platform/upstream/pytorch - Domain: Machine Learning / ML Framework; Licenses: BSD-3-Clause;

Age	Commit message (Expand)	Author	Files	Lines
2019-04-23	Use `fbgemm` for quantize/dequantize ops (#19500)	Zafar Takhirov	1	-12/+51
2019-04-23	Specify to use Float16UniformFill if necessary in sparse lookup layer (#18499)	Jiyan Yang	1	-2/+7
2019-04-23	Fix the Division by Zero Bug of CosineAnnealingLR (#19180)	Chandler Zuo	2	-3/+13
2019-04-23	Fix the documentation for BCEWithLogitsLoss (#17218, #16804) (#19212)	Vadim Velicodnii	1	-6/+19
2019-04-23	fix the docstring of `RandomSampler` (#19113)	crcrpar	1	-6/+8
2019-04-23	Avoid (future) cusparse name collision (#19591)	mruberry	1	-2/+2
2019-04-23	Add docs and test guaranteeing indices from torch.nonzero ordered C-style (#1...	jhultman	2	-30/+55
2019-04-23	Remove unnecessary printing from tests	Tongzhou Wang	2	-9/+18
2019-04-23	Fix lr_scheduler's last_epoch value at the time of initialization (BC BREAKIN...	Bado Lee	2	-6/+11
2019-04-23	Removes variable which is assigned but not used (#19194)	SebFar	1	-1/+0
2019-04-23	add torch.cuda.synchronize(device=None) (#19573)	SsnL	3	-16/+39
2019-04-23	Port adaptive_max_pool2d() to ATen (#19409)	Stefan Krah	13	-699/+847
2019-04-23	Fix math formatting of PairwiseDistance and CosineSimilarity docs and fix mat...	zhiqiang	2	-10/+14
2019-04-22	Revert D15039713: [pytorch][PR] add torch.tensor requires grad	Michael Suo	7	-432/+266
2019-04-22	Bugfix for fusion device check (#19594)	James Reed	1	-1/+1
2019-04-22	add torch.tensor requires grad (#19445)	Elias Ellison	7	-266/+432
2019-04-22	Add onnx support for _unique2 operator	Vitaly Fedyunin	1	-0/+7
2019-04-22	Automatic update of fbcode/onnx to 0e8d2bc5e51455c70ef790b9f65aa632ed9bc8a7 (...	Lu Fang	2	-0/+2
2019-04-22	Don't create FusionGroups for known-CPU producer values (#19342)	James Reed	4	-21/+35
2019-04-22	Explicitly define supported types (#19516)	Sebastian Messmer	4	-14/+107
2019-04-22	IRParser: optionally create name->value map of the parsed IR. (#19551)	Mikhail Zolotukhin	3	-7/+43
2019-04-22	Profiling : Adding Profile Op to provide storage for profiling lambdas	Nikolay Korovaiko	5	-23/+40
2019-04-22	Step 5: remove _unique_dim in favor of unique_dim (#18654)	Xiang Gao	4	-28/+1
2019-04-22	Add back option to not adjust output batch size (#19442)	Yinghai Lu	6	-12/+38
2019-04-22	Add debug logic to c2_ref_test and its helpers (#19359)	Michael Antonov	3	-7/+27
2019-04-22	fix variable shadowing issus (#19567)	Dehua Cheng	1	-2/+2
2019-04-22	Add manual_seed in script (#19510)	Elias Ellison	4	-0/+21
2019-04-22	Automatic update of fbcode/onnx to 83dd62659fc07d5b7fa93b5d1c1879f93509c7db (...	Lu Fang	2	-0/+5
2019-04-22	Get rid of unnecessary matches_jit_signature: True specifications. (#19549)	Gregory Chanan	1	-2/+0
2019-04-22	Rename potri to cholesky_inverse (#19498)	vishwakftw	15	-100/+131
2019-04-21	Add assertion to make sure init op is always fp16 compatible in fp16 training	Jiyan Yang	1	-9/+20
2019-04-21	Generate only one Type class per backend (#19295)	Roy Li	39	-298/+162
2019-04-21	Make complex its own backend (#19275)	Roy Li	6	-8/+64
2019-04-21	Add ScalarType argument to Type::options() (#19270)	Roy Li	18	-73/+76
2019-04-21	Generate cases for all ScalarTypes in Type functions that call to TH (#19230)	Roy Li	8	-269/+331
2019-04-21	Fix clang-format. (#19550)	Mikhail Zolotukhin	1	-7/+8
2019-04-21	Fix some typos in jit README	Shen Li	1	-18/+18
2019-04-21	Match JIT signature with triu_indices / tril_indices. (#19484)	Gregory Chanan	2	-4/+18
2019-04-21	Make one_hot non-differentiable. (#19524)	Gregory Chanan	1	-0/+3
2019-04-21	Remove 'BoolTensor', 'IndexTensor' from frontend specifications. (#19523)	Gregory Chanan	3	-12/+2
2019-04-21	Have _embedding_bag_dense_backward match JIT signature. (#19522)	Gregory Chanan	2	-2/+8
2019-04-21	Have embedding_dense_backward match JIT signature. (#19521)	Gregory Chanan	2	-2/+2
2019-04-21	Update mkldnn-bridge to fix crash issue in DNNLOWP dequantize op (#19159)	Gu, Jinghui	1	-0/+0
2019-04-21	Hook up non_differentiability in derivatives.yaml when no autograd function i...	Gregory Chanan	4	-3/+7
2019-04-21	Move non_differentiable_arg_names from autograd functions to differentiabilit...	Gregory Chanan	2	-6/+9
2019-04-20	Move cuFFT plan cache note outside Best Practices (#19538)	Tongzhou Wang	1	-28/+28
2019-04-20	Revert D14689639: [pytorch] Allow passing lists as trace inputs.	Michael Suo	3	-108/+11
2019-04-20	Improve optimizations for DNNLOWP support on MKL-DNN (#18843)	Gu, Jinghui	2	-195/+740
2019-04-19	Make Observer class as template Quant class for QuantConfig (#19418)	Nishant Pandit	1	-135/+124
2019-04-19	Support compilation on gcc-7.4.0 (#19470)	Sam Leeman-Munk	2	-2/+3