Documentation
¶
Index ¶
- func ConvEval(k *model.Kernel, x []float32, s, inDim, outDim int) ([]float32, error)
- func ConvEvalWithStats(k *model.Kernel, x []float32, s, inDim, outDim int) ([]float32, model.EvalStats, error)
- func FromChannelFirst(x []float32, s, d int) ([]float32, error)
- func ToChannelFirst(x []float32, s, d int) ([]float32, error)
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func ConvEval ¶
ConvEval transposes row-major [S, inDim] to channel-first [inDim, S], evaluates the kernel, and transposes the output back to row-major.
func ConvEvalWithStats ¶
func ConvEvalWithStats(k *model.Kernel, x []float32, s, inDim, outDim int) ([]float32, model.EvalStats, error)
ConvEvalWithStats is ConvEval plus kernel eval stats from EvalWithStats.
func FromChannelFirst ¶
FromChannelFirst transposes channel-first [d, s] to row-major [s, d].
Types ¶
This section is empty.
Source Files
Click to show internal directories.
Click to hide internal directories.