infini::ops::MoeWna16Gemm Member List

This is the complete list of members for infini::ops::MoeWna16Gemm, including all inherited members.

active_implementation_indices(Device::Type dev_type)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
active_implementation_indices(Device::Type dev_type)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
bit_	infini::ops::MoeWna16Gemm	protected
block_size_k_	infini::ops::MoeWna16Gemm	protected
block_size_m_	infini::ops::MoeWna16Gemm	protected
block_size_n_	infini::ops::MoeWna16Gemm	protected
Call(const Handle &handle, const Config &config, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	static
Call(const Tensor tensor, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Call(const TensorLike &tensor, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Call(const Handle &handle, const Config &config, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Call(const Tensor tensor, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Call(const TensorLike &tensor, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
clear_cache()	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
clear_cache()	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
config_ptr_	infini::ops::OperatorBase	protected
device_index_	infini::ops::MoeWna16Gemm	protected
device_type_	infini::ops::MoeWna16Gemm	protected
dtype_	infini::ops::MoeWna16Gemm	protected
expert_ids_size_	infini::ops::MoeWna16Gemm	protected
group_size_	infini::ops::MoeWna16Gemm	protected
handle_ptr_	infini::ops::OperatorBase	protected
has_qzeros_	infini::ops::MoeWna16Gemm	protected
has_topk_weights_	infini::ops::MoeWna16Gemm	protected
implementation_index_	infini::ops::Operator< MoeWna16Gemm >	protectedstatic
k_	infini::ops::MoeWna16Gemm	protected
m_	infini::ops::MoeWna16Gemm	protected
Make(const Config &config, const Tensor tensor, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const Tensor tensor, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const Config &config, const std::vector< Tensor > tensors, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const std::vector< Tensor > tensors, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const Config &config, const Tensor tensor, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const Tensor tensor, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const Config &config, const std::vector< Tensor > tensors, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
Make(const std::vector< Tensor > tensors, Args &&... args)	infini::ops::Operator< MoeWna16Gemm >	inlinestatic
MoeWna16Gemm(const Tensor input, const Tensor b_qweight, const Tensor b_scales, std::optional< Tensor > b_qzeros, std::optional< Tensor > topk_weights, const Tensor sorted_token_ids, const Tensor expert_ids, const Tensor num_tokens_post_pad, const int64_t top_k, const int64_t block_size_m, const int64_t block_size_n, const int64_t block_size_k, const int64_t bit, Tensor output)	infini::ops::MoeWna16Gemm	inline
n_	infini::ops::MoeWna16Gemm	protected
num_experts_	infini::ops::MoeWna16Gemm	protected
num_groups_	infini::ops::MoeWna16Gemm	protected
operator()(const Tensor input, const Tensor b_qweight, const Tensor b_scales, std::optional< Tensor > b_qzeros, std::optional< Tensor > topk_weights, const Tensor sorted_token_ids, const Tensor expert_ids, const Tensor num_tokens_post_pad, const int64_t top_k, const int64_t block_size_m, const int64_t block_size_n, const int64_t block_size_k, const int64_t bit, Tensor output) const =0	infini::ops::MoeWna16Gemm	pure virtual
Operator< MoeWna16Gemm >::operator()(const Handle &handle, const Args &... args)	infini::ops::Operator< MoeWna16Gemm >	inline
Operator< MoeWna16Gemm >::operator()(const Args &... args) const	infini::ops::Operator< MoeWna16Gemm >	inline
set_config(const Config &config)	infini::ops::OperatorBase	inline
set_config(const Config &config)	infini::ops::OperatorBase	inline
set_handle(const Handle &handle)	infini::ops::OperatorBase	inline
set_handle(const Handle &handle)	infini::ops::OperatorBase	inline
set_stream(void *stream)	infini::ops::OperatorBase	inline
set_stream(void *stream)	infini::ops::OperatorBase	inline
set_workspace(void *workspace)	infini::ops::OperatorBase	inline
set_workspace(void *workspace)	infini::ops::OperatorBase	inline
set_workspace_size_in_bytes(std::size_t workspace_size_in_bytes)	infini::ops::OperatorBase	inline
set_workspace_size_in_bytes(std::size_t workspace_size_in_bytes)	infini::ops::OperatorBase	inline
sorted_token_ids_size_	infini::ops::MoeWna16Gemm	protected
stream_	infini::ops::OperatorBase	protected
top_k_	infini::ops::MoeWna16Gemm	protected
ValidateCallMetadata(const Tensor input, const Tensor b_qweight, const Tensor b_scales, std::optional< Tensor > b_qzeros, std::optional< Tensor > topk_weights, const Tensor sorted_token_ids, const Tensor expert_ids, const Tensor num_tokens_post_pad, const int64_t top_k, const int64_t block_size_m, const int64_t block_size_n, const int64_t block_size_k, const int64_t bit, const Tensor output) const	infini::ops::MoeWna16Gemm	inlineprotected
workspace_	infini::ops::OperatorBase	protected
workspace_size_in_bytes() const	infini::ops::OperatorBase	inlinevirtual
workspace_size_in_bytes() const	infini::ops::OperatorBase	inlinevirtual
workspace_size_in_bytes_	infini::ops::OperatorBase	protected
~OperatorBase()=default	infini::ops::OperatorBase	virtual
~OperatorBase()=default	infini::ops::OperatorBase	virtual