nextorch.plotting

Creates 1-dimensional, 2-dimensional and 3-dimensional visualizations The plots are rendered using matplotlib as a backend

nextorch.plotting.acq_func_1d(acq_func: botorch.acquisition.acquisition.AcquisitionFunction, X_test: Union[list, nextorch.utils.Matrix, torch.Tensor], n_dim: Optional[int] = 1, X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, x_index: Optional[int] = 0, X_train: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_names: Optional[str] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot 1-dimensional acquision function at the given dimension defined by x_index

Parameters

  • acq_func ('botorch.acquisition.AcquisitionFunction'_) – the acquision function object

  • X_test (MatrixLike2d) – Test data points for plotting

  • n_dim (Optional[int], optional) – Dimensional of X, i.e., number of columns by default 1

  • X_ranges (Optional[MatrixLike2d], optional) – list of x ranges, by default None

  • x_index (Optional[int], optional) – index of the x variable, by default 0

  • X_train (Optional[MatrixLike2d], optional) – Training data points, by default None

  • X_new (Optional[MatrixLike2d], optional) – The next data point, i.e the infill points, by default None

  • X_names (Optional[str], optional) – Name of X varibale shown as x-label

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[Union[str, int]], optional) – Iteration index to add to the figure name by default ‘’

  • .._'botorch.acquisition.AcquisitionFunction' (https://botorch.org/api/acquisition.html) –

nextorch.plotting.acq_func_1d_exp(Exp: nextorch.bo.Experiment, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, x_index: Optional[int] = 0, fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = None, fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = None, baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, save_fig: Optional[bool] = False)

Plot 1-dimensional acquision function at the given dimension defined by x_index Using Experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • X_new (Optional[MatrixLike2d], optional) – The next data point, i.e the infill points, by default None

  • x_index (Optional[int], optional) – index of two x variables, by default 0

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default None

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default None

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (int, optional) – mesh size, by default 41

nextorch.plotting.add_x_slice_2d(ax: matplotlib.axes._axes.Axes, xvalue: float, yrange: List[float], zrange: List[float], mesh_size: Optional[int] = 100) → matplotlib.axes._axes.Axes

Adds a 2-dimensional plane on x axis, parallel to y-z plane in the 3-dimensional (x, y, z) space

Parameters

  • ax (`matplotlib.axes.Axes.axis`_) – Ax of the plot

  • xvalue (float) – the value on x axis which the slice is made

  • yrange (list of float) – [left bound, right bound] of y value

  • zrange (list of float) – [left bound, right bound] of z value

  • mesh_size (Optional[int], optional) – mesh size on the slice, by default 100

Returns

  • ax (`matplotlib.axes.Axes.axis`_) – Axes of the plots

  • .. _`matplotlib.axes.Axes.axis` (https://matplotlib.org/api/_as_gen/matplotlib.axes.Axes.axis.html)

nextorch.plotting.add_y_slice_2d(ax: matplotlib.axes._axes.Axes, yvalue: float, xrange: List[float], zrange: List[float], mesh_size: Optional[int] = 100) → matplotlib.axes._axes.Axes

Adds a 2-dimensional plane on y axis, parallel to x-z plane in the 3-dimensional (x, y, z) space

Parameters

  • ax (`matplotlib.axes.Axes.axis`_) – Ax of the plot

  • yvalue (float) – the value on y axis which the slice is made

  • xrange (list of float) – [left bound, right bound] of x value

  • zrange (list of float) – [left bound, right bound] of z value

  • mesh_size (Optional[int], optional) – mesh size on the slice, by default 100

Returns

  • ax (`matplotlib.axes.Axes.axis`_) – Axes of the plots

  • .. _`matplotlib.axes.Axes.axis` (https://matplotlib.org/api/_as_gen/matplotlib.axes.Axes.axis.html)

nextorch.plotting.add_z_slice_2d(ax: matplotlib.axes._axes.Axes, zvalue: float, xrange: List[float], yrange: List[float], mesh_size: Optional[int] = 100) → matplotlib.axes._axes.Axes

