-
Notifications
You must be signed in to change notification settings - Fork 23.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Port TH operators to Aten (umbrella issue) #24507
Labels
better-engineering
Relatively self-contained tasks for better engineering contributors
module: porting
Issues related to porting TH/THNN legacy to ATen native
triaged
This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Comments
This was referenced Aug 16, 2019
facebook-github-bot
pushed a commit
that referenced
this issue
May 12, 2020
This was referenced May 13, 2020
facebook-github-bot
pushed a commit
that referenced
this issue
May 14, 2020
facebook-github-bot
pushed a commit
that referenced
this issue
May 26, 2020
Summary: References: #24521 #24522 #24547 #24548 #24507 Depends on #36308 Changes related to this PR are only in file : aten/src/ATen/Declarations.cwrap aten/src/ATen/native/cuda/ReduceOpsKernel.cu aten/src/ATen/native/native_functions.yaml aten/src/THC/generic/THCTensorMathScan.cu aten/src/THC/generic/THCTensorMathScan.h Please Review VitalyFedyunin Thanks. Pull Request resolved: #36458 Differential Revision: D21718384 Pulled By: ngimel fbshipit-source-id: 5af15164050c77be164397abd659a48c9ded2b29
facebook-github-bot
pushed a commit
that referenced
this issue
Sep 25, 2020
Summary: Related #24507 Fixes #24666 This PR is to modernize the CPU implementation of the vector `outer product`. The existing TH implementation for `torch.attr` is migrated to `aten`, as the `torch.ger` manipulates the `addr` functions to calculate outer product, Pull Request resolved: #44364 Reviewed By: ezyang Differential Revision: D23866733 Pulled By: mruberry fbshipit-source-id: 5159ea22f0e3c991123fe7c19cc9beb6ad00301e
This was referenced Dec 17, 2020
Mmm just created issues for |
facebook-github-bot
pushed a commit
that referenced
this issue
Jan 22, 2021
facebook-github-bot
pushed a commit
that referenced
this issue
Apr 3, 2021
facebook-github-bot
pushed a commit
that referenced
this issue
Jun 17, 2021
Summary: Fixes #24618 Related to #24507 <details><summary>Benchmark script:</summary> ```py import torch import torch.nn as nn import time torch.manual_seed(0) def _time(): torch.cuda.synchronize() return time.time() device = "cuda" m = nn.RReLU().cuda() for n in [100, 10_000, 100_000]: fwd_t = 0 bwd_t = 0 input = torch.randn(128, n, device=device) grad_output = torch.ones(128, n, device=device) for i in range(10000): t1 = _time() output = m(input) t2 = _time() fwd_t = fwd_t + (t2 -t1) fwd_avg = fwd_t / 10000 * 1000 print(f"input size(128, {n}) forward time is {fwd_avg:.2f} (ms)") ``` </details> ### Results from benchmark: #### This PR ``` input size(128, 100) forward time is 0.01 (ms) input size(128, 10000) forward time is 0.06 (ms) input size(128, 100000) forward time is 0.54 (ms) ``` #### On master ``` input size(128, 100) forward time is 0.01 (ms) input size(128, 10000) forward time is 0.08 (ms) input size(128, 100000) forward time is 0.66 (ms) ``` Pull Request resolved: #57864 Reviewed By: H-Huang Differential Revision: D29177169 Pulled By: ngimel fbshipit-source-id: 4572133db06f143d27e70a91ade977ea962c8f77
facebook-github-bot
pushed a commit
that referenced
this issue
Jun 22, 2021
Summary: Fixes #24609 Aten Umbrella issue #24507 Related to #59765 There are no performance differences when running the following benchmark: <details> <summary>Benchmark script</summary> ```python import torch import torch.nn as nn import time torch.manual_seed(0) def _time(): torch.cuda.synchronize() MS_PER_SECOND = 1000 return time.perf_counter() * MS_PER_SECOND device = "cuda" C = 30 softmax = nn.LogSoftmax(dim=1) n_runs = 250 for reduction in ["none", "mean", "sum"]: for N in [100_000, 500_000, 1_000_000]: elapsed = 0 for i in range(n_runs): data = torch.randn(N, C, device=device, requires_grad=True) target = torch.empty(N, dtype=torch.long, device=device).random_(0, C) loss = nn.NLLLoss(reduction=reduction) input = softmax(data) result = loss(input, target) if reduction == "none": gradient = torch.randn(N, device=device) else: gradient = torch.randn(1, device=device).squeeze() t1 = _time() result.backward(gradient) t2 = _time() elapsed = elapsed + (t2 - t1) elapsed_avg = elapsed / n_runs print( f"input size({N}, {C}), reduction: {reduction} " f"elapsed time is {elapsed_avg:.2f} (ms)" ) print() ``` </details> ## master ``` input size(100000, 30), reduction: none elapsed time is 0.19 (ms) input size(500000, 30), reduction: none elapsed time is 0.83 (ms) input size(1000000, 30), reduction: none