Summary:
* s/environmental/environment/g
* Casing (CUDA, InfiniBand, Ethernet)
* Don't embed torch.multiprocessing.spawn but link to it (not part of the package)
* spawn _function_ instead of _utility_ (it's mentioned after the launch utility which is a proper utility)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14605
Differential Revision: D13273480
Pulled By: pietern
fbshipit-source-id: da6b4b788134645f2dcfdd666d1bbfc9aabd97b1
Summary:
Removed an incorrect section. We don't support this. I wrote this from my memory :(
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14530
Differential Revision: D13253471
Pulled By: teng-li
fbshipit-source-id: c3f1ffc6c98ef8789157e885776e0b775ec47b15
Summary:
Add to the Tensor doc info about `.device`, `.is_cuda`, `.requires_grad`, `.is_leaf` and `.grad`.
Update the `register_backward_hook` doc with a warning stating that it does not work in all cases.
Add support in the `_add_docstr` function to add docstring to attributes.
There is an explicit cast here but I am not sure how to handle it properly. The thing is that the doc field for getsetdescr is written as being a const char * (as all other doc fields in descriptors objects) in cpython online documentation. But in the code, it is the only one that is not const.
I assumed here that it is a bug in the code because it does not follow the doc and the convention of the others descriptors and so I cast out the const.
EDIT: the online doc I was looking at is for 3.7 and in that version both the code and the doc are const. For older versions, both are non const.
Please let me know if this should not be done. And if it should be done if there is a cleaner way to do it !
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14339
Differential Revision: D13243266
Pulled By: ezyang
fbshipit-source-id: 75b7838f7cd6c8dc72b0c61950e7a971baefaeeb
Summary:
- to fix#12241
- add `_sparse_sum()` to ATen, and expose as `torch.sparse.sum()`, not support `SparseTensor.sum()` currently
- this PR depends on #11253, and will need to be updated upon it lands
- [x] implement forward
- [x] implement backward
- performance [benchmark script](https://gist.github.com/weiyangfb/f4c55c88b6092ef8f7e348f6b9ad8946#file-sparse_sum_benchmark-py):
- sum all dims is fastest for sparse tensor
- when input is sparse enough nnz = 0.1%, sum of sparse tensor is faster than dense in CPU, but not necessary in CUDA
- CUDA backward is comparable (<2x) between `sum several dims` vs `sum all dims` in sparse
- CPU backward uses binary search is still slow in sparse, takes `5x` time in `sum [0, 2, 3] dims` vs `sum all dims`
- optimize CUDA backward for now
- using thrust for sort and binary search, but runtime not improved
- both of CPU and CUDA forward are slow in sparse (`sum several dims` vs `sum all dims`), at most `20x` slower in CPU, and `10x` in CUDA
- improve CPU and CUDA forward kernels
(nnz, sizes, sum_dims, keepdim, sum all or dims, bk=backward) | CPU (sparse vs dense) | CUDA(sparse vs dense)
-- | -- | --
(1000, [1000, 1000, 2, 2], [0, 1], False, sumAll) | 8.77 µs vs 72.9 µs | 42.5 µs vs 108 µs
(1000, [1000, 1000, 2, 2], [0, 1], False, sumD) | 112 µs vs 4.47 ms | 484 µs vs 407 µs
(1000, [1000, 1000, 2, 2], [0, 1], False, sumAll, bk) | 141 µs vs 148 µs | 647 µs vs 231 µs
(1000, [1000, 1000, 2, 2], [0, 1], False, sumD, bk) | 235 µs vs 1.23 ms | 781 µs vs 213 µs
(1000, [1000, 1000, 2, 2], [2, 3], False, sumD) | 48.5 µs vs 360 µs | 160 µs vs 2.03 ms
(1000, [1000, 1000, 2, 2], [2, 3], False, sumD, bk) | 258 µs vs 1.22 ms | 798 µs vs 224 µs
(1000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD) | 204 µs vs 882 µs | 443 µs vs 133 µs
(1000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD, bk) | 709 µs vs 1.15 ms | 893 µs vs 202 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumAll) | 39.8 µs vs 81 µs | 42.4 µs vs 113 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumD) | 747 µs vs 4.7 ms | 2.4 ms vs 414 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumAll, bk) | 1.04 ms vs 126 µs | 5.03 ms vs 231 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumD, bk) | 1.12 ms vs 1.24 ms | 5.99 ms vs 213 µs
(10000, [1000, 1000, 2, 2], [2, 3], False, sumD) | 133 µs vs 366 µs | 463 µs vs 2.03 ms
(10000, [1000, 1000, 2, 2], [2, 3], False, sumD, bk) | 1.56 ms vs 1.22 ms | 6.11 ms vs 229 µs
(10000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD) | 1.53 ms vs 799 µs | 824 µs vs 134 µs
(10000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD, bk) | 5.15 ms vs 1.09 ms | 7.02 ms vs 205 µs
- after improving CPU and CUDA forward kernels
- in `(1000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD)` forward, CPU takes ~~`171 µs`~~, in which `130 µs` is spent on `coalesce()`, for CUDA, total time is ~~`331 µs`~~, in which `141 µs` is spent on `coalesce()`, we need to reduce time at other places outside `coalesce()`.
