cdist: pairwise distances between two sets of tensors with batch mode #20934
Conversation
pytorchbot
added
module: cpu
aten/src/ATen/native/Distance.cpp
Outdated
| tensor2_expand_size.insert(tensor2_expand_size.end(), {r2, c2}); | ||
|
|
||
| int expand_batch_product = std::accumulate(expand_batch_portion.begin(), expand_batch_portion.end(), 1, std::multiplies<int64_t>()); | ||
| std::vector<int64_t> tensor1_view({expand_batch_product}); |
There was a problem hiding this comment.
Can replace these three lines with std::vector<int64_t> tensor1_view {expand_batch_product, r1, c1};
aten/src/ATen/native/Distance.cpp
Outdated
| auto dim1 = x1.dim(); | ||
| auto dim2 = x2.dim(); | ||
|
|
||
| IntArrayRef batch_tensor1(x1.sizes().data(), std::max<int64_t>(dim1 - 2, 0)); |
There was a problem hiding this comment.
Can you add some comments to this bunch of code to explain what you are doing
There was a problem hiding this comment.
You are checking above that the dims are >= 2, so why do we need the std::max calls?
| Tensor tensor1_expanded = x1.expand(tensor1_expand_size).contiguous().view(tensor1_view); | ||
| Tensor tensor2_expanded = x2.expand(tensor2_expand_size).contiguous().view(tensor2_view); | ||
|
|
||
| std::vector<int64_t> output_shape(expand_batch_portion); |
There was a problem hiding this comment.
Why do sometimes you use foo.insert(foo.end(), {bar, quux}); and sometimes you use foo.push_back twice? Should be consistent.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ifedan has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ifedan has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ifedan has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ifedan has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
… (#20934) Summary: Batch implementation for cdist function Pull Request resolved: pytorch/pytorch#20934 Differential Revision: D15609458 Pulled By: ifedan fbshipit-source-id: 31c12e120d168baec6a6af913f599838a44034d7
5 participants
Batch implementation for cdist function