Mila
Deep Neural Network Library
|
Directories | |
Kernels | |
Files | |
CublasLtMatMulBias.ixx | |
CudaDataTypeTraits.ixx | |
CudaEncoderOp.ixx | |
Implementation of the CUDA-based Encoder operation for transformer models. | |
CudaGeluOp.ixx | |
Implementation of the CUDA-based GELU activation function for neural networks. | |
CudaLayerNormOp.ixx | |
Implementation of the CUDA-based Layer Normalization operation for neural networks. | |
CudaLinearOp.ixx | |
Implementation of the CUDA-based Fully Connected operation for neural networks. | |
CudaMultiHeadAttentionOp.ixx | |
Implementation of the CUDA-based Multi-Head Attention operation for transformer models. | |
CudaOperations.ixx | |
CudaResidualOp.ixx | |
Implementation of the CUDA-based residual operation for neural networks. | |
CudaSoftmaxOp.ixx | |
Implementation of the CUDA-based softmax operation for neural networks. | |
FusedSoftmaxCrossEntropyOp.ixx | |
Implementation of the CUDA-based fused softmax and cross entropy operation for neural networks. | |
MatMulBiasActivation.ixx | |
Implementation of fused matrix multiplication, bias addition, and activation operations. | |