- after a few simple tweaks, now in the forward, it is at most `10x` slower in CPU, and `7x` in CUDA. And time takes in `sum dense dims only [2, 3]` is `~2x` of `sum all dims`. Speed of `sum all sparse dims [0, 1]` is on bar with `sum all dims`
(nnz, sizes, sum_dims, keepdim, sum all or dims, bk=backward) | CPU (sparse vs dense) | CUDA(sparse vs dense)
-- | -- | --
(1000, [1000, 1000, 2, 2], [0, 1], False, sumAll) | 7 µs vs 69.5 µs | 31.5 µs vs 61.6 µs
(1000, [1000, 1000, 2, 2], [0, 1], False, sumD) | 11.3 µs vs 4.72 ms | 35.2 µs vs 285 µs
(1000, [1000, 1000, 2, 2], [0, 1], False, sumAll, bk) | 197 µs vs 124 µs | 857 µs vs 134 µs
(1000, [1000, 1000, 2, 2], [0, 1], False, sumD, bk) | 124 µs vs 833 µs | 796 µs vs 106 µs
(1000, [1000, 1000, 2, 2], [2, 3], False, sumD) | 20.5 µs vs 213 µs | 39.4 µs vs 1.24 ms
(1000, [1000, 1000, 2, 2], [2, 3], False, sumD, bk) | 131 µs vs 830 µs | 881 µs vs 132 µs
(1000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD) | 95.8 µs vs 409 µs | 246 µs vs 87.2 µs
(1000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD, bk) | 624 µs vs 820 µs | 953 µs vs 124 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumAll) | 45.3 µs vs 72.9 µs | 33.9 µs vs 57.2 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumD) | 81.4 µs vs 4.49 ms | 39.7 µs vs 280 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumAll, bk) | 984 µs vs 111 µs | 6.41 ms vs 121 µs
(10000, [1000, 1000, 2, 2], [0, 1], False, sumD, bk) | 1.45 ms vs 828 µs | 6.77 ms vs 113 µs
(10000, [1000, 1000, 2, 2], [2, 3], False, sumD) | 74.9 µs vs 209 µs | 37.7 µs vs 1.23 ms
(10000, [1000, 1000, 2, 2], [2, 3], False, sumD, bk) | 1.48 ms vs 845 µs | 6.96 ms vs 132 µs
(10000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD) | 1.14 ms vs 411 µs | 252 µs vs 87.8 µs
(10000, [1000, 1000, 2, 2], [0, 2, 3], False, sumD, bk) | 4.53 ms vs 851 µs | 7.12 ms vs 128 µs
- time takes in CUDA backward of sparse is super long with large variance (in case of nnz=10000, it normally takes 6-7ms). To improve backward of sparse ops, we will need to debug at places other than CUDA kernels. here is a benchmark of `torch.copy_()`:
```
>>> d = [1000, 1000, 2, 2]
>>> nnz = 10000
>>> I = torch.cat([torch.randint(0, d[0], size=(nnz,)),
torch.randint(0, d[1], size=(nnz,))], 0).reshape(2, nnz)
>>> V = torch.randn(nnz, d[2], d[3])
>>> size = torch.Size(d)
>>> S = torch.sparse_coo_tensor(I, V, size).coalesce().cuda()
>>> S2 = torch.sparse_coo_tensor(I, V, size).coalesce().cuda().requires_grad_()
>>> data = S2.clone()
>>> S.copy_(S2)
>>> y = S * 2
>>> torch.cuda.synchronize()
>>> %timeit y.backward(data, retain_graph=True); torch.cuda.synchronize()
7.07 ms ± 3.06 ms per loop (mean ± std. dev. of 7 runs, 1000 loops each)
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12430
Differential Revision: D12878313
Pulled By: weiyangfb
fbshipit-source-id: e16dc7681ba41fdabf4838cf05e491ca9108c6fe
Summary:
The doc covers pretty much all we have had on distributed for PT1 stable release, tracked in https://github.com/pytorch/pytorch/issues/14080
Tested by previewing the sphinx generated webpages. All look good.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14444
Differential Revision: D13227675
Pulled By: teng-li
fbshipit-source-id: 752f00df096af38dd36e4a337ea2120ffea79f86
Summary:
This issue was noticed, and fix proposed, by raulpuric.
