Paddle

Author	SHA1	Message	Date
Markus Kliegl	164898277c	MatMul operator (#4856 ) * initial matmul operator Similar to np.matmul, but also has transpose_X and transpose_Y flags, and only supports tensors from rank 1 to 3 inclusive. For GPU, uses cublas?gemmStridedBatched. For CPU, uses cblas_?gemm_batch if available via MKL; otherwise a simple serial implementation that loops over the batch dimension is employed for now.	8 years ago
qijun	ab5dc9fe18	remove SelectedRows functors to selected_rows_functor.h	8 years ago
qijun	931572e210	SelectedRowsAddTensor method	8 years ago
qijun	5be10872f9	add selected_rows add cpu functor	8 years ago
qijun	c634a8480a	add SetConstant method in math_function.h	8 years ago
guosheng	9ffa79cd91	Add gemm with stride	8 years ago
Yu Yang	81d56ca86b	Remove lazy-initialization in device_context * Also use `const DeviceContext&` all the time, to prevent `const_cast` Fix #4169 Fix #3468 Fix #3475	8 years ago
qijun	960a525550	fix gpu build error	8 years ago
qijun	2ec8dab4c7	follow comments	8 years ago
qijun	de967fcefe	set gemm support continuous memory now	8 years ago
qijun	7eb07b3399	fix conflict	8 years ago
qijun	8de4e3bdd6	disable gpu implementation temporarily	8 years ago
qijun	7307b439e1	fix gpu build error	8 years ago
qijun	97d8175a5e	add global matmul function for Tensor	8 years ago
qijun	f190a79538	fix gpu build error	8 years ago
qijun	22dac40c3a	add gemm for both cpu and gpu	8 years ago