#8518 Support for empty tuples #10027
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
…compatibility (pytorch#9403) Summary: Pull Request resolved: pytorch#9403 In BBoxTransform and GenerateProposal ops, clip_boxes makes sure the bbox fits within the images. For rotated boxes, this doesn't always make sense as there could be multiple ways to clip a rotated box within an image boundary. Moreover, clipping to a horizontal box means we leave out pixels of interest potentially. Therefore, we clip only boxes with angle almost equal to 0 (with a specified `angle_thresh` tolerance). Reviewed By: pjh5 Differential Revision: D8828588 fbshipit-source-id: 39c1eafdb5d39d383780faa0a47e76149145e50c
Summary: Enable fusion for IDEEP in optimizeForIdeep including Conv+ReLU, Conv+Sum, Conv+Sum+ReLU, Conv+BN Pull Request resolved: pytorch#9255 Reviewed By: bddppq Differential Revision: D8809030 Pulled By: yinghai fbshipit-source-id: af30bad3b96cb965bd26a4dfa810370faec4bb88
Summary: I noticed that `Sequential::clone()` does not work. This is because `Sequential` does not use `reset()` which is normally where modules have to initialize and register its submodules. Further, this is because of the way `Sequential` allows its modules to be passed in the constructor, which doesn't work with `reset()` (since it does "late" initialization). I've added some better error messages inside `Cloneable::clone()` which makes this kind of mistake clearer for other users, and tests for `Sequential::clone()`. I also had to give `AnyModule` a deep `clone()` method. ebetica ezyang Pull Request resolved: pytorch#9372 Differential Revision: D8865189 Pulled By: goldsborough fbshipit-source-id: b81586e0d3157cd3c4265b19ac8dd87c5d8dcf94
Summary: onnx/onnx@b2817a6 Pull Request resolved: pytorch#9476 Reviewed By: houseroad Differential Revision: D8868253 Pulled By: bddppq fbshipit-source-id: b1f14bab47f020f0bc0239da7e2bbf959a407d6a
Summary: This change makes README.md compatible with both Github and VSTS markdown engines. Images can be reduced if necessary Pull Request resolved: pytorch#9296 Differential Revision: D8874931 Pulled By: soumith fbshipit-source-id: 0c530c1e00b06fc891301644c92c33007060bf27
Summary: 1. Added tests 2. Added doc string 3. Remove view_as redundant definition from tensor.py Closes pytorch#9416 Pull Request resolved: pytorch#9452 Differential Revision: D8851794 Pulled By: ezyang fbshipit-source-id: 0aa0430dd0a174e1a5caddbc50a7e2c9eb7802bc
…ytorch#9475) Summary: test_cuda.py uses routine 'number' to prepare many testscases. number should return a floating point value for float-type tensor types, or integer otherwise. But number's test to classify the type is incorrect, so it always returns the integer value. (type(t).__name__ is always 'torch.tensortype' so never matches 'Double', 'Float', or 'Half'.) Update number to use the existing is_floating() helper to make the check. The change to number causes a few tests to fail for HalfTensor. Relax the tolerance for those in line with other HalfTensor testcases. The failing tests--for addcdiv and fill--were not previously relaxed for HalfTensor so are held to the over-strict 1e-5 default tolerance. Finally, update a couple other tests for HalfTensor type to use the existing is_half() helper. Pull Request resolved: pytorch#9475 Reviewed By: yf225 Differential Revision: D8872112 Pulled By: ezyang fbshipit-source-id: 016e3e15adb23f6606bd4c08218954c1396699db
Summary: Pull Request resolved: pytorch#9497 Fixes pytorch#7883 by using `rfft`. It's worth noting that this is BC breaking. And it's impossible to detect the change because the two signatures before and after this change supports a common subset of calling patterns, e.g., `stft(Tensor, int, int)`. (some other calling patterns will raise error). soumith and I plan to change the current `stft` interface because it is a bit messy and non-standard. rafaelvalle suggested us that `librosa` is a good reference API to align with. After discussing with soumith and ezyang , and given that `stft` is only out for 1 release, I decide to go with directly changing the signature. Also, my understanding is that most researchers in this field will welcome this change as `librosa` seems to be the golden-standard here. (it doesn't yet support all `pad_mode` but those will become available if added to `F.pad`.) Pull Request resolved: pytorch#9308 Reviewed By: ezyang Differential Revision: D8806148 Pulled By: SsnL fbshipit-source-id: f6e8777d0c34d4a4d7024e638dc9c63242e8bb58
Summary: This issue was fixed in 976f925 Fixes pytorch#5311. Signed-off-by: Edward Z. Yang <[email protected]> Pull Request resolved: pytorch#9498 Differential Revision: D8875605 Pulled By: ezyang fbshipit-source-id: 449ffe975d35c959f92874437ba9be37d4d3a1f2
