mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-15 21:00:47 +00:00

PyExtreme e1d13f4f8b C++ API parity: NLLLoss & CrossEntropyLoss (#29812 )

Summary:
Hi yf225 , I have added **NLLLoss and CrossEntropyLoss.**
```

Also, while using log_softmax in cross_entropy_loss, I am getting an error
../caffe2/../torch/csrc/api/include/torch/nn/functional/loss.h:537:63: error: no matching function for call to  log_softmax(const at::Tensor&)’
     const Tensor& log_softmax_input = torch::log_softmax(input);

aten/src/ATen/Functions.h:5551:22: note: candidate: at::Tensor at::log_softmax(const at::Tensor&, int64_t, c10::optional<c10::ScalarType>)
 static inline Tensor log_softmax(const Tensor & self, int64_t dim, c10::optional<ScalarType> dtype) {
                      ^~~~~~~~~~~
aten/src/ATen/Functions.h:5551:22: note:   candidate expects 3 arguments, 1 provided
```

I think the other two parameters should be optional as in python frontend(shown in documentation here at https://pytorch.org/docs/stable/nn.functional.html#torch.nn.functional.log_softmax ). Rest, there were no errors in build and tests have passed
Pull Request resolved: https://github.com/pytorch/pytorch/pull/29812

Differential Revision: D18548249

Pulled By: yf225

fbshipit-source-id: 2ab350abd2a6f498d4dba2345f51ad87471f3038

2019-11-16 10:49:09 -08:00

4 KiB

Raw Blame History

Python / C++ API parity tracker

torch.nn

API	Implementation Parity	Doc Parity
torch.nn.Sequential	Yes	No
torch.nn.ModuleList	Yes	No
torch.nn.ModuleDict	No	No
torch.nn.ParameterList	No	No
torch.nn.ParameterDict	No	No
torch.nn.Conv1d	Yes	No
torch.nn.Conv2d	Yes	No
torch.nn.Conv3d	Yes	No
torch.nn.ConvTranspose1d	No	No
torch.nn.ConvTranspose2d	No	No
torch.nn.ConvTranspose3d	No	No
torch.nn.Unfold	Yes	No
torch.nn.Fold	Yes	No
torch.nn.MaxPool1d	Yes	No
torch.nn.MaxPool2d	Yes	No
torch.nn.MaxPool3d	Yes	No
torch.nn.MaxUnpool1d	Yes	No
torch.nn.MaxUnpool2d	Yes	No
torch.nn.MaxUnpool3d	Yes	No
torch.nn.AvgPool1d	Yes	No
torch.nn.AvgPool2d	Yes	No
torch.nn.AvgPool3d	Yes	No
torch.nn.FractionalMaxPool2d	No	No
torch.nn.LPPool1d	Yes	No
torch.nn.LPPool2d	Yes	No
torch.nn.AdaptiveMaxPool1d	Yes	No
torch.nn.AdaptiveMaxPool2d	Yes	No
torch.nn.AdaptiveMaxPool3d	Yes	No
torch.nn.AdaptiveAvgPool1d	Yes	No
torch.nn.AdaptiveAvgPool2d	Yes	No
torch.nn.AdaptiveAvgPool3d	Yes	No
torch.nn.ReflectionPad1d	Yes	No
torch.nn.ReflectionPad2d	Yes	No
torch.nn.ReplicationPad1d	Yes	No
torch.nn.ReplicationPad2d	Yes	No
torch.nn.ReplicationPad3d	Yes	No
torch.nn.ZeroPad2d	Yes	No
torch.nn.ConstantPad1d	Yes	No
torch.nn.ConstantPad2d	Yes	No
torch.nn.ConstantPad3d	Yes	No
torch.nn.ELU	Yes	No
torch.nn.Hardshrink	Yes	No
torch.nn.Hardtanh	Yes	No
torch.nn.LeakyReLU	Yes	No
torch.nn.LogSigmoid	Yes	No
torch.nn.MultiheadAttention	No	No
torch.nn.PReLU	Yes	No