elapsed time is 1.66 (ms) input size(100000, 30), reduction: mean elapsed time is 1.50 (ms) input size(500000, 30), reduction: mean elapsed time is 7.19 (ms) input size(1000000, 30), reduction: mean elapsed time is 14.35 (ms) input size(100000, 30), reduction: sum elapsed time is 1.49 (ms) input size(500000, 30), reduction: sum elapsed time is 7.17 (ms) input size(1000000, 30), reduction: sum elapsed time is 14.21 (ms) ``` ## this PR ``` input size(100000, 30), reduction: none elapsed time is 0.19 (ms) input size(500000, 30), reduction: none elapsed time is 0.83 (ms) input size(1000000, 30), reduction: none elapsed time is 1.66 (ms) input size(100000, 30), reduction: mean elapsed time is 1.48 (ms) input size(500000, 30), reduction: mean elapsed time is 7.16 (ms) input size(1000000, 30), reduction: mean elapsed time is 14.29 (ms) input size(100000, 30), reduction: sum elapsed time is 1.49 (ms) input size(500000, 30), reduction: sum elapsed time is 7.15 (ms) input size(1000000, 30), reduction: sum elapsed time is 14.18 (ms) ``` Pull Request resolved: #60299 Reviewed By: albanD Differential Revision: D29287613 Pulled By: ngimel fbshipit-source-id: 21e15f2c518087e9fb797a379e1e0a3508c98509
facebook-github-bot
pushed a commit
that referenced
this issue
Jun 23, 2021
Summary: Ref #24507 (There doesn't seem to be an actual issue for cross) This also moves the remaining operator functors in `THCTensorMathPointwise.cuh` to `SparseCUDATensorMath.cu` which is the only file using them. Pull Request resolved: #60039 Reviewed By: mrshenli Differential Revision: D29314638 Pulled By: ngimel fbshipit-source-id: aa7b57f6e11a933fb44f044e26945bb4a9e3de5f
facebook-github-bot
pushed a commit
that referenced
this issue
Jun 24, 2021
Summary: Fixes #24610 Aten Umbrella issue #24507 Related to #59765 The performance does not change between this PR and master with the following benchmark script: <details> <summary>Benchmark script</summary> ```python import torch import torch.nn as nn import time torch.manual_seed(0) def _time(): torch.cuda.synchronize() MS_PER_SECOND = 1000 return time.perf_counter() * MS_PER_SECOND device = "cuda" C = 30 softmax = nn.LogSoftmax(dim=1) n_runs = 250 for reduction in ["none", "mean", "sum"]: for N in [100_000, 500_000, 1_000_000]: fwd_t = 0 bwd_t = 0 data = torch.randn(N, C, device=device) target = torch.empty(N, dtype=torch.long, device=device).random_(0, C) loss = nn.NLLLoss(reduction=reduction) input = softmax(data) for i in range(n_runs): t1 = _time() result = loss(input, target) t2 = _time() fwd_t = fwd_t + (t2 - t1) fwd_avg = fwd_t / n_runs print( f"input size({N}, {C}), reduction: {reduction} " f"forward time is {fwd_avg:.2f} (ms)" ) print() ``` </details> ## master ``` input size(100000, 30), reduction: none forward time is 0.02 (ms) input size(500000, 30), reduction: none forward time is 0.08 (ms) input size(1000000, 30), reduction: none forward time is 0.15 (ms) input size(100000, 30), reduction: mean forward time is 1.81 (ms) input size(500000, 30), reduction: mean forward time is 8.24 (ms) input size(1000000, 30), reduction: mean forward time is 16.46 (ms) input size(100000, 30), reduction: sum forward time is 1.66 (ms) input size(500000, 30), reduction: sum forward time is 8.24 (ms) input size(1000000, 30), reduction: sum forward time is 16.46 (ms) ``` ## this PR ``` input size(100000, 30), reduction: none forward time is 0.02 (ms) input size(500000, 30), reduction: none forward time is 0.08 (ms) input size(1000000, 30), reduction: none forward time is 0.15 (ms) input size(100000, 30), reduction: mean forward time is 1.80 (ms) input size(500000, 30), reduction: mean forward time is 8.24 (ms) input size(1000000, 30), reduction: mean forward time is 16.46 (ms) input size(100000, 30), reduction: sum forward time is 1.66 (ms) input size(500000, 30), reduction: sum forward time is 8.24 (ms) input size(1000000, 30), reduction: sum forward time is 16.46 (ms) ``` Pull Request resolved: #60097 Reviewed By: mrshenli Differential Revision: D29303099 Pulled By: ngimel fbshipit-source-id: fc0d636543a79ea81158d286dcfb84043bec079a
The last open checkboxes were all done already. 280/280 ported, let's declare victory here 🎉 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
better-engineering
Relatively self-contained tasks for better engineering contributors
module: porting
Issues related to porting TH/THNN legacy to ATen native
triaged
This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Porting guide: https://github.com/pytorch/pytorch/wiki/TH-to-ATen-porting-guide
