/packages/utilities/src/mod.ts | netsaur@0.4.2

import * as netsaur from "https://deno.land/x/netsaur@0.4.2/packages/utilities/src/mod.ts";

Metrics for Machine Learning outcomes.

Classes

c CategoricalEncoder
c ClassificationReport	A report with metrics for classification results
c ConfusionMatrix	Confusion matrix for the result.
c Image
c Matrix	Class for 2D Arrays. This is not akin to a mathematical Matrix (a collection of column vectors). This is a collection of row vectors. A special case of Tensor for 2D data.
c MultiHotEncoder	Convert 2D array of indices into multi-hot encoded vectors.
c OneHotEncoder	Convert an array of indices into one-hot encoded vectors.
c Tensor	A Tensor of order O.
c TextCleaner	Simple text cleaner
c TextVectorizer
c TfEncoder	Convert 2D array of indices into multi-hot encoded vectors where each index contains the number of times the respective value appears in a sample (term frequency encoder).

The fraction of positives that were predicted correctly

f accuracyScore	The fraction of predictions that were correct
f argmax
f cohensKappa	Compute Cohen's Kappa to find Agreement
f extractColors	Extract colors from an image.
f f1Score	Compute F1 Score
f getConstructor
f getHistogram	Get a histogram of frequency of colors.
f mae	Mean Absolute Error
f mse	Mean Square Error
f patch2d	Extract patches from a 2d image
f precisionScore	The fraction of "positive" predictions that were actually positive
f preprocess	Function for quick cleaning of text
f r2	R2 Score for regression
f rmse	Root Mean Square Error
f sensitivityScore	The fraction of positives that were predicted correctly
f specificityScore	The fraction of negatives that were predicted correctly
f transformSoftmaxMut	Convert a softmax output into one-hot vectors. Mutates the input.
f useNormal	Generate a normal random variate.
f useNormalArray	Generate a normally distributed array.
f useRange	Get n evenly distributed numbers in a range.
f useRearrange	Rearrange characters in a string randomly.
f useRNG	Get random number from range
f useSeries	Get an array of numbers between a given range, incremented by a step.
f useShuffle	Shuffle a given array in-place.
f useSplit	Split arrays by their first axis
f useUnique	Remove duplicate values in an array. Uses a strict = for identifying duplicates.
f useWeighted	Roll one from an array of weighted choices.

I NDArray	nDArray type
I TensorLike	The base type implemented by Tensor

T ImageData
T MatrixLike	The base type implemented by Matrix
T Order	Order of the tensor
T Shape	An array with n items