Adds a 2-dimensional plane on z axis, parallel to x-y plane in the 3-dimensional (x, y, z) space

Parameters

  • ax (`matplotlib.axes.Axes.axis`_) – Ax of the plot

  • zvalue (float) – the value on z axis which the slice is made

  • xrange (list of float) – [left bound, right bound] of x value

  • yrange (list of float) – [left bound, right bound] of y value

  • mesh_size (Optional[int], optional) – mesh size on the slice, by default 100

Returns

  • ax (`matplotlib.axes.Axes.axis`_) – Axes of the plots

  • .. _`matplotlib.axes.Axes.axis` (https://matplotlib.org/api/_as_gen/matplotlib.axes.Axes.axis.html)

nextorch.plotting.objective_heatmap(objective_func: object, X_ranges: Union[list, nextorch.utils.Matrix, torch.Tensor], Y_name: Optional[str] = None, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], X_names: Optional[List[str]] = None, X_train: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, save_fig: Optional[bool] = False, name: Optional[str] = 'simple_experiment')

Show a 3-dimensional response surface in a real scale Using the experiment object

Parameters

  • objective_func (function object) – a objective function to optimize

  • X_ranges (MatrixLike2d,) – list of x ranges

  • Y_name (Optional[str], optional) – Name of Y variable, by default None

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • X_name (Optional[List(str)], optional) – Names of X varibale shown as x,y,z-labels by default None

  • X_train (Optional[MatrixLike2d], optional) – Data points used in training, by default None

  • X_new (Optional[MatrixLike2d], optional) – The next data point, i.e the infill points, by default None

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • name (Optional[str], optional) – Name of the objective function, by default ‘simple_experiment’

nextorch.plotting.objective_heatmap_exp(Exp: nextorch.bo.Experiment, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, show_samples: Optional[bool] = True, save_fig: Optional[bool] = False)

Show a heat map for objective function in a real scale Using the experiment object :param Exp: Experiment object :type Exp: Experiment :param X_new: The next data point, i.e the infill points,

by default None

Parameters

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • show_samples (Optional[bool], optional) – if true show the sample points by default True

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.objective_surface(objective_func: object, X_ranges: Union[list, nextorch.utils.Matrix, torch.Tensor], Y_name: Optional[str] = None, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], baseline: Optional[str] = 'left', X_names: Optional[List[str]] = None, mesh_size: Optional[int] = 41, save_fig: Optional[bool] = False, name: Optional[str] = 'simple_experiment')

Show a 3-dimensional response surface in a real scale Using the experiment object

Parameters

  • objective_func (function object) – a objective function to optimize

  • X_ranges (MatrixLike2d,) – list of x ranges

  • Y_name (Optional[str], optional) – Name of Y variable, by default None

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • X_name (Optional[List(str)], optional) – Names of X varibale shown as x,y,z-labels by default None

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • name (Optional[str], optional) – Name of the objective function, by default ‘simple_experiment’

nextorch.plotting.objective_surface_exp(Exp: nextorch.bo.Experiment, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, save_fig: Optional[bool] = False)

Show a 3-dimensional response surface in a real scale Using the experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.opt_per_trial(Ys: Union[list, nextorch.utils.Array, torch.Tensor], maximize: Optional[bool] = True, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, Y_name: Optional[str] = None, log_flag: Optional[bool] = False, design_names: Optional[Union[str, List[str]]] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None)

Discovery plot show the optimum value performance versus the trial number i.e. the index of training data

Parameters

  • Ys (Union[list, ArrayLike1d]) – Response of each design in a real scale

  • maximize (Optional[bool], optional) – by default True, maximize the objective function Otherwise False, minimize the objective function

  • Y_real_range (ArrayLike1d) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • Y_name (Optional[str], optional) – Name of Y variable, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale, by default False

