Fit the regression model Inputs: x_train: training data (num_examples, num_timestep, num_channels) or (num_examples, num

Initialise pool workers for multi processing :param num_cores: :return:

Initialise the regression model

Fit Rocket Inputs: x_train: training data (num_examples, num_timestep, num_channels) y_train: training targe

Fit a transformer per dimension of the series and transform the series based on the number of coefficients :param series: A set of ti

Loads data from a .ts file into a Pandas DataFrame. Parameters ---------- full_file_path_and_name: str The full pathname of the .

This is a function to process the data, i.e. convert dataframe to numpy array :param X: :param min_len: :param normalise: :return

Save test time :param file_name: :param test_duration: :return:

Save training time :param file_name: :param test_duration: :return:

Transform the series based on the number of coefficients :param series: A set of time series with the shape N x L x D :return

Convert a set of time series to functional data :param series: a set of time series with the shape of N x L :param sample_points: sample

Build the SVR model Inputs: model_params: parameters for SVR in dictionary format Outputs: model: in

Build the RF model Inputs: model_params: parameters for SVR in dictionary format Outputs: model: ini

Build the XGBoost model Inputs: model_params: parameters for SVR in dictionary format Outputs: model

This is a function to calculate metrics for regression. The metrics being calculated are RMSE and MAE. :param y_true: :param y_pred:

Create a directory if path doesn't exists :param directory_path: :return:

Do cross validation to optimise hyperparameters of the model

This is a function to fit a regression model given the name and data :param output_directory: :param regressor_name: :param X_train:

Plot the train/test metrics of Deep Learning models Inputs: hist: training history file_name: save file name model:

Do prediction with Rocket Inputs: x: data for prediction (num_examples, num_timestep, num_channels) Outputs:

Build the Linear model Inputs: model_params: parameters for SVR in dictionary format Outputs: model:

Build the DL models Inputs: input_shape: input shape for the models

This is a function to create the regression model :param regressor_name: :param input_shape: :param output_directory: :param verb

Create a transformer :param transformer_name: :param kwargs: :return:

Fit a transformer for a set of time series :param transformer_name: :param X_train: :param flatten: :param kwargs: :return:

Do prediction using the regression model on x Inputs: x: data for prediction (num_examples, num_timestep, num_channels)

Provide a summary of the model

This is a function to scale the time series uniformly :param data: :param max_len: :return:

Initialise the SVR model Inputs: output_directory: path to store results/models verbose: verbosity

Initialise the RF model Inputs: output_directory: path to store results/models verbose: verbosity

Initialise the XGBoost model Inputs: output_directory: path to store results/models verbose: verbosity

Initialise the Linear model Inputs: output_directory: path to store results/models model_params: parameters

Initialise the Rocket model Inputs: output_directory: path to store results/models n_kernels: number of rand

Initialise the regression model

Initialise the FCN model Inputs: output_directory: path to store results/models input_shape: input shape for

Initialise the DL model Inputs: output_directory: path to store results/models input_shape: input shape for

Initialise the InceptionNetwork model Inputs: output_directory: path to store results/models input_shape: in

Initialise the DL model Inputs: output_directory: path to store results/models input_shape: input shape for

A function for multi-threading :param series: a set of time series to transform with the shape N x L :param pca_transformer: transformer

Build the FCN model Inputs: input_shape: input shape for the model

Build the InceptionNetwork model Inputs: input_shape: input shape for the model

Build the ResNet model Inputs: input_shape: input shape for the model

Cross validation on SVR Inputs: x_train: training data (num_examples, num_features) y_train: training target

Cross validation on RF Inputs: x_train: training data (num_examples, num_features) y_train: training target

Cross validation on XGBoost Inputs: x_train: training data (num_examples, num_features) y_train: training ta

Fit Linear model Inputs: x_train: training data (num_examples, num_timestep, num_channels) or (num_examples, num_feature

Fit the regression model

Fit DL models Inputs: x_train: training data (num_examples, num_timestep, num_channels) y_train: training ta

Convert the series to its functional form, fit the transformer per dimension and transform the series based on the number of coeffici

Do prediction with Rocket Inputs: x: data for prediction (num_examples, num_timestep, num_channels) or (num_examples, nu

Do prediction using the regression model on x

Do prediction with DL models Inputs: x: data for prediction (num_examples, num_timestep, num_channels) Outputs:

Convert a set of time series to functional data :param series: a set of time series with the shape of N x L :param sample_points: sample

Transform the series based on the number of coefficients :param series: A set of time series with the shape N x L x D :return

A function for multi-threading :param series: a set of time series to transform with the shape N x L :param transformer: transformer to t

A function for multi-threading :param series: a set of time series to transform with the shape N x L :param transformer: transformer to t