summaryrefslogtreecommitdiff
path: root/test/test_nn.py
AgeCommit message (Expand)AuthorFilesLines
2019-04-19Improve embedding_bag add kernel (#19329)James Reed1-8/+11
2019-04-16Revert replicate.py to disallow replicating multi-device modules (#19278)Shen Li1-136/+3
2019-04-16Enable half for CUDA dense EmbeddingBag backward. (#19293)Richard Zou1-11/+30
2019-04-12Enable working ROCm tests (#19169)Johannes M Dieterich1-5/+0
2019-04-11Import MultiheadAttention to PyTorch (#18334)Guanheng Zhang1-0/+187
2019-04-09EmbeddingBag w/ differentiable per_sample_weights (#18957)Richard Zou1-8/+25
2019-04-09EmbeddingBag w/ per_sample_weights CUDA fwd + bwd (#18800)Richard Zou1-2/+18
2019-04-09EmbeddingBag w/ per_sample_weights CPU backward (#18799)Richard Zou1-8/+21
2019-04-09EmbeddingBag CPU forward with per_sample_weights. (#18735)Richard Zou1-30/+155
2019-04-05add launch bounds, enable more tests (#18909)J M Dieterich1-7/+0
2019-04-04return missing keys from load_state_dict (#18668)Sepehr Sameni1-1/+23
2019-04-04Introduce DeprecatedTypeProperties class (#17991)Roy Li1-1/+1
2019-04-03Support replicating multi-GPU modules (#18687)Shen Li1-12/+146
2019-04-03Fix dense Embedding to work with double backward (#9078)kshitij123451-0/+30
2019-03-30Turn on F401: Unused import warning. (#18598)Edward Yang1-5/+3
2019-03-27enable more unit tests (#18537)jithunnair-amd1-3/+0
2019-03-24Added tensor size warning to F.mse_loss() (#18349)mc-robinson1-0/+12
2019-03-22Fix incorrect sparse add behavior when the sparse tensor has non-contiguous v...Will Feng1-0/+19
2019-03-22Using sqrt for better precision in cosine_similarity (#18250)Ailing Zhang1-0/+7
2019-03-21Fix B902 lint error: invalid first argument. (#18181)Edward Yang1-1/+2
2019-03-19fix cosine_similarity (#18168)Ailing Zhang1-0/+6
2019-03-14Fix lint in test_nn.py (#18006)Edward Yang1-10/+10
2019-03-14Report convolution size mismatch (#17436)bhushan1-0/+26
2019-03-05Fix nll_loss crash on cpu where ignore_index is out of bounds (#17328)Soumith Chintala1-1/+7
2019-02-26Fix autograd with buffers requiring grad in DataParallel (#13352)Tongzhou Wang1-0/+23
2019-02-25Fix reduction='none' in poisson_nll_loss (#17358)vishwakftw1-0/+13
2019-02-22Fix DataParallel(cpu_m).cuda() not working by checking at forward (#17363)Tongzhou Wang1-25/+62
2019-02-21fix double backward for half softmax/logsoftmax (#17330)Natalia Gimelshein1-2/+1
2019-02-21remove nn.Upsample deprecation warnings from tests (#17352)Soumith Chintala1-13/+23
2019-02-19Fix cuda softmax backward with empty input (#17259)Tongzhou Wang1-0/+24
2019-02-19at::native batch norm kernel launch config update (#17047)Jie1-0/+7
2019-02-19False alarm about leak in TestNN.test_variable_sequence_cuda (#17242)Sergei Nikolaev1-15/+16
2019-02-15Enforce module device at DataParallel construction time (#17129)Shen Li1-0/+37
2019-02-15Fix AdaptiveLogSoftmaxWithLoss's constructor (#16694)wbydo1-0/+6
2019-02-13Skip test_cudnn_multiple_threads_same_device on ROCm (flaky) (#17061)Junjie Bai1-0/+1
2019-02-12enable more unit tests in test_nn (#16994)Johannes M Dieterich1-1/+0
2019-02-11optionally zero infinite losses in CTCLoss (#16199)Thomas Viehmann1-0/+23
2019-02-09enable unit tests working on ROCm 2.1 (#16871)Johannes M Dieterich1-19/+1
2019-02-01Better bounds checks in ctcloss (#16269)Asher Mancinelli1-0/+29
2019-01-30Allow list and tuples to be passed as output_size to max_unpool1d (#16489)vishwakftw1-0/+6
2019-01-30Fix the flake8 linterLu Fang1-15/+15
2019-01-24Handle non-contiguous inputs with mkldnn convolution. (#16300)Gregory Chanan1-0/+33
2019-01-18add back NNPACK in PyTorch (#15924)Thomas Viehmann1-0/+28
2019-01-18Improve pack_sequence and pack_padded_sequence error message (#16084)Richard Zou1-1/+8
2019-01-17Cleanup gumbel_softmax (#13339)Egil Martinsson1-49/+75
2019-01-17Revert batched pdist, improve existing kernel, add test (#15901)Gregory Chanan1-3/+6
2019-01-08use all_weights instead of _parameters in _flat_weights in rnn (#15766)Natalia Gimelshein1-0/+23
2019-01-04Cudnn Handle Pool 3: At Wit's End (#15668)Michael Carilli1-0/+50
2019-01-03Add mkldnn conv double backward (#15686)Ailing Zhang1-14/+22
2018-12-26Move autograd metadata from VariableImpl to TensorImpl (#13827)Will Feng1-2/+4
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-23 12:09:36 +0000