Summary: It was only used to toggle refcounting, but we ALWAYS refcount tensors. Signed-off-by: Edward Z. Yang <[email protected]> Pull Request resolved: pytorch#9494 Differential Revision: D8875169 Pulled By: ezyang fbshipit-source-id: 3a8618fb288334e62942bbaf388f3c9e473e7524
Summary: Pull Request resolved: pytorch#9485 Reviewed By: houseroad Differential Revision: D8873733 Pulled By: bddppq fbshipit-source-id: 3a3cc351834cbbedce360760504ea16f5fa0ea06
Summary: Pull Request resolved: pytorch#9480 Ops like Reshape sometimes take a second input tensor of long with the new shape (can also be specified in arg). If this input tensor is passed in via external input (which ONNX does sometimes), LoadOp fails with an exception. Such ops anyway are executed by IDEEPFallbackOp, so this should be fine. Reviewed By: yinghai Differential Revision: D8872671 fbshipit-source-id: 659a02416c374e373ce041a7d65a174be828702d
pytorch#9482) Summary: …ors (CPU). This includes (mainly) CPU fixes; CUDA fixes are a little more involved because you can't use an empty grid. This also includes a fix for index_copy, which checked that self.size(dim) == src.size(0), which isn't correct (the same dimension should be compared). Finally, also includes a fix for CUDA flip (although it's not tested yet), to get the stride using multiplication rather than division to avoid divide-by-0. Pull Request resolved: pytorch#9482 Reviewed By: ezyang Differential Revision: D8873047 Pulled By: gchanan fbshipit-source-id: 86523afd3d50277834f654cd559dfbc7875cdffe
Summary: ebetica asked for a way to add parameters to `Optimizer`s after they are created. ebetica ezyang Pull Request resolved: pytorch#9472 Differential Revision: D8872176 Pulled By: goldsborough fbshipit-source-id: 39a4032c519a6d3b458dd3596361b04afea10365
Summary: This is enabled by the allocator patch; previously we could not deduplicate THStorage_free/THCStorage_free; now we can. Pull Request resolved: pytorch#9495 Reviewed By: SsnL Differential Revision: D8875497 Pulled By: ezyang fbshipit-source-id: 387198dff446eb9f84d2d6187066fae1d595dea7
Summary: Pull Request resolved: pytorch#9017 Closes pytorch#9017 Added "get_blob_size_bytes" to "pybind_state.cc" in Caffe2 to expose the size of blob in bytes. Reviewed By: kuttas Differential Revision: D8685696 fbshipit-source-id: 9a9d38f207c8c59ef534217181e8ce1514617628
Summary: Pull Request resolved: pytorch#9470 Reviewed By: pjh5 Differential Revision: D8826713 fbshipit-source-id: 47674af86b3a5ae0752056faf3b93f0d96e38fc2
Summary: It implements per-channel alpha_dropout. It also creates corresponding function classes and unifies the process of dropout and alpha_dropout. Pull Request resolved: pytorch#9073 Differential Revision: D8727008 Pulled By: ezyang fbshipit-source-id: 9d509f9c5db4e98f7b698cdfc4443505a4d2b331
Summary: If this is good, I could write some tests to ensure collision doesn't occur within a given range. Closes pytorch#7228 Pull Request resolved: pytorch#9246 Differential Revision: D8872608 Pulled By: ezyang fbshipit-source-id: 0ed29a73188f4167b42756f59a5c9a3d5cb37326
…nd (pytorch#9458) Summary: Pull Request resolved: pytorch#9458 The goal is to support count_include_pad in Caffe2 ONNX backend. This commit contains the first step - support 4-D tensor cases. AveragePool with count_include_pad can be expressed as PadImage + AveragePool. Reviewed By: houseroad Differential Revision: D8852180 fbshipit-source-id: 4db00e9771be7a000a2d92850dfd066d9c9c38bf
Summary: Pull Request resolved: pytorch#9126 Closes pytorch#9126 Allow concurrent read and writes in dispatcher table Reviewed By: smessmer Differential Revision: D8722560 fbshipit-source-id: e376bcd59f1b9f6b0e6fd3dd376a55561ea3c9c3
Summary: Stacked on pytorch#9495 Pull Request resolved: pytorch#9496 Differential Revision: D8875528 Pulled By: ezyang fbshipit-source-id: 6419d2ffb07aaf49c1462e7b64737019abbb7f61
Summary: - I ran into this couple days ago, and thought it might be useful to take note on that Pull Request resolved: pytorch#9504 Reviewed By: soumith Differential Revision: D8887396 Pulled By: weiyangfb fbshipit-source-id: d2061cf379ce140d6e43ef6c18241f7ce00dbab6
Summary: Signed-off-by: Edward Z. Yang <[email protected]> Pull Request resolved: pytorch#9516 Differential Revision: D8886493 Pulled By: ezyang fbshipit-source-id: fea974fd96c7d81126a129eb5b8b06eb1b028526
…ytorch#9520) Summary: Pull Request resolved: pytorch#9520 Add random data filler to predictor bench to support production nets Reviewed By: salexspb Differential Revision: D8712757 fbshipit-source-id: 2c732b2ba71ab210f9222adf94d08442ca71dc03
Summary: Pull Request resolved: pytorch#9523 Differential Revision: D8890124 Pulled By: soumith fbshipit-source-id: dea8d153fc352c36b219298c52f2c97caf9999f4
