-
-
Save cpuhrsch/f400a6614c0f67dda601aad98a84767c to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| diff --git a/build/aten/src/ATen/Declarations.yaml b/build/aten/src/ATen/Declarations.yaml | |
| index 7ca5b4659..249b9a9b5 100644 | |
| --- a/build/aten/src/ATen/Declarations.yaml | |
| +++ b/build/aten/src/ATen/Declarations.yaml | |
| @@ -25896,8 +25896,8 @@ | |
| with_gil: false | |
| deprecated: false | |
| - name: embedding_backward | |
| - matches_jit_signature: false | |
| - schema_string: aten::embedding_backward(Tensor grad, IndexTensor indices, int num_weights, | |
| + matches_jit_signature: true | |
| + schema_string: aten::embedding_backward(Tensor grad, Tensor indices, int num_weights, | |
| int padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| @@ -25907,7 +25907,7 @@ | |
| name: grad | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| @@ -26037,9 +26037,9 @@ | |
| with_gil: false | |
| deprecated: false | |
| - name: embedding_sparse_backward | |
| - matches_jit_signature: false | |
| - schema_string: aten::embedding_sparse_backward(Tensor grad, IndexTensor indices, | |
| - int num_weights, int padding_idx, bool scale_grad_by_freq) -> Tensor | |
| + matches_jit_signature: true | |
| + schema_string: aten::embedding_sparse_backward(Tensor grad, Tensor indices, int | |
| + num_weights, int padding_idx, bool scale_grad_by_freq) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| @@ -26048,7 +26048,7 @@ | |
| name: grad | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| @@ -26085,9 +26085,9 @@ | |
| deprecated: false | |
| - name: embedding_bag | |
| matches_jit_signature: false | |
| - schema_string: aten::embedding_bag(Tensor weight, IndexTensor indices, IndexTensor | |
| - offsets, bool scale_grad_by_freq=False, int mode=0, bool sparse=False) -> (Tensor, | |
| - Tensor, Tensor, Tensor) | |
| + schema_string: aten::embedding_bag(Tensor weight, Tensor indices, Tensor offsets, | |
| + bool scale_grad_by_freq=False, int mode=0, bool sparse=False) -> (Tensor, Tensor, | |
| + Tensor, Tensor) | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| @@ -26096,12 +26096,12 @@ | |
| name: weight | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: offsets | |
| type: const Tensor & | |
| @@ -26149,10 +26149,10 @@ | |
| with_gil: false | |
| deprecated: false | |
| - name: _embedding_bag | |
| - matches_jit_signature: false | |
| - schema_string: aten::_embedding_bag(Tensor weight, IndexTensor indices, IndexTensor | |
| - offsets, bool scale_grad_by_freq=False, int mode=0, bool sparse=False) -> (Tensor, | |
| - Tensor, Tensor, Tensor) | |
| + matches_jit_signature: true | |
| + schema_string: aten::_embedding_bag(Tensor weight, Tensor indices, Tensor offsets, | |
| + bool scale_grad_by_freq=False, int mode=0, bool sparse=False) -> (Tensor, Tensor, | |
| + Tensor, Tensor) | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| @@ -26161,12 +26161,12 @@ | |
| name: weight | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: offsets | |
| type: const Tensor & | |
| @@ -26214,10 +26214,10 @@ | |
| with_gil: false | |
| deprecated: false | |
| - name: _embedding_bag_backward | |
| - matches_jit_signature: false | |
| - schema_string: aten::_embedding_bag_backward(Tensor grad, IndexTensor indices, IndexTensor | |
| - offsets, IndexTensor offset2bag, IndexTensor bag_size, IndexTensor maximum_indices, | |
| - int num_weights, bool scale_grad_by_freq, int mode, bool sparse) -> Tensor | |
| + matches_jit_signature: true | |
| + schema_string: aten::_embedding_bag_backward(Tensor grad, Tensor indices, Tensor | |
| + offsets, Tensor offset2bag, Tensor bag_size, Tensor maximum_indices, int num_weights, | |
| + bool scale_grad_by_freq, int mode, bool sparse) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| @@ -26226,27 +26226,27 @@ | |
| name: grad | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: offsets | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: offset2bag | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: bag_size | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: maximum_indices | |
| type: const Tensor & | |
| @@ -26287,10 +26287,10 @@ | |
| with_gil: false | |
| deprecated: false | |
| - name: _embedding_bag_sparse_backward | |
| - matches_jit_signature: false | |
| - schema_string: aten::_embedding_bag_sparse_backward(Tensor grad, IndexTensor indices, | |
| - IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, int num_weights, | |
| - bool scale_grad_by_freq, int mode) -> Tensor | |
| + matches_jit_signature: true | |
| + schema_string: aten::_embedding_bag_sparse_backward(Tensor grad, Tensor indices, | |
| + Tensor offsets, Tensor offset2bag, Tensor bag_size, int num_weights, bool scale_grad_by_freq, | |
| + int mode) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| @@ -26299,22 +26299,22 @@ | |