Example PR with porting of the adaptive_avg_pool2d: #14714
Example PR with porting of the point-wise operator
addcmul
: #22874How to use TensorIterator (if needed): https://github.com/pytorch/pytorch/wiki/How-to-use-TensorIterator
Notes:
[namedtensor ci]
into commit message, it will start names propagation test.Issues were automatically generated based on
legacy::
dispatch rules innative_functions.yaml
and operatorscname
inDeclarations.cwrap
CUDA Ops
Migrate
_index_copy_
from the TH to Aten (CUDA) #24523 Migrate_index_copy_
from the TH to Aten (CUDA)Migrate
_mode
from the TH to Aten (CUDA) #24526 Migrate_mode
from the TH to Aten (CUDA)Migrate
_multinomial_alias_draw
from the TH to Aten (CUDA) #24527 Migrate_multinomial_alias_draw
from the TH to Aten (CUDA)Migrate
_multinomial_alias_setup
from the TH to Aten (CUDA) #24528 Migrate_multinomial_alias_setup
from the TH to Aten (CUDA)Migrate
cholesky_inverse
from the TH to Aten (CUDA) #24543 Migratecholesky_inverse
from the TH to Aten (CUDA)Migrate
fmod
andfmod_
from the TH to Aten (CUDA) #24565 Migratefmod
andfmod_
from the TH to Aten (CUDA)Migrate
geqrf
from the TH to Aten (CUDA) #24569 Migrategeqrf
from the TH to Aten (CUDA)Migrate
glu
from the TH to Aten (CUDA) #24571 Migrateglu
from the TH to Aten (CUDA)Migrate
glu_backward
from the TH to Aten (CUDA) #24572 Migrateglu_backward
from the TH to Aten (CUDA)Migrate
index_fill_
from the TH to Aten (CUDA) #24577 Migrateindex_fill_
from the TH to Aten (CUDA)Migrate
log_sigmoid_backward
from the TH to Aten (CUDA) #24590 Migratelog_sigmoid_backward
from the TH to Aten (CUDA)Migrate
log_sigmoid_forward
from the TH to Aten (CUDA) #24591 Migratelog_sigmoid_forward
from the TH to Aten (CUDA)Migrate
lstsq
from the TH to Aten (CUDA) #24592 Migratelstsq
from the TH to Aten (CUDA)Migrate
mode
from the TH to Aten (CUDA) #24597 Migratemode
from the TH to Aten (CUDA)Migrate
multi_margin_loss
from the TH to Aten (CUDA) #24600 Migratemulti_margin_loss
from the TH to Aten (CUDA)Migrate
multi_margin_loss_backward
from the TH to Aten (CUDA) #24601 Migratemulti_margin_loss_backward
from the TH to Aten (CUDA)Migrate
multilabel_margin_loss_backward
from the TH to Aten (CUDA) #24602 Migratemultilabel_margin_loss_backward
from the TH to Aten (CUDA)Migrate
multilabel_margin_loss_forward
from the TH to Aten (CUDA) #24603 Migratemultilabel_margin_loss_forward
from the TH to Aten (CUDA)Migrate
nll_loss2d_backward
from the TH to Aten (CUDA) #24607 Migratenll_loss2d_backward
from the TH to Aten (CUDA)Migrate
nll_loss2d_forward
from the TH to Aten (CUDA) #24608 Migratenll_loss2d_forward
from the TH to Aten (CUDA)Migrate
nll_loss_backward
from the TH to Aten (CUDA) #24609 Migratenll_loss_backward
from the TH to Aten (CUDA)Migrate
nll_loss_forward
from the TH to Aten (CUDA) #24610 Migratenll_loss_forward
from the TH to Aten (CUDA)Migrate
put_
from the TH to Aten (CUDA) #24614 Migrateput_
from the TH to Aten (CUDA)Migrate
renorm
andrenorm_
from the TH to Aten (CUDA) #24616 Migraterenorm
andrenorm_
from the TH to Aten (CUDA)Migrate
rrelu_with_noise
andrrelu_with_noise_
from the TH to Aten (CUDA) #24618 Migraterrelu_with_noise
andrrelu_with_noise_
from the TH to Aten (CUDA)Migrate
sigmoid_backward
from the TH to Aten (CUDA) #24625 Migratesigmoid_backward
from the TH to Aten (CUDA)Migrate
sort
from the TH to Aten (CUDA) #24637 Migratesort
from the TH to Aten (CUDA)Migrate
take
from the TH to Aten (CUDA) #24640 Migratetake
from the TH to Aten (CUDA)Migrate
tanh_backward
from the TH to Aten (CUDA) #24643 Migratetanh_backward
from the TH to Aten (CUDA)Migrate
thnn_conv2d_backward
from the TH to Aten (CUDA) #24644 Migratethnn_conv2d_backward
from the TH to Aten (CUDA)Migrate
thnn_conv2d_forward
from the TH to Aten (CUDA) #24645 Migratethnn_conv2d_forward
from the TH to Aten (CUDA)Migrate
thnn_conv_depthwise2d_backward
from the TH to Aten (CUDA) #24646 Migratethnn_conv_depthwise2d_backward
from the TH to Aten (CUDA)Migrate
thnn_conv_depthwise2d_forward
from the TH to Aten (CUDA) #24647 Migratethnn_conv_depthwise2d_forward
from the TH to Aten (CUDA)Migrate
topk
from the TH to Aten (CUDA) #24648 Migratetopk
from the TH to Aten (CUDA)Migrate
_addr
and_addr_
from the TH to Aten (CUDA) #24519 Migrate_addr
and_addr_
from the TH to Aten (CUDA)Migrate
_cat
from the TH to Aten (CUDA) #24520 Migrate_cat
from the TH to Aten (CUDA)Migrate
_cumprod
from the TH to Aten (CUDA) #24521 Migrate_cumprod
from the TH to Aten (CUDA)Migrate
_cumsum
from the TH to Aten (CUDA) #24522 Migrate_cumsum