Checkpointing is implemented by rerunning a forward-pass segment for each checkpointed segment during backward. This can result in the RNG state advancing more than it would without checkpointing, which can cause checkpoints that include dropout invocations to lose end-to-end bitwise accuracy as compared to non-checkpointed passes.
The present PR contains optional logic to juggle the RNG states such that checkpointed passes containing dropout achieve bitwise accuracy with non-checkpointed equivalents.** The user requests this behavior by supplying `preserve_rng_state=True` to `torch.utils.checkpoint` or `torch.utils.checkpoint_sequential`.
Currently, `preserve_rng_state=True` may incur a moderate performance hit because restoring MTGP states can be expensive. However, restoring Philox states is dirt cheap, so syed-ahmed's [RNG refactor](https://github.com/pytorch/pytorch/pull/13070#discussion_r235179882), once merged, will make this option more or less free.
I'm a little wary of the [def checkpoint(function, *args, preserve_rng_state=False):](https://github.com/pytorch/pytorch/pull/14253/files#diff-58da227fc9b1d56752b7dfad90428fe0R75) argument-passing method (specifically, putting a kwarg after a variable argument list). Python 3 seems happy with it.
Edit: It appears Python 2.7 is NOT happy with a [kwarg after *args](https://travis-ci.org/pytorch/pytorch/builds/457706518?utm_source=github_status&utm_medium=notification). `preserve_rng_state` also needs to be communicated in a way that doesn't break any existing usage. I'm open to suggestions (a global flag perhaps)?
**Batchnorm may still be an issue, but that's a battle for another day.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14253
Differential Revision: D13166665
Pulled By: soumith
fbshipit-source-id: 240cddab57ceaccba038b0276151342344eeecd7
Summary:
Update range documentation to show that we don't support start or increment parameters
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13730
Differential Revision: D12982016
Pulled By: eellison
fbshipit-source-id: cc1462fc1af547ae80c6d3b87999b7528bade8af
Summary:
The stylesheet at docs/source/_static/css/pytorch_theme.css is no longer necessary for the html docs build. The new html docs theme styles are located at https://github.com/pytorch/pytorch_sphinx_theme.
The Lato font is also no longer used in the new theme.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13699
Differential Revision: D12967448
Pulled By: soumith
fbshipit-source-id: 7de205162a61e3acacfd8b499660d328ff3812ec
Summary:
Also add docs for get_backend, Backend, and reduce_op
fixes#11803
cc The controller you requested could not be found. pietern apaszke
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11830
Differential Revision: D9927991
Pulled By: SsnL
fbshipit-source-id: a2ffb70826241ba84264f36f2cb173e00b19af48
Summary:
This PR performs a renaming of the function `potrf` responsible for the Cholesky
decomposition on positive definite matrices to `cholesky` as NumPy and TF do.
Billing of changes
- make potrf cname for cholesky in Declarations.cwrap
- modify the function names in ATen/core
- modify the function names in Python frontend
- issue warnings when potrf is called to notify users of the change
Reviewed By: soumith
Differential Revision: D10528361
Pulled By: zou3519
fbshipit-source-id: 19d9bcf8ffb38def698ae5acf30743884dda0d88
Summary:
[Edit: after applied colesbury 's suggestions]
* Hub module enable users to share code + pretrained weights through github repos.
Example usage:
```
hub_model = hub.load(
'ailzhang/vision:hub', # repo_owner/repo_name:branch
'wrapper1', # entrypoint
1234, # args for callable [not applicable to resnet18]
pretrained=True) # kwargs for callable
```
* Protocol on repo owner side: example https://github.com/ailzhang/vision/tree/hub
* The "published" models should be at least in a branch/tag. It can't be a random commit.