| name: grad | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: offsets | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: offset2bag | |
| type: const Tensor & | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: bag_size | |
| type: const Tensor & | |
| @@ -39554,11 +39554,11 @@ | |
| deprecated: false | |
| - name: where | |
| matches_jit_signature: false | |
| - schema_string: aten::where(BoolTensor condition, Tensor self, Tensor other) -> Tensor | |
| + schema_string: aten::where(Tensor condition, Tensor self, Tensor other) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| - dynamic_type: BoolTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: condition | |
| type: const Tensor & | |
| @@ -41714,12 +41714,12 @@ | |
| deprecated: false | |
| - name: sparse_coo_tensor | |
| matches_jit_signature: false | |
| - schema_string: aten::sparse_coo_tensor(IndexTensor indices, Tensor values, *, TensorOptions | |
| + schema_string: aten::sparse_coo_tensor(Tensor indices, Tensor values, *, TensorOptions | |
| options=[]) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| @@ -41753,12 +41753,12 @@ | |
| deprecated: false | |
| - name: sparse_coo_tensor | |
| matches_jit_signature: false | |
| - schema_string: aten::sparse_coo_tensor(IndexTensor indices, Tensor values, int[] | |
| - size, *, TensorOptions options=[]) -> Tensor | |
| + schema_string: aten::sparse_coo_tensor(Tensor indices, Tensor values, int[] size, | |
| + *, TensorOptions options=[]) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| @@ -41797,12 +41797,12 @@ | |
| deprecated: false | |
| - name: _sparse_coo_tensor_unsafe | |
| matches_jit_signature: false | |
| - schema_string: aten::_sparse_coo_tensor_unsafe(IndexTensor indices, Tensor values, | |
| - int[] size, *, TensorOptions options=[]) -> Tensor | |
| + schema_string: aten::_sparse_coo_tensor_unsafe(Tensor indices, Tensor values, int[] | |
| + size, *, TensorOptions options=[]) -> Tensor | |
| method_prefix_derived: '' | |
| arguments: | |
| - annotation: null | |
| - dynamic_type: IndexTensor | |
| + dynamic_type: Tensor | |
| is_nullable: false | |
| name: indices | |
| type: const Tensor & | |
| diff --git a/build/aten/src/ATen/MSNPUType.cpp b/build/aten/src/ATen/MSNPUType.cpp | |
| index 35f9d639c..ff75cbf78 100644 | |
| --- a/build/aten/src/ATen/MSNPUType.cpp | |
| +++ b/build/aten/src/ATen/MSNPUType.cpp | |
| @@ -2107,7 +2107,7 @@ Tensor MSNPUType::embedding(const Tensor & weight, const Tensor & indices, int64 | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, bool, bool)>("embedding(Tensor weight, Tensor indices, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")(weight, indices, padding_idx, scale_grad_by_freq, sparse); | |
| } | |
| Tensor MSNPUType::embedding_backward(const Tensor & grad, const Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool, bool)>("embedding_backward(Tensor grad, IndexTensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool, bool)>("embedding_backward(Tensor grad, Tensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse); | |
| } | |
| Tensor MSNPUType::embedding_dense_backward(const Tensor & grad, const Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) const { | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool)>("embedding_dense_backward(Tensor grad, IndexTensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq); | |
| @@ -2116,19 +2116,19 @@ Tensor & MSNPUType::embedding_renorm_(Tensor & self, const Tensor & indices, dou | |
| return MSNPUTypeDispatch::get_function<Tensor & (*)(Tensor &, const Tensor &, double, double)>("embedding_renorm_(Tensor self, Tensor indices, double max_norm, double norm_type) -> Tensor")(self, indices, max_norm, norm_type); | |
| } | |
| Tensor MSNPUType::embedding_sparse_backward(const Tensor & grad, const Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool)>("embedding_sparse_backward(Tensor grad, IndexTensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool)>("embedding_sparse_backward(Tensor grad, Tensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq); | |
| } | |
| std::tuple<Tensor,Tensor,Tensor,Tensor> MSNPUType::embedding_bag(const Tensor & weight, const Tensor & indices, const Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) const { | |
| - return MSNPUTypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("embedding_bag(Tensor weight, IndexTensor indices, IndexTensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| + return MSNPUTypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("embedding_bag(Tensor weight, Tensor indices, Tensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| } | |
| std::tuple<Tensor,Tensor,Tensor,Tensor> MSNPUType::_embedding_bag(const Tensor & weight, const Tensor & indices, const Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) const { | |