from the TH to Aten (CUDA)Migrate
_max
from the TH to Aten (CUDA) #24524 Migrate_max
from the TH to Aten (CUDA)Migrate
_min
from the TH to Aten (CUDA) #24525 Migrate_min
from the TH to Aten (CUDA)Migrate
_std
from the TH to Aten (CUDA) #24529 Migrate_std
from the TH to Aten (CUDA)Migrate
_var
from the TH to Aten (CUDA) #24530 Migrate_var
from the TH to Aten (CUDA)Migrate
addbmm
andaddbmm_
from the TH to Aten (CUDA) #24533 Migrateaddbmm
andaddbmm_
from the TH to Aten (CUDA)Migrate
addmm
andaddmm_
from the TH to Aten (CUDA) #24534 Migrateaddmm
andaddmm_
from the TH to Aten (CUDA)Migrate
addmv
andaddmv_
from the TH to Aten (CUDA) #24535 Migrateaddmv
andaddmv_
from the TH to Aten (CUDA)Migrate
addr
andaddr_
from the TH to Aten (CUDA) #24536 Migrateaddr
andaddr_
from the TH to Aten (CUDA)Migrate
atan
andatan_
from the TH to Aten (CUDA) #24538 Migrateatan
andatan_
from the TH to Aten (CUDA)Migrate
baddbmm
andbaddbmm_
from the TH to Aten (CUDA) #24539 Migratebaddbmm
andbaddbmm_
from the TH to Aten (CUDA)Migrate
binary_cross_entropy
from the TH to Aten (CUDA) #24540 Migratebinary_cross_entropy
from the TH to Aten (CUDA)Migrate
binary_cross_entropy_backward
from the TH to Aten (CUDA) #24541 Migratebinary_cross_entropy_backward
from the TH to Aten (CUDA)Migrate
clamp
andclamp_
from the TH to Aten (CUDA) #24544 Migrateclamp
andclamp_
from the TH to Aten (CUDA)Migrate
cos
andcos_
from the TH to Aten (CUDA) #24545 Migratecos
andcos_
from the TH to Aten (CUDA)Migrate
cosh
andcosh_
from the TH to Aten (CUDA) #24546 Migratecosh
andcosh_
from the TH to Aten (CUDA)Migrate
cumprod
from the TH to Aten (CUDA) #24547 Migratecumprod
from the TH to Aten (CUDA)Migrate
cumsum
from the TH to Aten (CUDA) #24548 Migratecumsum
from the TH to Aten (CUDA)Migrate
diag
from the TH to Aten (CUDA) #24549 Migratediag
from the TH to Aten (CUDA)Migrate
dist
from the TH to Aten (CUDA) #24551 Migratedist
from the TH to Aten (CUDA)Migrate
dot
from the TH to Aten (CUDA) #24552 Migratedot
from the TH to Aten (CUDA)Migrate
eig
from the TH to Aten (CUDA) #24553 Migrateeig
from the TH to Aten (CUDA)Migrate
elu
andelu_
from the TH to Aten (CUDA) #24554 Migrateelu
andelu_
from the TH to Aten (CUDA)Migrate
elu_backward
from the TH to Aten (CUDA) #24555 Migrateelu_backward
from the TH to Aten (CUDA)Migrate
equal
from the TH to Aten (CUDA) #24557 Migrateequal
from the TH to Aten (CUDA)Migrate
erf
anderf_
from the TH to Aten (CUDA) #24558 Migrateerf
anderf_
from the TH to Aten (CUDA)Migrate
erfc
anderfc_
from the TH to Aten (CUDA) #24559 Migrateerfc
anderfc_
from the TH to Aten (CUDA)Migrate
exp
andexp_
from the TH to Aten (CUDA) #24561 Migrateexp
andexp_
from the TH to Aten (CUDA)Migrate
gather
from the TH to Aten (CUDA) #24567 Migrategather
from the TH to Aten (CUDA)Migrate
ger
from the TH to Aten (CUDA) #24570 Migrateger
from the TH to Aten (CUDA)Migrate
hardtanh
andhardtanh_
from the TH to Aten (CUDA) #24574 Migratehardtanh
andhardtanh_
from the TH to Aten (CUDA)Migrate
hardtanh_backward
from the TH to Aten (CUDA) #24575 Migratehardtanh_backward
from the TH to Aten (CUDA)Migrate
index_add_
from the TH to Aten (CUDA) #24576 Migrateindex_add_
from the TH to Aten (CUDA)Migrate
index_select
from the TH to Aten (CUDA) #24578 Migrateindex_select
from the TH to Aten (CUDA)Migrate
leaky_relu
andleaky_relu_
from the TH to Aten (CUDA) #24583 Migrateleaky_relu
andleaky_relu_
from the TH to Aten (CUDA)Migrate
leaky_relu_backward
from the TH to Aten (CUDA) #24584 Migrateleaky_relu_backward
from the TH to Aten (CUDA)Migrate
masked_select
from TH to ATen (CUDA) #33054 Migratemasked_select
from TH to Aten (CUDA)Migrate
masked_scatter
from TH to ATen (CUDA) #49542 Migratemasked_scatter
from TH to Aten (CUDA)Migrate
masked_fill
from TH to ATen (CUDA) #49543 Migratemasked_fill
from TH to Aten (CUDA)Migrate
max
from the TH to Aten (CUDA) #24594 Migratemax
from the TH to Aten (CUDA)Migrate
min
from the TH to Aten (CUDA) #24595 Migratemin
from the TH to Aten (CUDA)Migrate
mm
from the TH to Aten (CUDA) #24596 Migratemm
from the TH to Aten (CUDA)Migrate
mv
from the TH to Aten (CUDA) #24605 Migratemv
from the TH to Aten (CUDA)Migrate
nonzero
from the TH to Aten (CUDA) #24611 Migratenonzero
from the TH to Aten (CUDA)Migrate
remainder
andremainder_
from the TH to Aten (CUDA) #24615 Migrateremainder
andremainder_
from the TH to Aten (CUDA)Migrate
rrelu_with_noise_backward
from the TH to Aten (CUDA) #24619 Migraterrelu_with_noise_backward
from the TH to Aten (CUDA)Migrate
scatter
andscatter_
from the TH to Aten (CUDA) #24621 Migratescatter
