Struct neuro::data::TabularDataSet

pub struct TabularDataSet { /* fields omitted */ }

Structure representing tabular data.

Methods

impl TabularDataSet

pub fn from_csv(
    inputs: &Path,
    outputs: &Path,
    valid_frac: f64,
    header: bool
) -> Result<TabularDataSet, Error>

Creates a TabularDataSet from a set of csv files.

The data are shuffled before being split into training and validation sets.

Arguments

• inputs - The path to the csv file containing the input features.
• outputs - The path to the csv file containing the output labels.
• valid_frac - The fraction of the data used for validation.
• header - Flag indicating whether the files have a header.

pub fn from_tensor(
    x_train: Tensor,
    y_train: Tensor,
    x_valid: Option<Tensor>,
    y_valid: Option<Tensor>,
    x_test: Option<Tensor>,
    y_test: Option<Tensor>
) -> Result<TabularDataSet, Error>

Creates a TabularDataSet from Tensors.

The samples must be stacked along the fourth dimension.

pub fn normalize_input(&mut self)

Normalizes the features of the training, validation, and test (if any) sets.

The minimum and maximum values of the training features are computed and used to normalize the training, validation, and test sets. After normalization, the distribution of the features in the training sets is a uniform distribution within 0 and 1 (assuming that the features originally come from a uniform distribution).

pub fn standardize_input(&mut self)

Standardizes the features of the training, validation, and test (if any) sets.

The mean and standard deviation of the training features are computed and used to standardize the training, validation, and test sets. After standardization, the distribution of the features in the training set is a normal distribution with a mean of 0 and standard deviation 1 (assuming that the features originally come from a Gaussian distribution).

pub fn normalize_output(&mut self)

Normalizes the labels of the training, validation, and test (if any) sets.

The minimum and maximum values of the training labels are computed and used to normalize the training, validation, and test sets. After normalization, the distribution of the labels in the training sets is a uniform distribution within 0 and 1 (assuming that the labels originally come from a uniform distribution).

pub fn standardize_output(&mut self)

Standardizes the labels of the training, validation, and test (if any) sets.

The mean and standard deviation of the training labels are computed and used to standardize the training, validation, and test sets. After standardization, the distribution of the labels in the training set is a normal distribution with a mean of 0 and standard deviation 1 (assuming that the labels originally come from a Gaussian distribution).

Trait Implementations

impl DataSet for TabularDataSet

fn input_shape(&self) -> Dim4

Returns the dimension of the samples.

fn output_shape(&self) -> Dim4

Returns the dimension of the labels.

fn num_train_samples(&self) -> u64

Returns the number of samples in the training set.

fn num_valid_samples(&self) -> u64

Returns the number of samples in the validation set.

fn x_train(&self) -> &Tensor

Returns a reference to the training samples.

fn y_train(&self) -> &Tensor

Returns a reference to the training labels.

fn x_valid(&self) -> Option<&Tensor>

Returns a reference to the validation samples.

fn y_valid(&self) -> Option<&Tensor>

Returns a reference to the validation labels.

fn x_test(&self) -> Option<&Tensor>

Returns a reference to the test samples.

fn y_test(&self) -> Option<&Tensor>

Returns a reference to the test labels.

fn x_train_stats(&self) -> &Option<(Scaling, Tensor, Tensor)>

Returns a reference to the type of scaling that has been applied to the input features and the values used for the scaling.

fn y_train_stats(&self) -> &Option<(Scaling, Tensor, Tensor)>

Returns a reference to the type of scaling that has been applied to the output labels and the values used for the scaling.

fn classes(&self) -> Option<Vec<String>>

Returns the classes in the data set.

impl Display for TabularDataSet

fn fmt(&self, f: &mut Formatter) -> Result

Formats the value using the given formatter.