1 | #pragma once |
2 | |
3 | // @generated by torchgen/gen.py from NativeFunction.h |
4 | |
5 | #include <c10/core/Scalar.h> |
6 | #include <c10/core/Storage.h> |
7 | #include <c10/core/TensorOptions.h> |
8 | #include <c10/util/Deprecated.h> |
9 | #include <c10/util/Optional.h> |
10 | #include <c10/core/QScheme.h> |
11 | #include <ATen/core/Reduction.h> |
12 | #include <ATen/core/Tensor.h> |
13 | #include <tuple> |
14 | #include <vector> |
15 | |
16 | |
17 | namespace at { |
18 | namespace native { |
19 | TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> _native_decoder_only_multi_head_attention_out(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, int64_t embed_dim, int64_t num_head, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, const c10::optional<at::Tensor> & mask, const c10::optional<at::Tensor> & incr_key, const c10::optional<at::Tensor> & incr_value, bool need_weights, bool average_attn_weights, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3); |
20 | TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> native_decoder_only_multi_head_attention(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, int64_t embed_dim, int64_t num_head, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, const c10::optional<at::Tensor> & mask={}, const c10::optional<at::Tensor> & incr_key={}, const c10::optional<at::Tensor> & incr_value={}, bool need_weights=true, bool average_attn_weights=true); |
21 | } // namespace native |
22 | } // namespace at |
23 | |