andscatter_
from the TH to Aten (CUDA)Migrate
scatter_add_
from the TH to Aten (CUDA) #24622 Migratescatter_add_
from the TH to Aten (CUDA)Migrate
set_
from the TH to Aten (CUDA) #24623 Migrateset_
from the TH to Aten (CUDA)Migrate
soft_margin_loss
from the TH to Aten (CUDA) #24631 Migratesoft_margin_loss
from the TH to Aten (CUDA)Migrate
soft_margin_loss_backward
from the TH to Aten (CUDA) #24632 Migratesoft_margin_loss_backward
from the TH to Aten (CUDA)Migrate
softplus
from the TH to Aten (CUDA) #24633 Migratesoftplus
from the TH to Aten (CUDA)Migrate
softplus_backward
from the TH to Aten (CUDA) #24634 Migratesoftplus_backward
from the TH to Aten (CUDA)Migrate
softshrink
from the TH to Aten (CUDA) #24635 Migratesoftshrink
from the TH to Aten (CUDA)Migrate
softshrink_backward
from the TH to Aten (CUDA) #24636 Migratesoftshrink_backward
from the TH to Aten (CUDA)Migrate
std
from the TH to Aten (CUDA) #24639 Migratestd
from the TH to Aten (CUDA)Migrate
tan
andtan_
from the TH to Aten (CUDA) #24641 Migratetan
andtan_
from the TH to Aten (CUDA)Migrate
tanh
andtanh_
from the TH to Aten (CUDA) #24642 Migratetanh
andtanh_
from the TH to Aten (CUDA)Migrate
trace
from the TH to Aten (CUDA) #24649 Migratetrace
from the TH to Aten (CUDA)Migrate
var
from the TH to Aten (CUDA) #24652 Migratevar
from the TH to Aten (CUDA)CPU Ops
addmm
andaddmm_
from the TH to Aten (CPU) #24679 Migrate_addmm
and_addmm_
from the TH to Aten (CPU)_addr
and_addr_
from the TH to Aten (CPU) #24666 Migrate_addr
and_addr_
from the TH to Aten (CPU)_cat
from the TH to Aten (CPU) #24667 Migrate_cat
from the TH to Aten (CPU)_cumprod
from the TH to Aten (CPU) #24668 Migrate_cumprod
from the TH to Aten (CPU)_cumsum
from the TH to Aten (CPU) #24669 Migrate_cumsum
from the TH to Aten (CPU)_index_copy_
from the TH to Aten (CPU) #24670 Migrate_index_copy_
from the TH to Aten (CPU)_max
from the TH to Aten (CPU) #24671 Migrate_max
from the TH to Aten (CPU)_min
from the TH to Aten (CPU) #24672 Migrate_min
from the TH to Aten (CPU)_mode
from the TH to Aten (CPU) #24673 Migrate_mode
from the TH to Aten (CPU)_multinomial_alias_draw
from the TH to Aten (CPU) #24674 Migrate_multinomial_alias_draw
from the TH to Aten (CPU)_multinomial_alias_setup
from the TH to Aten (CPU) #24675 Migrate_multinomial_alias_setup
from the TH to Aten (CPU)_std
from the TH to Aten (CPU) #24676 Migrate_std
from the TH to Aten (CPU)_var
from the TH to Aten (CPU) #24677 Migrate_var
from the TH to Aten (CPU)addbmm
andaddbmm_
from the TH to Aten (CPU) #24678 Migrateaddbmm
andaddbmm_
from the TH to Aten (CPU)addmm
andaddmm_
from the TH to Aten (CPU) #24679 Migrateaddmm
andaddmm_
from the TH to Aten (CPU)addmv
andaddmv_
from the TH to Aten (CPU) #24680 Migrateaddmv
andaddmv_
from the TH to Aten (CPU)addr
andaddr_
from the TH to Aten (CPU) #24681 Migrateaddr
andaddr_
from the TH to Aten (CPU)binary_cross_entropy
from the TH to Aten (CPU) #24682 Migratebinary_cross_entropy
from the TH to Aten (CPU)binary_cross_entropy_backward
from the TH to Aten (CPU) #24683 Migratebinary_cross_entropy_backward
from the TH to Aten (CPU)cauchy_
from the TH to Aten (CPU) #24684 Migratecauchy_
from the TH to Aten (CPU)cholesky_inverse
from the TH to Aten (CPU) #24685 Migratecholesky_inverse
from the TH to Aten (CPU)cumprod
from the TH to Aten (CPU) #24687 Migratecumprod
from the TH to Aten (CPU)cumsum
from the TH to Aten (CPU) #24688 Migratecumsum
from the TH to Aten (CPU)diag
from the TH to Aten (CPU) #24689 Migratediag
from the TH to Aten (CPU)dist
from the TH to Aten (CPU) #24691 Migratedist
from the TH to Aten (CPU)dot
from the TH to Aten (CPU) #24692 Migratedot
from the TH to Aten (CPU)eig
from the TH to Aten (CPU) #24693 Migrateeig
from the TH to Aten (CPU)elu
andelu_
from the TH to Aten (CPU) #24694 Migrateelu
andelu_
from the TH to Aten (CPU)elu_backward
from the TH to Aten (CPU) #24695 Migrateelu_backward
from the TH to Aten (CPU)equal
from the TH to Aten (CPU) #24697 Migrateequal
from the TH to Aten (CPU)exponential_
from the TH to Aten (CPU) #24699 Migrateexponential_
from the TH to Aten (CPU)fmod
andfmod_
from the TH to Aten (CPU) #24701 Migratefmod
andfmod_
from the TH to Aten (CPU)gather
from the TH to Aten (CPU) #24702 Migrategather
from the TH to Aten (CPU)geometric_
from the TH to Aten (CPU) #24704 Migrategeometric_
from the TH to Aten (CPU)geqrf
from the TH to Aten (CPU) #24705 Migrategeqrf
from the TH to Aten (CPU)ger
from the TH to Aten (CPU) #24706 Migrateger
from the TH to Aten (CPU)glu
from the TH to Aten (CPU) #24707 Migrateglu