* Repo owner should have the following field defined in `hubconf.py`
* function/entrypoint with function signature `def wrapper1(pretrained=False, *args, **kwargs):`
* `pretrained` allows users to load pretrained weights from repo owner.
* `args` and `kwargs` are passed to the callable `resnet18`, repo owner should clearly specify their help message in the docstring
```
def wrapper1(pretrained=False, *args, **kwargs):
"""
pretrained (bool): a recommended kwargs for all entrypoints
args & kwargs are arguments for the function
"""
from torchvision.models.resnet import resnet18
model = resnet18(*args, **kwargs)
checkpoint = 'https://download.pytorch.org/models/resnet18-5c106cde.pth'
if pretrained:
model.load_state_dict(model_zoo.load_url(checkpoint, progress=False))
return model
```
* Hub_dir
* `hub_dir` specifies where the intermediate files/folders will be saved. By default this is `~/.torch/hub`.
* Users can change it by either setting the environment variable `TORCH_HUB_DIR` or calling `hub.set_dir(PATH_TO_HUB_DIR)`.
* By default, we don't cleanup files after loading so that users can use cache next time.
* Cache logic :
* We used the cache by default if it exists in `hub_dir`.
* Users can force a fresh reload by calling `hub.load(..., force_reload=True)`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12228
Differential Revision: D10511470
Pulled By: ailzhang
fbshipit-source-id: 12ac27f01d33653f06b2483655546492f82cce38
Summary:
Here is my stab at ```dense.to_sparse```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12171
Differential Revision: D10859078
Pulled By: weiyangfb
fbshipit-source-id: 5df72f72ba4f8f10e283402ff7731fd535682664
Summary:
include atomicAdd commentary as this is less well known
There is some discussion in #12207
Unfortunately, I cannot seem to get the ..include working in `_tensor_docs.py` and `_torch_docs.py`. I could use a hint for that.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12217
Differential Revision: D10419739
Pulled By: SsnL
fbshipit-source-id: eecd04fb7486bd9c6ee64cd34859d61a0a97ec4e
Summary:
This PR gets rid of unnecessary copy of weight gradients in cudnn rnn. Also removes unnecessary check for input size when deciding whether to use persistent rnn, and adds doc string explaining when persistent rnn can be used. cc ezyang
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12600
Differential Revision: D10359981
Pulled By: soumith
fbshipit-source-id: 0fce11b527d543fabf21e6e9213fb2879853d7fb
Summary:
- This was one of the few functions left out from the list of functions in
NumPy's `linalg` module
- `multi_mm` is particularly useful for DL research, for quick analysis of
deep linear networks
- Added tests and doc string
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12380
Differential Revision: D10357136
Pulled By: SsnL
fbshipit-source-id: 52b44fa18d6409bdeb76cbbb164fe4e88224458e
Summary:
- fix https://github.com/pytorch/pytorch/issues/12120
- add `torch.argsort`, `torch.pdist`, `broadcast_tensors` to *.rst files
- add parameter dim to `torch.unique` doc
- fix table and args for `torch.norm`
- test plan: make html and check docs in browser
gchanan
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12126
Differential Revision: D10087006
Pulled By: weiyangfb
fbshipit-source-id: 25f65c43d14e02140d0da988d8742c7ade3d8cc9
Summary:
goldsborough Modify the docs to match the changes made in #4999
Pull Request resolved: https://github.com/pytorch/pytorch/pull/12158
Differential Revision: D10103964
Pulled By: SsnL
fbshipit-source-id: 1b8692da86aca1a52e8d2e6cea76a5ad1f71e058
Summary:
Couple questions:
1) I used the log1p implementation in #8969 as a guide especially for testing. I'm not sure what the ```skipIfROCM``` annotation is for, so unsure if i need it for my test.
2) I implemented the branching logic in the narrow function itself; is this the right place to do so? I noticed that there a number of places where sparse-specific logic is handled with just an if statement in this file. Or should I implement a separate dispatch in native_functions.yml as in the log1p?
And of course, happy to make any any other updates/changes that I may have missed as well. This is my first PR to the project.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11342
Differential Revision: D9978430
Pulled By: weiyangfb
fbshipit-source-id: e73dc20302ab58925afb19e609e31f4a38c634ad
Summary:
Deleted this section by mistake in last PR.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11938
Reviewed By: SsnL
Differential Revision: D9993258
Pulled By: brianjo
fbshipit-source-id: 2552178cebd005a1105a22930c4d128c67247378
Summary:
A couple fixes I deem necessary to the TorchScript C++ API after writing the tutorial:
1. When I was creating the custom op API, I created `torch/op.h` as the one-stop header for creating custom ops. I now notice that there is no good header for the TorchScript C++ story altogether, i.e. when you just want to load a script module in C++ without any custom ops necessarily. The `torch/op.h` header suits that purpose just as well of course, but I think we should rename it to `torch/script.h`, which seems like a great name for this feature.
2. The current API for the CMake we provided was that we defined a bunch of variables like `TORCH_LIBRARY_DIRS` and `TORCH_INCLUDES` and then expected users to add those variables to their targets. We also had a CMake function that did that for you automatically. I now realized a much smarter way of doing this is to create an `IMPORTED` target for the libtorch library in CMake, and then add all this stuff to the link interface of that target. Then all downstream users have to do is `target_link_libraries(my_target torch)` and they get all the proper includes, libraries and compiler flags added to their target. This means we can get rid of the CMake function and all that stuff. orionr AFAIK this is a much, much better way of doing all of this, no?
3. Since we distribute libtorch with `D_GLIBCXX_USE_CXX11_ABI=0`, dependent libraries must set this flag too. I now add this to the interface compile options of this imported target.
4. Fixes to JIT docs.
These could likely be 4 different PRs but given the release I wouldn't mind landing them all asap.
zdevito dzhulgakov soumith
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11682
Differential Revision: D9839431
Pulled By: goldsborough
fbshipit-source-id: fdc47b95f83f22d53e1995aa683e09613b4bfe65
Summary:
This adds a Note on making experiments reproducible.
It also adds Instructions for building the Documentation to `README.md`. Please ping if I missed any requirements.
I'm not sure what to do about the submodule changes. Please advise.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11329
Differential Revision: D9784939
Pulled By: ezyang
fbshipit-source-id: 5c5acbe343d1fffb15bdcb84c6d8d925c2ffcc5e
Summary:
Ping ezyang
This addresses your comment in #114. Strangely, when running the doc build (`make html`) none of my changes are actually showing, could you point out what I'm doing wrong?
Once #11329 is merged it might make sense to link to the reproducibility note everywhere.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11434
Differential Revision: D9751208
Pulled By: ezyang
fbshipit-source-id: cc672472449564ff099323c39603e8ff2b2d35c9
Summary:
I'm 80% sure that this fixes the math bug. But I can't repro locally so I don't know.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11472
Differential Revision: D9755328
Pulled By: SsnL
fbshipit-source-id: 130be664d3c6ceee3c0c166c1a86fc9ec3b79d74
Summary:
vishwakftw Your patch needed some updates because the default native function dispatches changed from `[function, method]` to `[function]`. The CI was run before that change happened so it still shows green, but the internal test caught it.
I did some changes when rebasing and updating so I didn't just force push to your branch. Let's see if this passes CI and internal test. If it does, let me know if you want me to force push to your branch or use this PR instead.
Note to reviewers: patch was already approved at #10068 .
cc yf225
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11421
Differential Revision: D9733407
Pulled By: SsnL
fbshipit-source-id: cf2ed293bb9942dcc5158934ff4def2f63252599
Summary:
In addition to documentation, this cleans up a few error message formats.
It also adds infra to find which operators are supported by the JIT automatically, which is then used in the generation of the docs.
The wording and formatting of the docs is not yet polished, but having this will allow our document writers to make faster progress.
Followup PRs will polish the docs and fix formatting issues.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11357
Differential Revision: D9721277
Pulled By: zdevito
fbshipit-source-id: 153a0d5be1efb314511bcfc0cec48643d78ea48b
Summary:
This PR cleans up the `at::Tensor` class by removing all methods that start with an underscore in favor of functions in the `at::` namespace. This greatly cleans up the `Tensor` class and makes it clearer what is the public and non-public API.
For this I changed `native_functions.yaml` and `Declarations.cwrap` to make all underscore methods `variant: function` (or add such a statement to begin with), and then fixed all code locations using the underscore methods.
ezyang colesbury gchanan
Pull Request resolved: https://github.com/pytorch/pytorch/pull/11152
Differential Revision: D9683607
Pulled By: goldsborough
fbshipit-source-id: 97f869f788fa56639c05a439e2a33be49f10f543
