implement concatenation of sparse tensors #13577
Closed
Conversation
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
umanwizard
force-pushed
the
sparse_cat_and_stack
branch
from
807ba97 to
54ddd12
Compare
gchanan
reviewed
Nov 7, 2018
aten/src/ATen/native/TensorShape.cpp
Outdated
| std::vector<Tensor> values; | ||
| int64_t sparse_dim = tensors[0].sparse_dim(); | ||
| int64_t dense_dim = tensors[0].dense_dim(); | ||
| AT_CHECK(dim < sparse_dim, |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
| auto const &t = tensors[i]; | ||
| AT_CHECK(t.is_sparse(), | ||
| "Can't cat dense tensor at position ", i, " with sparse tensor(s)."); | ||
| AT_CHECK(t.sparse_dim() == sparse_dim && t.dense_dim() == dense_dim, |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
| return false; | ||
| } | ||
| for (int64_t i = 0; i < s1.size(); ++i) { | ||
| if (i != dim_except && s1[i] != s2[i]) { |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
aten/src/ATen/native/TensorShape.cpp
Outdated
| col += this_piece_size; | ||
| } | ||
| auto sizes_copy = sizes.vec(); | ||
| sizes_copy[dim] = cumulative_offset; |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Contributor
Author
|
@gchanan thanks for the review. I have updated the PR. |
gchanan
approved these changes
Nov 8, 2018
Contributor
gchanan
left a comment
gchanan
left a comment
There was a problem hiding this comment.
lgtm, just fix up a few nits please.
aten/src/ATen/native/TensorShape.cpp
Outdated
| // E.g.: catting [[1,2],[0,0]] and [[0,0],[3,4]] | ||
| // yields [[1,2,0,0],[0,0,3,4]] | ||
| AT_CHECK(wrapped < sparse_dim, | ||
| "Can't cat or stack tensors of sparse dim ", sparse_dim, "along non-sparse dimension ", dim); |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
aten/src/ATen/native/TensorShape.cpp
Outdated
| " along dimension ", dim); | ||
| AT_CHECK(t.sparse_dim() == sparse_dim && t.dense_dim() == dense_dim, | ||
| "Tensor at position ", i, " has dimension: sparse ", t.sparse_dim(), ", dense ", t.dense_dim(), | ||
| ". Can't cat with tensor of dimensions ", sparse_dim, ", ", dense_dim); |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
test/test_sparse.py
Outdated
| test_shapes( | ||
| [(3, 10, [2, 3, 4]), (3, 10, [2, 1, 4]), (3, 10, [2, 4, 4])], 1) | ||
| test_shapes([(3, 10, [2, 3, 4]), (3, 10, [2, 1, 4])], 0, | ||
| "can't be concatenated.*along dimension 0") |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@umanwizard has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
zdevito
pushed a commit
to zdevito/ATen
that referenced
this pull request
Nov 8, 2018
Summary: With this change applied, `torch.cat` works for sparse tensors. The algorithm is just to concatenate the values, and give the new values the proper indices (which will be the same as their old indices in every dimension except the catted dimension, and their old indices plus the sum of the size of every previous tensor in the catted dimension). This is my first time contributing to PyTorch so please feel free to tell me if this approach seems totally wrong. Coming next: `torch.stack` for sparse tensors. Pull Request resolved: pytorch/pytorch#13577 Differential Revision: D12980948 Pulled By: umanwizard fbshipit-source-id: 51ebdafee7fcd56d9762dcae9ebe5b4ab8e1dd6b
14 tasks
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 26, 2018
… dimensions (#13761) Summary: Follow-up to #13577 The idea is to take each values tensor, concatenate it with zeros before and after itself (along the dimension corresponding to the one we're catting the tensors along), to get a tensor corresponding to the values for that tensor in the result. Then we concatenate all of those together to get the final values tensor. (Hopefully, this will be more clear from the example in the comments). The indices are more straightforward: since we aren't concatenating along a sparse dimension, they don't change at all, so all we need to do are concatenate the indices from the different tensors together. Pull Request resolved: #13761 Differential Revision: D13160343 Pulled By: umanwizard fbshipit-source-id: 13d7adecd369e0eebdf5bce3d90a51029b66bd1d
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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.
With this change applied,
torch.catworks for sparse tensors.The algorithm is just to concatenate the values, and give the new values the proper indices (which will be the same as their old indices in every dimension except the catted dimension, and their old indices plus the sum of the size of every previous tensor in the catted dimension).
This is my first time contributing to PyTorch so please feel free to tell me if this approach seems totally wrong.
Coming next:
torch.stackfor sparse tensors.