from the TH to Aten (CPU)glu_backward
from the TH to Aten (CPU) #24708 Migrateglu_backward
from the TH to Aten (CPU)hardtanh
andhardtanh_
from the TH to Aten (CPU) #24710 Migratehardtanh
andhardtanh_
from the TH to Aten (CPU)hardtanh_backward
from the TH to Aten (CPU) #24711 Migratehardtanh_backward
from the TH to Aten (CPU)histc
from the TH to Aten (CPU) #24712 Migratehistc
from the TH to Aten (CPU)index_add_
from the TH to Aten (CPU) #24713 Migrateindex_add_
from the TH to Aten (CPU)index_fill_
from the TH to Aten (CPU) #24714 Migrateindex_fill_
from the TH to Aten (CPU)index_select
from the TH to Aten (CPU) #24715 Migrateindex_select
from the TH to Aten (CPU)leaky_relu
andleaky_relu_
from the TH to Aten (CPU) #24720 Migrateleaky_relu
andleaky_relu_
from the TH to Aten (CPU)leaky_relu_backward
from the TH to Aten (CPU) #24721 Migrateleaky_relu_backward
from the TH to Aten (CPU)log_normal_
from the TH to Aten (CPU) #24723 Migratelog_normal_
from the TH to Aten (CPU)log_sigmoid_backward
from the TH to Aten (CPU) #24724 Migratelog_sigmoid_backward
from the TH to Aten (CPU)log_sigmoid_forward
from the TH to Aten (CPU) #24725 Migratelog_sigmoid_forward
from the TH to Aten (CPU)lstsq
from the TH to Aten (CPU) #24726 Migratelstsq
from the TH to Aten (CPU)masked_select
from TH to ATen (CPU) #33053 Migratemasked_select
from TH to Aten (CPU)masked_scatter
from TH to ATen (CPU) #49541 Migratemasked_scatter
from TH to Aten (CPU)max
from the TH to Aten (CPU) #24728 Migratemax
from the TH to Aten (CPU)min
from the TH to Aten (CPU) #24729 Migratemin
from the TH to Aten (CPU)mm
from the TH to Aten (CPU) #24730 Migratemm
from the TH to Aten (CPU)mode
from the TH to Aten (CPU) #24731 Migratemode
from the TH to Aten (CPU)mv
from the TH to Aten (CPU) #24739 Migratemv
from the TH to Aten (CPU)nonzero
from the TH to Aten (CPU) #24745 Migratenonzero
from the TH to Aten (CPU)normal
andnormal_
from the TH to Aten (CPU) #24746 Migratenormal
andnormal_
from the TH to Aten (CPU)orgqr
from the TH to Aten (CPU) #24747 Migrateorgqr
from the TH to Aten (CPU)ormqr
from the TH to Aten (CPU) #24748 Migrateormqr
from the TH to Aten (CPU)put_
from the TH to Aten (CPU) #24751 Migrateput_
from the TH to Aten (CPU)random_
from the TH to Aten (CPU) #24752 Migraterandom_
from the TH to Aten (CPU)remainder
andremainder_
from the TH to Aten (CPU) #24753 Migrateremainder
andremainder_
from the TH to Aten (CPU)renorm
andrenorm_
from the TH to Aten (CPU) #24754 Migraterenorm
andrenorm_
from the TH to Aten (CPU)rrelu_with_noise
andrrelu_with_noise_
from the TH to Aten (CPU) #24755 Migraterrelu_with_noise
andrrelu_with_noise_
from the TH to Aten (CPU)rrelu_with_noise_backward
from the TH to Aten (CPU) #24756 Migraterrelu_with_noise_backward
from the TH to Aten (CPU)scatter
andscatter_
from the TH to Aten (CPU) #24757 Migratescatter
andscatter_
from the TH to Aten (CPU)scatter_add_
from the TH to Aten (CPU) #24758 Migratescatter_add_
from the TH to Aten (CPU)set_
from the TH to Aten (CPU) #24759 Migrateset_
from the TH to Aten (CPU)sigmoid_backward
from the TH to Aten (CPU) #24760 Migratesigmoid_backward
from the TH to Aten (CPU)soft_margin_loss
from the TH to Aten (CPU) #24764 Migratesoft_margin_loss
from the TH to Aten (CPU)soft_margin_loss_backward
from the TH to Aten (CPU) #24765 Migratesoft_margin_loss_backward
from the TH to Aten (CPU)softplus
from the TH to Aten (CPU) #24766 Migratesoftplus
from the TH to Aten (CPU)softplus_backward
from the TH to Aten (CPU) #24767 Migratesoftplus_backward
from the TH to Aten (CPU)softshrink
from the TH to Aten (CPU) #24768 Migratesoftshrink
from the TH to Aten (CPU)softshrink_backward
from the TH to Aten (CPU) #24769 Migratesoftshrink_backward
from the TH to Aten (CPU)sort
from the TH to Aten (CPU) #24770 Migratesort
from the TH to Aten (CPU)std
from the TH to Aten (CPU) #24771 Migratestd
from the TH to Aten (CPU)take
from the TH to Aten (CPU) #24772 Migratetake
from the TH to Aten (CPU)tanh_backward
from the TH to Aten (CPU) #24773 Migratetanh_backward
from the TH to Aten (CPU)trace
from the TH to Aten (CPU) #24779 Migratetrace
from the TH to Aten (CPU)uniform_
from the TH to Aten (CPU) #24781 Migrateuniform_
from the TH to Aten (CPU)var
from the TH to Aten (CPU) #24782 Migratevar
from the TH to Aten (CPU)Pointwise bit-ops
__and__
from the TH to Aten (CUDA) #24508 Migrate__and__
from the TH to Aten (CUDA)__iand__
from the TH to Aten (CUDA) #24509 Migrate__iand__
from the TH to Aten (CUDA)__ilshift__
from the TH to Aten (CUDA) #24510 Migrate__ilshift__
from the TH to Aten (CUDA)__ior__