…h#9524) Summary: This command (suggested by albanD when I raised a related question in pytorch slack) is super useful to me. I have used it several times and it worked like a charm (without it, I have to delete entire pytorch folder and clone things again). So I guess it is nice to have in the CONTRIBUTING doc. Pull Request resolved: pytorch#9524 Differential Revision: D8890126 Pulled By: soumith fbshipit-source-id: c1798ff1ab2423627fcd8e0662a66c4e85cb2413
Summary: This PR contains the ROCm contributions of last week: * documentation of pyHIPIFY data format originating from pytorch#8812 reviewing comments by ezyang * removal of most patch files from the `amd_build` directory and integration into the code base * enabling of previously disabled_features that do compile now * improvement to the static_cast feature in pyHIPIFY (it will only apply static_cast to kernel arguments, not launch arguments) * addition of two workarounds to pyHIPIFY for ROCm/HIP shortcomings: a) `__forceinline__` does not imply `static`, hence change to `__inline__`, b) `std::[exp,log,pow]` math functions cannot be selected in device code, use `::[exp,log,pow]` instead. Both of these workarounds will be removed once the issues are fixed upstream. Neither of these issues have surfaced on the CI but were reproduced internally. Pull Request resolved: pytorch#9432 Differential Revision: D8887441 Pulled By: ezyang fbshipit-source-id: 71cf5c6b13772a66d10be369a45ebf06e4e268e1
Summary: A 0-dimensional tensor is now returned when squeezing a tensor with a single element. Pull Request resolved: pytorch#9529 Differential Revision: D8893103 Pulled By: soumith fbshipit-source-id: 658189ecfff283b2b7281feb16a397692d6dbd8f
Summary: fixes pytorch#9132 Pull Request resolved: pytorch#9487 Reviewed By: soumith Differential Revision: D8875529 Pulled By: SsnL fbshipit-source-id: d1b8aa825d202cfbdca27897da6a8bc1b714f856
Summary: Which test should I look at, bddppq? Pull Request resolved: pytorch#10022 Reviewed By: bddppq Differential Revision: D9068732 Pulled By: yinghai fbshipit-source-id: 241ef72c7fac0ed0b8c58ecdffbb5e24eb956217
Summary: Pull Request resolved: pytorch#9890 Minor cleanups for Graph.h to make it more consistent with our style guide Also fix opt/device.cc and binary_match_test.cc to not access subgraph.nodes_ which is now private Reviewed By: bwasti Differential Revision: D9017108 fbshipit-source-id: 9f5cba4a2cd2a452a955005f4704f6c120bbc1d5
Reviewed By: Maratyszcza Differential Revision: D9068091 fbshipit-source-id: 4aeac45f9732a86979a08488637bf0ba6cc79b34
Summary: Pull Request resolved: pytorch#10039 Reviewed By: houseroad Differential Revision: D9074261 Pulled By: bddppq fbshipit-source-id: 26df516633d5a4ec539a03a62cf9e7839e1e1964
Summary: I was dumb lol Pull Request resolved: pytorch#10047 Differential Revision: D9076023 Pulled By: bddppq fbshipit-source-id: 10587875d04ac2aed2e015846fc73ce9e4717a4f
Summary: ATenCore.h is a dummy header to just test that this is working at all. Pull Request resolved: pytorch#10019 Reviewed By: smessmer Differential Revision: D9067262 Pulled By: ezyang fbshipit-source-id: 58bab9c0aa83b56335e36b719b9b6505400d8dee
Summary: We missed the upsample symbolic when bumping up the opset to 7. Pull Request resolved: pytorch#10001 Reviewed By: bddppq Differential Revision: D9067212 Pulled By: houseroad fbshipit-source-id: 3e285d2800a32cb04fa82f8e7f261bdd010a8883
Summary: Pull Request resolved: pytorch#10064 Differential Revision: D9082082 Pulled By: gchanan fbshipit-source-id: ae49470f5b4c89b13beb55fd825de1ba05b6a4fa
…es (pytorch#9948) Summary: zdevito Pull Request resolved: pytorch#9948 Reviewed By: ezyang Differential Revision: D9033666 Pulled By: apaszke fbshipit-source-id: 02d75e391ed6dee62500842df50f0b6ee5e38846
jramseyer
requested review from
Yangqing,
anderspapitto,
bddppq,
dzhulgakov,
ebetica,
goldsborough,
houseroad,
jamesr66a and
smessmer
as code owners
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
jramseyer has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
jramseyer has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
Contributor
|
You may want to set |
Contributor
Author
|
Thanks! I will do that! |
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
jramseyer is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
goodlux
pushed a commit
to goodlux/pytorch
that referenced
this pull request
Aug 15, 2018
Summary: Fixing pytorch#8518 Sorry for the pile of commits; I forgot to rebase. Pull Request resolved: pytorch#10027 Reviewed By: ezyang Differential Revision: D9070028 Pulled By: jramseyer fbshipit-source-id: 49729c9755ab8a586711e9f6d6a574f3035a7e75
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 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.
Fixing #8518
Sorry for the pile of commits; I forgot to rebase.