| - return MSNPUTypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("_embedding_bag(Tensor weight, IndexTensor indices, IndexTensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| + return MSNPUTypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("_embedding_bag(Tensor weight, Tensor indices, Tensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| } | |
| Tensor MSNPUType::_embedding_bag_backward(const Tensor & grad, const Tensor & indices, const Tensor & offsets, const Tensor & offset2bag, const Tensor & bag_size, const Tensor & maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, bool sparse) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t, bool)>("_embedding_bag_backward(Tensor grad, IndexTensor indices, IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, IndexTensor maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, bool sparse) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, maximum_indices, num_weights, scale_grad_by_freq, mode, sparse); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t, bool)>("_embedding_bag_backward(Tensor grad, Tensor indices, Tensor offsets, Tensor offset2bag, Tensor bag_size, Tensor maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, bool sparse) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, maximum_indices, num_weights, scale_grad_by_freq, mode, sparse); | |
| } | |
| Tensor MSNPUType::_embedding_bag_sparse_backward(const Tensor & grad, const Tensor & indices, const Tensor & offsets, const Tensor & offset2bag, const Tensor & bag_size, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t)>("_embedding_bag_sparse_backward(Tensor grad, IndexTensor indices, IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, num_weights, scale_grad_by_freq, mode); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t)>("_embedding_bag_sparse_backward(Tensor grad, Tensor indices, Tensor offsets, Tensor offset2bag, Tensor bag_size, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, num_weights, scale_grad_by_freq, mode); | |
| } | |
| Tensor MSNPUType::_embedding_bag_dense_backward(const Tensor & grad, const Tensor & indices, const Tensor & offsets, const Tensor & offset2bag, const Tensor & bag_size, const Tensor & maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) const { | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t)>("_embedding_bag_dense_backward(Tensor grad, IndexTensor indices, IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, IndexTensor maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, maximum_indices, num_weights, scale_grad_by_freq, mode); | |
| @@ -3151,7 +3151,7 @@ Tensor MSNPUType::view_as(const Tensor & self, const Tensor & other) const { | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &)>("view_as(Tensor self, Tensor other) -> Tensor")(self, other); | |
| } | |
| Tensor MSNPUType::where(const Tensor & condition, const Tensor & self, const Tensor & other) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &)>("where(BoolTensor condition, Tensor self, Tensor other) -> Tensor")(condition, self, other); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &)>("where(Tensor condition, Tensor self, Tensor other) -> Tensor")(condition, self, other); | |
| } | |
| Tensor MSNPUType::_s_where(const Tensor & condition, const Tensor & self, const Tensor & other) const { | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &)>("_s_where(Tensor condition, Tensor self, Tensor other) -> Tensor")(condition, self, other); | |
| @@ -3319,13 +3319,13 @@ Tensor MSNPUType::sparse_coo_tensor(IntArrayRef size, const TensorOptions & opti | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(IntArrayRef, const TensorOptions &)>("sparse_coo_tensor(IntArrayRef size, TensorOptions options) -> Tensor")(size, options); | |
| } | |
| Tensor MSNPUType::sparse_coo_tensor(const Tensor & indices, const Tensor & values, const TensorOptions & options) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const TensorOptions &)>("sparse_coo_tensor(IndexTensor indices, Tensor values, TensorOptions options) -> Tensor")(indices, values, options); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const TensorOptions &)>("sparse_coo_tensor(Tensor indices, Tensor values, TensorOptions options) -> Tensor")(indices, values, options); | |
| } | |
| Tensor MSNPUType::sparse_coo_tensor(const Tensor & indices, const Tensor & values, IntArrayRef size, const TensorOptions & options) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("sparse_coo_tensor(IndexTensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("sparse_coo_tensor(Tensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| } | |
| Tensor MSNPUType::_sparse_coo_tensor_unsafe(const Tensor & indices, const Tensor & values, IntArrayRef size, const TensorOptions & options) const { | |