from the TH to Aten (CUDA) #24511 Migrate__ior__
from the TH to Aten (CUDA)__irshift__
from the TH to Aten (CUDA) #24512 Migrate__irshift__
from the TH to Aten (CUDA)__lshift__
from the TH to Aten (CUDA) #24514 Migrate__lshift__
from the TH to Aten (CUDA)__or__
from the TH to Aten (CUDA) #24515 Migrate__or__
from the TH to Aten (CUDA)__rshift__
from the TH to Aten (CUDA) #24516 Migrate__rshift__
from the TH to Aten (CUDA)__and__
from the TH to Aten (CPU) #24655 Migrate__and__
from the TH to Aten (CPU)__iand__
from the TH to Aten (CPU) #24656 Migrate__iand__
from the TH to Aten (CPU)__ilshift__
from the TH to Aten (CPU) #24657 Migrate__ilshift__
from the TH to Aten (CPU)__ior__
from the TH to Aten (CPU) #24658 Migrate__ior__
from the TH to Aten (CPU)__irshift__
from the TH to Aten (CPU) #24659 Migrate__irshift__
from the TH to Aten (CPU)__lshift__
from the TH to Aten (CPU) #24661 Migrate__lshift__
from the TH to Aten (CPU)__or__
from the TH to Aten (CPU) #24662 Migrate__or__
from the TH to Aten (CPU)__rshift__
from the TH to Aten (CPU) #24663 Migrate__rshift__
from the TH to Aten (CPU)Completed (2019-11-20)
ne
andne_
from the TH to Aten (CUDA) #24606 Migratene
andne_
from the TH to Aten (CUDA)ne
andne_
from the TH to Aten (CPU) #24740 Migratene
andne_
from the TH to Aten (CPU)gt
andgt_
from the TH to Aten (CUDA) #24573 Migrategt
andgt_
from the TH to Aten (CUDA)gt
andgt_
from the TH to Aten (CPU) #24709 Migrategt
andgt_
from the TH to Aten (CPU)lt
andlt_
from the TH to Aten (CUDA) #24593 Migratelt
andlt_
from the TH to Aten (CUDA)lt
andlt_
from the TH to Aten (CPU) #24727 Migratelt
andlt_
from the TH to Aten (CPU)eq
andeq_
from the TH to Aten (CUDA) #24556 Migrateeq
andeq_
from the TH to Aten (CUDA)eq
andeq_
from the TH to Aten (CPU) #24696 Migrateeq
andeq_
from the TH to Aten (CPU)ge
andge_
from the TH to Aten (CUDA) #24568 Migratege
andge_
from the TH to Aten (CUDA)ge
andge_
from the TH to Aten (CPU) #24703 Migratege
andge_
from the TH to Aten (CPU)le
andle_
from the TH to Aten (CUDA) #24582 Migratele
andle_
from the TH to Aten (CUDA)le
andle_
from the TH to Aten (CPU) #24719 Migratele
andle_
from the TH to Aten (CPU)__ixor__
from the TH to Aten (CPU) #24660 Migrate__ixor__
from the TH to Aten (CPU)__xor__
from the TH to Aten (CUDA) #24517 Migrate__xor__
from the TH to Aten (CUDA)__xor__
from the TH to Aten (CPU) #24664 Migrate__xor__
from the TH to Aten (CPU)__ixor__
from the TH to Aten (CUDA) #24513 Migrate__ixor__
from the TH to Aten (CUDA)is_set_to
from the TH to Aten (CPU) #24716 Migrateis_set_to
from the TH to Aten (CPU)l1_loss
from the TH to Aten (CPU) #24717 Migratel1_loss
from the TH to Aten (CPU)l1_loss_backward
from the TH to Aten (CPU) #24718 Migratel1_loss_backward
from the TH to Aten (CPU)lgamma
andlgamma_
from the TH to Aten (CPU) #24722 Migratelgamma
andlgamma_
from the TH to Aten (CPU)mse_loss
from the TH to Aten (CPU) #24732 Migratemse_loss
from the TH to Aten (CPU)mse_loss_backward
from the TH to Aten (CPU) #24733 Migratemse_loss_backward
from the TH to Aten (CPU)multi_margin_loss
from the TH to Aten (CPU) #24734 Migratemulti_margin_loss
from the TH to Aten (CPU)multi_margin_loss_backward
from the TH to Aten (CPU) #24735 Migratemulti_margin_loss_backward
from the TH to Aten (CPU)multilabel_margin_loss_backward
from the TH to Aten (CPU) #24736 Migratemultilabel_margin_loss_backward
from the TH to Aten (CPU)multilabel_margin_loss_forward
from the TH to Aten (CPU) #24737 Migratemultilabel_margin_loss_forward
from the TH to Aten (CPU)multinomial
from the TH to Aten (CPU) #24738 Migratemultinomial
from the TH to Aten (CPU)polygamma
andpolygamma_
from the TH to Aten (CPU) #24749 Migratepolygamma
andpolygamma_
from the TH to Aten (CPU)pow
andpow_
from the TH to Aten (CPU) #24750 Migratepow
andpow_
from the TH to Aten (CPU)nll_loss2d_backward
from the TH to Aten (CPU) #24741 Migratenll_loss2d_backward
from the TH to Aten (CPU)nll_loss2d_forward
from the TH to Aten (CPU) #24742 Migratenll_loss2d_forward
from the TH to Aten (CPU)nll_loss_backward
from the TH to Aten (CPU) #24743 Migratenll_loss_backward
from the TH to Aten (CPU)nll_loss_forward
from the TH to Aten (CPU) #24744 Migratenll_loss_forward
from the TH to Aten (CPU)zero_
from the TH to Aten (CPU) #24783 Migratezero_
from the TH to Aten (CPU)unfold
from the TH to Aten (CPU) #24780 Migrateunfold
from the TH to Aten (CPU)thnn_conv2d_backward
from the TH to Aten (CPU) #24774 Migratethnn_conv2d_backward
from the TH to Aten (CPU)thnn_conv2d_forward
from the TH to Aten (CPU) #24775 Migratethnn_conv2d_forward
