|
MiniBatchesCreateTInput, TOutput Method
|
Creates a method to partition a given dataset into mini-batches of equal size.
Namespace:
Accord.MachineLearning
Assembly:
Accord.MachineLearning (in Accord.MachineLearning.dll) Version: 3.8.0
Syntax public static Batches<TInput, TOutput> Create<TInput, TOutput>(
TInput[] input,
TOutput[] output,
double[] weights = null,
int batchSize = 32,
int maxIterations = 0,
int maxEpochs = 0,
ShuffleMethod shuffle = ShuffleMethod.EveryEpoch
)
Public Shared Function Create(Of TInput, TOutput) (
input As TInput(),
output As TOutput(),
Optional weights As Double() = Nothing,
Optional batchSize As Integer = 32,
Optional maxIterations As Integer = 0,
Optional maxEpochs As Integer = 0,
Optional shuffle As ShuffleMethod = ShuffleMethod.EveryEpoch
) As Batches(Of TInput, TOutput)
Request Example
View SourceParameters
- input
- Type: TInput
The input data to be partitioned into mini-batches. - output
- Type: TOutput
The output data to be partitioned into mini-batches. - weights (Optional)
- Type: SystemDouble
The weights for the data to be partitioned into mini-batches. - batchSize (Optional)
- Type: SystemInt32
The size of the batch. - maxIterations (Optional)
- Type: SystemInt32
The maximum number of mini-batches that should be created until the method stops. - maxEpochs (Optional)
- Type: SystemInt32
The maximum number of epochs that should be run until the method stops. - shuffle (Optional)
- Type: Accord.MachineLearningShuffleMethod
The data shuffling options.
Type Parameters
- TInput
- The type of the input data.
- TOutput
- The type of the output output.
Return Value
Type:
BatchesTInput,
TOutputSee Also