torch.nn.ReLU	Yes	No
torch.nn.ReLU6	Yes	No
torch.nn.RReLU	Yes	No
torch.nn.SELU	Yes	No
torch.nn.CELU	Yes	No
torch.nn.GELU	Yes	No
torch.nn.Sigmoid	Yes	No
torch.nn.Softplus	Yes	No
torch.nn.Softshrink	Yes	No
torch.nn.Softsign	Yes	No
torch.nn.Tanh	Yes	No
torch.nn.Tanhshrink	Yes	No
torch.nn.Threshold	Yes	No
torch.nn.GLU	No	No
torch.nn.Softmin	Yes	No
torch.nn.Softmax	Yes	No
torch.nn.Softmax2d	Yes	No
torch.nn.LogSoftmax	Yes	No
torch.nn.AdaptiveLogSoftmaxWithLoss	No	No
torch.nn.BatchNorm1d	Yes	No
torch.nn.BatchNorm2d	Yes	No
torch.nn.BatchNorm3d	Yes	No
torch.nn.GroupNorm	No	No
torch.nn.SyncBatchNorm	No	No
torch.nn.InstanceNorm1d	No	No
torch.nn.InstanceNorm2d	No	No
torch.nn.InstanceNorm3d	No	No
torch.nn.LayerNorm	Yes	No
torch.nn.LocalResponseNorm	Yes	No
torch.nn.CrossMapLRN2d	Yes	No
torch.nn.RNN	No	No
torch.nn.LSTM	No	No
torch.nn.GRU	No	No
torch.nn.RNNCell	No	No
torch.nn.LSTMCell	No	No
torch.nn.GRUCell	No	No
torch.nn.Transformer	No	No
torch.nn.TransformerEncoder	No	No
torch.nn.TransformerDecoder	No	No
torch.nn.TransformerEncoderLayer	No	No
torch.nn.TransformerDecoderLayer	No	No
torch.nn.Identity	Yes	No
torch.nn.Linear	Yes	No
torch.nn.Bilinear	Yes	No
torch.nn.Flatten	Yes	No
torch.nn.Dropout	Yes	No
torch.nn.Dropout2d	Yes	No
torch.nn.Dropout3d	Yes	No
torch.nn.AlphaDropout	No	No
torch.nn.Embedding	Yes	No
torch.nn.EmbeddingBag	Yes	No
torch.nn.CosineSimilarity	Yes	No
torch.nn.PairwiseDistance	Yes	No
torch.nn.L1Loss	Yes	No
torch.nn.MSELoss	Yes	No
torch.nn.CrossEntropyLoss	Yes	No
torch.nn.CTCLoss	Yes	No
torch.nn.NLLLoss	Yes	No
torch.nn.PoissonNLLLoss	Yes	No
torch.nn.KLDivLoss	Yes	No
torch.nn.BCELoss	Yes	No
torch.nn.BCEWithLogitsLoss	No	No
torch.nn.MarginRankingLoss	Yes	No
torch.nn.HingeEmbeddingLoss	Yes	No
torch.nn.MultiLabelMarginLoss	Yes	No
torch.nn.SmoothL1Loss	Yes	No
torch.nn.SoftMarginLoss	Yes	No
torch.nn.MultiLabelSoftMarginLoss	Yes	No
torch.nn.CosineEmbeddingLoss	Yes	No
torch.nn.MultiMarginLoss	Yes	No
torch.nn.TripletMarginLoss	Yes	No
torch.nn.PixelShuffle	Yes	No
torch.nn.Upsample	Yes	No
torch.nn.DataParallel	No	No
torch.nn.parallel.DistributedDataParallel	No	No
torch.nn.utils.clip_grad_norm_	Yes	No
torch.nn.utils.clip_grad_value_	No	No
torch.nn.utils.parameters_to_vector	No	No
torch.nn.utils.vector_to_parameters	No	No
torch.nn.utils.weight_norm	No	No
torch.nn.utils.remove_weight_norm	No	No
torch.nn.utils.spectral_norm	No	No
torch.nn.utils.remove_spectral_norm	No	No
torch.nn.utils.rnn.PackedSequence	No	No
torch.nn.utils.rnn.pack_padded_sequence	No	No
torch.nn.utils.rnn.pad_packed_sequence	No	No
torch.nn.utils.rnn.pad_sequence	No	No
torch.nn.utils.rnn.pack_sequence	No	No
torch.nn.FractionalMaxPool3d	No	No
torch.nn.SampleModule	Yes	Yes

4 KiB Raw Blame History

Python / C++ API parity tracker

torch.nn

4 KiB

Raw Blame History