from the TH to Aten (CPU)thnn_conv3d_backward
from the TH to Aten (CPU) #24776 Migratethnn_conv3d_backward
from the TH to Aten (CPU)thnn_conv3d_forward
from the TH to Aten (CPU) #24777 Migratethnn_conv3d_forward
from the TH to Aten (CPU)topk
from the TH to Aten (CPU) #24778 Migratetopk
from the TH to Aten (CPU)sign
andsign_
from the TH to Aten (CPU) #24761 Migratesign
andsign_
from the TH to Aten (CPU)smooth_l1_loss
from the TH to Aten (CPU) #24762 Migratesmooth_l1_loss
from the TH to Aten (CPU)smooth_l1_loss_backward
from the TH to Aten (CPU) #24763 Migratesmooth_l1_loss_backward
from the TH to Aten (CPU)digamma
anddigamma_
from the TH to Aten (CPU) #24690 Migratedigamma
anddigamma_
from the TH to Aten (CPU)erfinv
anderfinv_
from the TH to Aten (CPU) #24698 Migrateerfinv
anderfinv_
from the TH to Aten (CPU)fill_
from the TH to Aten (CPU) #24700 Migratefill_
from the TH to Aten (CPU)clamp
andclamp_
from the TH to Aten (CPU) #24686 Migrateclamp
andclamp_
from the TH to Aten (CPU)trunc
andtrunc_
from the TH to Aten (CUDA) #24650 Migratetrunc
andtrunc_
from the TH to Aten (CUDA)unfold
from the TH to Aten (CUDA) #24651 Migrateunfold
from the TH to Aten (CUDA)sqrt
andsqrt_
from the TH to Aten (CUDA) #24638 Migratesqrt
andsqrt_
from the TH to Aten (CUDA)sign
andsign_
from the TH to Aten (CUDA) #24626 Migratesign
andsign_
from the TH to Aten (CUDA)sin
andsin_
from the TH to Aten (CUDA) #24627 Migratesin
andsin_
from the TH to Aten (CUDA)sinh
andsinh_
from the TH to Aten (CUDA) #24628 Migratesinh
andsinh_
from the TH to Aten (CUDA)smooth_l1_loss
from the TH to Aten (CUDA) #24629 Migratesmooth_l1_loss
from the TH to Aten (CUDA)smooth_l1_loss_backward
from the TH to Aten (CUDA) #24630 Migratesmooth_l1_loss_backward
from the TH to Aten (CUDA)sigmoid
andsigmoid_
from the TH to Aten (CUDA) #24624 Migratesigmoid
andsigmoid_
from the TH to Aten (CUDA)zero_
from the TH to Aten (CUDA) #24653 Migratezero_
from the TH to Aten (CUDA)abs
andabs_
from the TH to Aten (CUDA) #24531 Migrateabs
andabs_
from the TH to Aten (CUDA)acos
andacos_
from the TH to Aten (CUDA) #24532 Migrateacos
andacos_
from the TH to Aten (CUDA)_addmm
and_addmm_
from the TH to Aten (CUDA) #24518 Migrate_addmm
and_addmm_
from the TH to Aten (CUDA)asin
andasin_
from the TH to Aten (CUDA) #24537 Migrateasin
andasin_
from the TH to Aten (CUDA)ceil
andceil_
from the TH to Aten (CUDA) #24542 Migrateceil
andceil_
from the TH to Aten (CUDA)digamma
anddigamma_
from the TH to Aten (CUDA) #24550 Migratedigamma
anddigamma_
from the TH to Aten (CUDA)frac
andfrac_
from the TH to Aten (CUDA) #24566 Migratefrac
andfrac_
from the TH to Aten (CUDA)expm1
andexpm1_
from the TH to Aten (CUDA) #24562 Migrateexpm1
andexpm1_
from the TH to Aten (CUDA)fill_
from the TH to Aten (CUDA) #24563 Migratefill_
from the TH to Aten (CUDA)floor
andfloor_
from the TH to Aten (CUDA) #24564 Migratefloor
andfloor_
from the TH to Aten (CUDA)erfinv
anderfinv_
from the TH to Aten (CUDA) #24560 Migrateerfinv
anderfinv_
from the TH to Aten (CUDA)is_set_to
from the TH to Aten (CUDA) #24579 Migrateis_set_to
from the TH to Aten (CUDA)l1_loss
from the TH to Aten (CUDA) #24580 Migratel1_loss
from the TH to Aten (CUDA)l1_loss_backward
from the TH to Aten (CUDA) #24581 Migratel1_loss_backward
from the TH to Aten (CUDA)lgamma
andlgamma_
from the TH to Aten (CUDA) #24585 Migratelgamma
andlgamma_
from the TH to Aten (CUDA)log
andlog_
from the TH to Aten (CUDA) #24586 Migratelog
andlog_
from the TH to Aten (CUDA)log10
andlog10_
from the TH to Aten (CUDA) #24587 Migratelog10
andlog10_
from the TH to Aten (CUDA)log1p
andlog1p_
from the TH to Aten (CUDA) #24588 Migratelog1p
andlog1p_
from the TH to Aten (CUDA)log2
andlog2_
from the TH to Aten (CUDA) #24589 Migratelog2
andlog2_
from the TH to Aten (CUDA)mse_loss
from the TH to Aten (CUDA) #24598 Migratemse_loss
from the TH to Aten (CUDA)mse_loss_backward
from the TH to Aten (CUDA) #24599 Migratemse_loss_backward
from the TH to Aten (CUDA)multinomial
from the TH to Aten (CUDA) #24604 Migratemultinomial
from the TH to Aten (CUDA)polygamma
andpolygamma_
from the TH to Aten (CUDA) #24612 Migratepolygamma
andpolygamma_
from the TH to Aten (CUDA)pow
andpow_
from the TH to Aten (CUDA) #24613 Migratepow
andpow_
from the TH to Aten (CUDA)round
andround_
from the TH to Aten (CUDA) #24617 Migrateround
andround_
from the TH to Aten (CUDA)rsqrt
andrsqrt_
from the TH to Aten (CUDA) #24620 Migratersqrt
andrsqrt_
from the TH to Aten (CUDA)The text was updated successfully, but these errors were encountered: