[AIR - Datasets] Encode number of dimensions in variable-shaped tensor extension type. #29281
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
8 tasks
clarkzinzow
force-pushed
the
datasets/fix/ragged-tensor-ndim
branch
from
dc64bb1
to
d97c6ad
Compare
clarkzinzow
requested review from
ericl,
scv119,
jjyao,
jianoaix and
c21
as code owners
clarkzinzow
changed the title
bveeramani
reviewed
Oct 20, 2022
clarkzinzow
force-pushed
the
datasets/fix/ragged-tensor-ndim
branch
from
dc53d80
to
14dc07e
Compare
clarkzinzow
force-pushed
the
datasets/fix/ragged-tensor-ndim
branch
from
c7d8566
to
a85626a
Compare
7 tasks
jianoaix
reviewed
Oct 24, 2022
| if ndim is not None and a.ndim != ndim: | ||
| raise ValueError( | ||
| "ArrowVariableShapedTensorArray only supports tensor elements that " | ||
| "all have the same number of dimensinos, but got tensor elements " |
There was a problem hiding this comment.
nit: "dimensinos" seems not fixed yet
There was a problem hiding this comment.
Oh weird, I thought I fixed that, good catch!
clarkzinzow
commented
Oct 24, 2022
Signed-off-by: Clark Zinzow <[email protected]>
WeichenXu123
pushed a commit
to WeichenXu123/ray
that referenced
this pull request
Dec 19, 2022
…r extension type. (ray-project#29281) Knowing the number of dimensions in a variable-shaped tensor column is useful for e.g. inferring a ragged tensor spec when constructing a tf.data Dataset; by encoding this ndim data in the extension type, we can do this type inference base on Dataset metadata, which is required. Note that this change will disallow variable-shaped tensor columns containing tensor elements that have a variable number of dimensions. This isn't supported by TensorFlow and Torch ragged tensors, so sacrificing this feature seems tenable. Signed-off-by: Weichen Xu <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in 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.
Knowing the number of dimensions in a variable-shaped tensor column is useful for e.g. inferring a ragged tensor spec when constructing a
tf.dataDataset; by encoding thisndimdata in the extension type, we can do this type inference based on Dataset metadata, which is required.Note that this change will disallow variable-shaped tensor columns containing tensor elements that have a variable number of dimensions. This isn't supported by TensorFlow and Torch ragged tensors, so sacrificing this feature seems tenable.
Related issue number
Closes #29135
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.