,
nit to change extended_header to flat_tensor_header for flat_tensor?
dbort
left a comment
dbort
left a comment
There was a problem hiding this comment.
Looks good! Thanks for making this change.
A good double-check would be to try parsing a .ptd file using the flatc commandline tool, like
flatc --json --defaults-json --strict-json -o /tmp extension/flat_tensor/serialize/flat_tensor.fbs -- input.ptd
and see if it was able to create /tmp/input.json
Summary: Embed the header inside the flatbuffer. We do this for .pte and it lets us reuse a lot of flatbuffer tools natively. Reviewed By: lucylq Differential Revision: D68578075
JacobSzwejbka
force-pushed
the
export-D68578075
branch
from
f2987eb to
b73baeb
Compare
|
This pull request was exported from Phabricator. Differential Revision: D68578075 |
|
Ran the command Dave suggested on a local update to the train xor stack that makes it use .ptd and it worked (ill land that change later once the runtime can load .ptds |
Differential Revision: D68578075 Pull Request resolved: #7965
Differential Revision: D68578075 Pull Request resolved: pytorch#7965
4 participants
Summary: Embed the header inside the flatbuffer. We do this for .pte and it lets us reuse a lot of flatbuffer tools natively.
Differential Revision: D68578075