| - return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("_sparse_coo_tensor_unsafe(IndexTensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| + return MSNPUTypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("_sparse_coo_tensor_unsafe(Tensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| } | |
| Tensor MSNPUType::_sparse_coo_tensor_with_dims(int64_t sparse_dim, int64_t dense_dim, IntArrayRef size, const TensorOptions & options) const { | |
| return MSNPUTypeDispatch::get_function<Tensor (*)(int64_t, int64_t, IntArrayRef, const TensorOptions &)>("_sparse_coo_tensor_with_dims(int64_t sparse_dim, int64_t dense_dim, IntArrayRef size, TensorOptions options) -> Tensor")(sparse_dim, dense_dim, size, options); | |
| diff --git a/build/aten/src/ATen/XLAType.cpp b/build/aten/src/ATen/XLAType.cpp | |
| index f4e919eee..cbd587522 100644 | |
| --- a/build/aten/src/ATen/XLAType.cpp | |
| +++ b/build/aten/src/ATen/XLAType.cpp | |
| @@ -2107,7 +2107,7 @@ Tensor XLAType::embedding(const Tensor & weight, const Tensor & indices, int64_t | |
| return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, bool, bool)>("embedding(Tensor weight, Tensor indices, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")(weight, indices, padding_idx, scale_grad_by_freq, sparse); | |
| } | |
| Tensor XLAType::embedding_backward(const Tensor & grad, const Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool, bool)>("embedding_backward(Tensor grad, IndexTensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool, bool)>("embedding_backward(Tensor grad, Tensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse); | |
| } | |
| Tensor XLAType::embedding_dense_backward(const Tensor & grad, const Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) const { | |
| return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool)>("embedding_dense_backward(Tensor grad, IndexTensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq); | |
| @@ -2116,19 +2116,19 @@ Tensor & XLAType::embedding_renorm_(Tensor & self, const Tensor & indices, doubl | |
| return XLATypeDispatch::get_function<Tensor & (*)(Tensor &, const Tensor &, double, double)>("embedding_renorm_(Tensor self, Tensor indices, double max_norm, double norm_type) -> Tensor")(self, indices, max_norm, norm_type); | |
| } | |
| Tensor XLAType::embedding_sparse_backward(const Tensor & grad, const Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool)>("embedding_sparse_backward(Tensor grad, IndexTensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, int64_t, int64_t, bool)>("embedding_sparse_backward(Tensor grad, Tensor indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq) -> Tensor")(grad, indices, num_weights, padding_idx, scale_grad_by_freq); | |
| } | |
| std::tuple<Tensor,Tensor,Tensor,Tensor> XLAType::embedding_bag(const Tensor & weight, const Tensor & indices, const Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) const { | |
| - return XLATypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("embedding_bag(Tensor weight, IndexTensor indices, IndexTensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| + return XLATypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("embedding_bag(Tensor weight, Tensor indices, Tensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| } | |
| std::tuple<Tensor,Tensor,Tensor,Tensor> XLAType::_embedding_bag(const Tensor & weight, const Tensor & indices, const Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) const { | |
| - return XLATypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("_embedding_bag(Tensor weight, IndexTensor indices, IndexTensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| + return XLATypeDispatch::get_function<std::tuple<Tensor,Tensor,Tensor,Tensor> (*)(const Tensor &, const Tensor &, const Tensor &, bool, int64_t, bool)>("_embedding_bag(Tensor weight, Tensor indices, Tensor offsets, bool scale_grad_by_freq, int64_t mode, bool sparse) -> std::tuple<Tensor,Tensor,Tensor,Tensor>")(weight, indices, offsets, scale_grad_by_freq, mode, sparse); | |
| } | |
| Tensor XLAType::_embedding_bag_backward(const Tensor & grad, const Tensor & indices, const Tensor & offsets, const Tensor & offset2bag, const Tensor & bag_size, const Tensor & maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, bool sparse) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t, bool)>("_embedding_bag_backward(Tensor grad, IndexTensor indices, IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, IndexTensor maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, bool sparse) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, maximum_indices, num_weights, scale_grad_by_freq, mode, sparse); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t, bool)>("_embedding_bag_backward(Tensor grad, Tensor indices, Tensor offsets, Tensor offset2bag, Tensor bag_size, Tensor maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, bool sparse) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, maximum_indices, num_weights, scale_grad_by_freq, mode, sparse); | |
