Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17619
--filler hive --iter -1 will let debugger exhaust all batches from a hive partition before exiting.
add README that summarizes command line options and usage.
Reviewed By: yinghai
Differential Revision: D14220166
fbshipit-source-id: daa23b7e8a9184481c6d7b67acf1599e5c99d74a
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17481
Usually, feature macros are either defined or undefined and checked accordingly.
C10_MOBILE was a weird special case that was always defined but either defined to 1 or to 0.
This caused a lot of confusion for me when trying to disable something from mobile build and it also disabled it
from the server build (because I was using ifdef). Also, I found a place in the existing code base that made
that wrong assumption and used the macro wrongly, see https://fburl.com/y4icohts
Reviewed By: dzhulgakov
Differential Revision: D14214825
fbshipit-source-id: f3a155b6d43d334e8839e2b2e3c40ed2c773eab6
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/16597
This diff fixes some bugs in shape inference for `SparseLengthsSumFused8BitRowwise`. And added input shape inference for `Concat` when `add_axis=1`.
Reviewed By: bertmaher
Differential Revision: D13892452
fbshipit-source-id: 6cd95697a6fabe6d78a5ce3cb749a3a1e51c68e7
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/16620
LogfiledbNetLoader loads all external input blobs into a workspace instance, we pack a shared pointer to this loaded workspace into the SingleLoadedNetSupplier.
SingleLoadedNetSupplier will pass this workspace to BlackBoxPredictor to be executed. (D13891759 is a WIP of how it all comes together)
Reviewed By: pjh5
Differential Revision: D13901467
fbshipit-source-id: 20589f898922f5f1aec50be131dad17a8c38e9b2
Summary:
Based on offline discussion it should be less surprising to the users of existing code. Thus caffe2::Tensor is now a move-only class (as it used to be), explicit calls to UnsafeSharedInstance() are necessary to get shared_ptr behavior.
This change also identified a few places that misused the copy constructor - those are fixed
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15416
Reviewed By: Yangqing
Differential Revision: D13524598
fbshipit-source-id: aea12d6dff77342606fa88ce4ddddbff266245a7
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15418
Previously we are using Resize + ShareData.
Instead, we'll create a function on Tensor that clones itself with same storage.
Suppose we want `t` to `ShareData` with `t0`, Previous:
```
Tensor t(dims, CPU);
t.Resize(t0.sizes());
t.ShareData(t0);
```
Now:
```
Tensor t = t0.Alias();
```
Reviewed By: dzhulgakov
Differential Revision: D13507609
fbshipit-source-id: 6e4275d02f4c3356cbce91127f1b01111dc86b9f
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15027
- Make DataRandomFiller able to accept input_dims and input_types for only non intermediate inputs. Add a helper to fill input directly to a workspace
Reviewed By: highker
Differential Revision: D13408345
fbshipit-source-id: 5fc54d33da12e3f0a200e79380d4c695b0339b17
Summary:
Hi guys,
I'd like to build Caffe2 with more supported options in Windows with Microsoft Visual Studios.
This is the first pull request.
Running scripts/build_windows_shared.bat is able to build Caffe2 with both CMAKE_BUILD_TYPE=Debug and CMAKE_BUILD_TYPE=Release with Visual Studio 14 2015.
CUDA is 9.0, cudnn is 7.0.5, glog, gflags and lmdb are supported on my system.
Python is 3.5, Detectron works from python interface as well.
It was even possible to debug detectron code and step into caffe2_gpu.dll with pdbs built.
What is disappointing, that c10/experimental ops don't build with this Visual Studio generator, I added special option INCLUDE_EXPERIMENTAL_C10_OPS (default ON) to deal with it in build_windows_shared.bat.
After this pull request the next step is to add Visual Studio 2017 support in the script.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13550
Reviewed By: ezyang
Differential Revision: D13042597
Pulled By: orionr
fbshipit-source-id: f313f909f599cd582a1d000eff766eef3a9fc4fc
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13949
This diff adds support to fillers for `SparseLengthsWeight*` ops. It does 3 things:
1. Add the fillers for `SparseLengthsWeight*` ops
2. Add filling heuristics to consider the path of `LengthsRangeFill` -> `Gather` -> `SparseLengthsWeightedSum`, where the length input is shared by `LengthsRangeFill` and `SparseLengthsWeightedSum`. Therefore, we need to carefully bound the value of that length input so that at `Gather`, it does not index out-of-bound for the weight input of `Gather`.
3. Fix and simplify the logic of `math::RandFixedSum`, where we just keep rejecting the generated value if it violates the invariants.
Reviewed By: highker
Differential Revision: D13048216
fbshipit-source-id: bfe402e07e6421b28548047d18b298c148e0ec87
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13820
We would like to provide an option to show additional info of the net to be benchmarked.
Reviewed By: highker, rdzhabarov
Differential Revision: D13018219
fbshipit-source-id: d3ec69901bdae58117a482ddd2c327b0f8cf7cb6
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13522
Currently Tensor is a shared pointer to the underlying implementation, rather than a value, copying
the pointer will share the underlying TensorImpl, ShareData probably don't make sense anymore.
Reviewed By: dzhulgakov
Differential Revision: D12871708
fbshipit-source-id: d3773c66b7ed0bf1c37e886f69f59aec158b216b
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12905
This diff does some clean up of the multithread benchmark code:
1. Split implementation to `.cc` file to separate implementation and improve build
2. Make `MutatingNetSupplier` more generic by providing the mutating function as an argument instead of virtual method.
3. Fix AI benchmark by sticking to the original option names
Reviewed By: highker
Differential Revision: D10479238
fbshipit-source-id: afa201fc287e3fdbb232db24513ecf8024501f66
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11135
This diff does not have any logic change; it simply move files/functions/classes around.
Open source (almost all) necessary dependency for multithreaded predictor bench.
The benchmark itself can be open sourced once the predictor is open sourced.
Reviewed By: salexspb
Differential Revision: D9602006
fbshipit-source-id: 386c9483e2c64c8b7d36e4600189c4e0b7e159ff
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12043
Re-trying D9979976, this time with all call sites fixed.
D9979976 got reverted because there was a call site that wasn't covered by sandcastle it seems.
I fixed it and used 'grep' to ensure there aren't any more call sites in fbsource.
Reviewed By: ezyang
Differential Revision: D10026392
fbshipit-source-id: cd341514a8e53a40147ea0ee3e52f63bb6444157
Summary: The controller you requested could not be found. Original commit changeset: 2ea17724e223
Differential Revision:
D10026321
Ninja: stable broken
fbshipit-source-id: faf87cb7cc0f78c2c10d4aa6fceea279cd27acd6
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11923
This is pre-work to allow moving Blob to ATen/core, which cannot depend on caffe2 anymore.
(1) Removing the Blob -> Tensor dependency allows us to move Blob to ATen/core and use it inside IValue without having to wait for the Tensor merge to be complete.
(2) In the final Blob design, we want it to be a very small class that doesn't have any special treatment for Tensor (or to be more correct, doesn't allow storing Tensor anymore), so this is anyhow the direction we want to go.
This changes call sites that will have to be moved to IValue later, but they cannot be moved to IValue directly, because for that, IValue first needs to be able to store Blob, which in turn first needs this diff and some other changes coming up in future diffs.
Codemods:
$ codemod --extensions h,hpp,c,cpp,cc "([a-zA-Z0-9_]+)\\.IsTensorType\\(" "BlobIsTensorType(\\1, "
$ codemod --extensions h,hpp,c,cpp,cc "([a-zA-Z0-9_]+)->IsTensorType\\(" "BlobIsTensorType(*\\1, "
$ codemod --extensions h,hpp,c,cpp,cc "([a-zA-Z0-9_]+)\\.GetMutableTensor\\(" "BlobGetMutableTensor(\\1, "
$ codemod --extensions h,hpp,c,cpp,cc "([a-zA-Z0-9_]+)->GetMutableTensor\\(" "BlobGetMutableTensor(*\\1, "
It is, however, not only these codemods because regex based refactoring was only able to match a small amount of the call sites. To catch more, I wouldn've needed a AST aware tool like clangr, which I didn't figure out how to use.
Reviewed By: ezyang
Differential Revision: D9979976
fbshipit-source-id: 2ea17724e223b5b73b44f99362727759ca689e61
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11817
Blob::Serialize() and Blob::Deserialize() are now free functions SerializeBlob(), DeserializeBlob() instead.
This takes away access to Blob internals from them and makes future refactorings easier.
Reviewed By: ezyang
Differential Revision: D9882726
fbshipit-source-id: 3251ebd4b53fc12f5e6924a6e4a8db3846ab3729
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11875
Seems like the refactor to predictor_config dropped some functionality that is now blocking other teams
rFBS2b30208263c14ce7039f27c618a3b232bf11ee33 is the change that was missed
hoping to land this quickly :)
Reviewed By: jonmorton
Differential Revision: D9948324
fbshipit-source-id: 1628f7c51c06319fa7ca5dc9d59799135bb82c5f
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11779
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11731
This Predictor provides threadsafe interface and also
cleans-up activations after each run. So in multi-model setup
activation space doesn't explode
Reviewed By: highker
Differential Revision: D9842374
fbshipit-source-id: bfe253ae5fc813e73a347c5147ff6b58d50781ea
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11167
Narrow the Blob API as preparation for merging Blob/IValue
- get rid of templated IsType and Operator::InputIsType / OutputIsType
- Use 'using' instead of 'typedef' for DestroyCall (just for readability)
Reviewed By: ezyang
Differential Revision: D9623916
fbshipit-source-id: 952f0b0cf5a525094b02e8d2798dd57a56a9e1d8
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11052
Delete the test case for Predictor with constructing by MetaNetDef since the constructor
actually has been deprecated. The broken PR is for construcing predictor from DB instance.
Reviewed By: highker
Differential Revision: D9566935
fbshipit-source-id: 5511883953a2d3f6eb0a4f1c5518a1bc4b3ffbdc
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/10920
Update the black box predictor and the related code to use the
constructor with PredictorConfig.
Reviewed By: highker
Differential Revision: D9516972
fbshipit-source-id: fbd7ece934d527e17dc6bcc740b4e67e778afa1d
Summary: Update all the caller for the new interface
Reviewed By: highker
Differential Revision: D9323167
fbshipit-source-id: a39335ceb402db0719f5f2314085ba9a81380308
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/10835
The last diff of constructor cause performance regression in cold run.
This one tried to fix this.
Reviewed By: highker
Differential Revision: D9489617
fbshipit-source-id: a77c2e2c903a73e2ad9806b4f9c209cdb751442f
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/9483
The interface is updated to accept the config to construct the predictor.
Reviewed By: highker
Differential Revision: D8872999
fbshipit-source-id: 3ca54d644970823fc33c0ade9a005e12f52e2b24
Summary:
Properly annotated all apis for cpu front. Checked with cmake using
cmake -DUSE_ATEN=ON -DUSE_CUDA=OFF -DBUILD_ATEN=ON
and resulting libcaffe2.so has about 11k symbols.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/10504
Reviewed By: ezyang
Differential Revision: D9316491
Pulled By: Yangqing
fbshipit-source-id: 215659abf350af7032e9a4b0f28a856babab2454
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/9939
Pull Request resolved: https://github.com/facebookresearch/weakly-supervised-action-detection/pull/13
Pull Request resolved: https://github.com/pytorch/translate/pull/166
Pull Request resolved: https://github.com/pytorch/pytorch/pull/9125
Closes https://github.com/pytorch/pytorch/pull/9125
Use inheritance for polymorphism, and remove template parameter
This is to change the templating in call sites, the core implementations will change later
Before Caffe2 Tensor class was compile-time fixed to bind to a particular device/context. With this change, we're making it a runtime property (stored inside the tensor), but preserve the same semantics. For example, one has to specify device type in order to create a Tensor - there are no uninitialized tensors. More specifically the changes are:
1. We added an extra argument *DeviceType* to most of the constructors of the tensor, e.g. (Tensor(DeviceType type)),
2. Semantics of constructor Tensor(const Tensor<SrcContext>& src, ContextForCopy* context); is changed, in this constructor, the second context is passed in to enable us to call the templated Copy function, it could be in a different context as source and target previously, now we'll enforce that the context should have same device type as src, if it is provided.
3. To preserve 'get-or-construct' semantics of Blob, we added specialized getter Blob::GetMutableTensor that verifies both that Blob contains a Tensor and that it's of a correct type
4. Specifically, Tensor type is not default-constructible any more (as we don't have unknown device tensors) and thus some of the code handling STL containers needs to change
Note: Some changes are postponed just to keep this diff a bit smaller. Please see `TODO`s.
Reviewed By: ezyang, houseroad
Differential Revision: D9024330
fbshipit-source-id: e0b8295d2dc6ebe2963383ded5af799ad17164ba
