TensorDistribution<T> class abstract

A distribution that samples tensor element values from independent distributions.

Example

import 'package:calc/calc.dart';

void main() {
  // Construct a distribution for matrices 2 rows and 3 columns.
  // Each element is sampled from unit normal distribution.
  const distribution = TensorDistribution.filled(
    TensorShape([2,3]),
    NormalDistribution(mean: 0.0, variance:1.0),
  );

  // Sample the distribution.
  final tensor = distribution.sample();
}
Inheritance

Constructors

TensorDistribution.filled(TensorShape shape, Distribution<T> distribution)
Constructing a distribution where all indices have identical scalar distribution.
factory
TensorDistribution.fitScalarsIndependently(Iterable<Tensor<double>> iterable, Distribution<T> scalarFitter(Iterable<double> iterable))
Constructs a distribution by fitting a distribution to each index independently.
factory
TensorDistribution.generate(TensorShape shape, Distribution<T> generator(TensorShape shape, int index))
Generates distribution for each index independently.
factory

Properties

distributions → List<Distribution<T>>
Distribution for each (flat) index of the tensor.
no setter
hashCode → int
The hash code for this object.
no setterinherited
runtimeType → Type
A representation of the runtime type of the object.
no setterinherited
shape → TensorShape
Shape of the tensor.
no setter

Methods

map<R>(R f(Tensor<T> value)) → Distribution<R>
Maps samples of the distribution with the given function.
inherited
noSuchMethod(Invocation invocation) → dynamic
Invoked when a nonexistent method or property is accessed.
inherited
sample({Random? random}) → Tensor<T>
Generates a random sample.
inherited
sampleList(int length, {Random? random}) → List<Tensor<T>>
Generates a list of N random samples.
inherited
toString() → String
A string representation of this object.
inherited

Operators

operator ==(Object other) → bool
The equality operator.
inherited