| } | |
| Tensor XLAType::_embedding_bag_sparse_backward(const Tensor & grad, const Tensor & indices, const Tensor & offsets, const Tensor & offset2bag, const Tensor & bag_size, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t)>("_embedding_bag_sparse_backward(Tensor grad, IndexTensor indices, IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, num_weights, scale_grad_by_freq, mode); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t)>("_embedding_bag_sparse_backward(Tensor grad, Tensor indices, Tensor offsets, Tensor offset2bag, Tensor bag_size, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, num_weights, scale_grad_by_freq, mode); | |
| } | |
| Tensor XLAType::_embedding_bag_dense_backward(const Tensor & grad, const Tensor & indices, const Tensor & offsets, const Tensor & offset2bag, const Tensor & bag_size, const Tensor & maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) const { | |
| return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, const Tensor &, int64_t, bool, int64_t)>("_embedding_bag_dense_backward(Tensor grad, IndexTensor indices, IndexTensor offsets, IndexTensor offset2bag, IndexTensor bag_size, IndexTensor maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode) -> Tensor")(grad, indices, offsets, offset2bag, bag_size, maximum_indices, num_weights, scale_grad_by_freq, mode); | |
| @@ -3151,7 +3151,7 @@ Tensor XLAType::view_as(const Tensor & self, const Tensor & other) const { | |
| return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &)>("view_as(Tensor self, Tensor other) -> Tensor")(self, other); | |
| } | |
| Tensor XLAType::where(const Tensor & condition, const Tensor & self, const Tensor & other) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &)>("where(BoolTensor condition, Tensor self, Tensor other) -> Tensor")(condition, self, other); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &)>("where(Tensor condition, Tensor self, Tensor other) -> Tensor")(condition, self, other); | |
| } | |
| Tensor XLAType::_s_where(const Tensor & condition, const Tensor & self, const Tensor & other) const { | |
| return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const Tensor &)>("_s_where(Tensor condition, Tensor self, Tensor other) -> Tensor")(condition, self, other); | |
| @@ -3319,13 +3319,13 @@ Tensor XLAType::sparse_coo_tensor(IntArrayRef size, const TensorOptions & option | |
| return XLATypeDispatch::get_function<Tensor (*)(IntArrayRef, const TensorOptions &)>("sparse_coo_tensor(IntArrayRef size, TensorOptions options) -> Tensor")(size, options); | |
| } | |
| Tensor XLAType::sparse_coo_tensor(const Tensor & indices, const Tensor & values, const TensorOptions & options) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const TensorOptions &)>("sparse_coo_tensor(IndexTensor indices, Tensor values, TensorOptions options) -> Tensor")(indices, values, options); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, const TensorOptions &)>("sparse_coo_tensor(Tensor indices, Tensor values, TensorOptions options) -> Tensor")(indices, values, options); | |
| } | |
| Tensor XLAType::sparse_coo_tensor(const Tensor & indices, const Tensor & values, IntArrayRef size, const TensorOptions & options) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("sparse_coo_tensor(IndexTensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("sparse_coo_tensor(Tensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| } | |
| Tensor XLAType::_sparse_coo_tensor_unsafe(const Tensor & indices, const Tensor & values, IntArrayRef size, const TensorOptions & options) const { | |
| - return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("_sparse_coo_tensor_unsafe(IndexTensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| + return XLATypeDispatch::get_function<Tensor (*)(const Tensor &, const Tensor &, IntArrayRef, const TensorOptions &)>("_sparse_coo_tensor_unsafe(Tensor indices, Tensor values, IntArrayRef size, TensorOptions options) -> Tensor")(indices, values, size, options); | |
| } | |
| Tensor XLAType::_sparse_coo_tensor_with_dims(int64_t sparse_dim, int64_t dense_dim, IntArrayRef size, const TensorOptions & options) const { | |
| return XLATypeDispatch::get_function<Tensor (*)(int64_t, int64_t, IntArrayRef, const TensorOptions &)>("_sparse_coo_tensor_with_dims(int64_t sparse_dim, int64_t dense_dim, IntArrayRef size, TensorOptions options) -> Tensor")(sparse_dim, dense_dim, size, options); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment