Documentation
¶
Index ¶
- Variables
- func FillTensorRows(batchSize int, inputs tensor.Tensor) (x tensor.Tensor, err error)
- func InitForwardOnlyVm(uProfileDim, uBehaviorSize, uBehaviorDim, iFeatureDim, cFeatureDim int, ...) (err error)
- func Predict(m Model, numExamples, batchSize int, si *rcmd.SampleInfo, inputs tensor.Tensor) (y []float32, err error)
- func Train(uProfileDim, uBehaviorSize, uBehaviorDim, iFeatureDim, cFeatureDim int, ...) (err error)
- type Model
Constants ¶
This section is empty.
Variables ¶
View Source
var DT = tensor.Float32
Functions ¶
func FillTensorRows ¶
FillTensorRows fills the batch samples with the zero data to make sample size fit the batch size it sames tensor.Concat is not optimized for large dataset. we should avoid using FillTensorRows while input data is large.
func InitForwardOnlyVm ¶
Types ¶
Click to show internal directories.
Click to hide internal directories.