  • design_names (Optional[List[str]], optional) – Names of the designs, by default None

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

nextorch.plotting.opt_per_trial_exp(Exp: nextorch.bo.Experiment, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, save_fig: Optional[bool] = False)

Discovery plot show the optimum value performance versus the trial number i.e. the index of training data Using the experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • Y_real_range (ArrayLike1d) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale, by default False

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.pareto_front(y1: Union[list, nextorch.utils.Matrix, torch.Tensor], y2: Union[list, nextorch.utils.Matrix, torch.Tensor], Y_names: Optional[List[str]] = None, fill: Optional[bool] = True, diagonal: Optional[bool] = True, save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot parity plot comparing the ground true objective function values against predicted model mean

Parameters

  • y1 (MatrixLike2d) – Ground truth values

  • y2 (MatrixLike2d) – Model predicted values

  • fill (Optional[bool], optional) – if true fill the space enclosed by the points by default True

  • diagonal (Optional[bool], optional) – if true plot the y = x line by default True

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[Union[str, int]], optional) – Iteration number to add to the figure name by default ‘’

nextorch.plotting.pareto_front_exp(Exp: Union[nextorch.bo.WeightedMOOExperiment, nextorch.bo.EHVIMOOExperiment], fill: Optional[bool] = True, diagonal: Optional[bool] = True, save_fig: Optional[bool] = False, design_name: Optional[Union[str, int]] = 'final')

Plot parity plot comparing the ground true objective function values against predicted model mean Using MOOExperiment object

Parameters

  • Exp (Union[WeightedMOOExperiment, EHVIMOOExperiment]) – MOOExperiment object

  • fill (Optional[bool], optional) – if true fill the space enclosed by the points by default True

  • diagonal (Optional[bool], optional) – if true plot the y = x line by default True

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • design_name (Optional[Union[str, int]], optional) – Design name to add to the figure name by default ‘final’

nextorch.plotting.parity(y1: Union[list, nextorch.utils.Matrix, torch.Tensor], y2: Union[list, nextorch.utils.Matrix, torch.Tensor], save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot parity plot comparing the ground true objective function values against predicted model mean

Parameters

  • y1 (MatrixLike2d) – Ground truth values

  • y2 (MatrixLike2d) – Model predicted values

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[Union[str, int]], optional) – Iteration number to add to the figure name by default ‘’

nextorch.plotting.parity_exp(Exp: nextorch.bo.Experiment, save_fig: Optional[bool] = False, design_name: Optional[Union[str, int]] = 'final')

Plot parity plot comparing the ground true objective function values against predicted model mean Using Experiment object

Parameters

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • design_name (Optional[Union[str, int]], optional) – Design name to add to the figure name by default ‘final’

nextorch.plotting.parity_with_ci(y1: Union[list, nextorch.utils.Matrix, torch.Tensor], y2: Union[list, nextorch.utils.Matrix, torch.Tensor], y2_lower: Union[list, nextorch.utils.Matrix, torch.Tensor], y2_upper: Union[list, nextorch.utils.Matrix, torch.Tensor], save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot parity plot comparing the ground true objective function values against predicted model mean with predicted confidence interval as error bars

Parameters

  • y1 (MatrixLike2d) – Ground truth values

  • y2 (MatrixLike2d) – Model predicted values

  • y2_lower (MatrixLike2d) –

  • y2_upper (MatrixLike2d) –

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[Union[str, int]], optional) – Iteration number to add to the figure name by default ‘’

nextorch.plotting.parity_with_ci_exp(Exp: nextorch.bo.Experiment, save_fig: Optional[bool] = False, design_name: Optional[Union[str, int]] = 'final')

Plot parity plot comparing the ground true objective function values against predicted model mean with predicted confidence interval as error bars Using Experiment object

Parameters

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • design_name (Optional[Union[str, int]], optional) – Design name to add to the figure name by default ‘final’

nextorch.plotting.response_1d(model: botorch.models.model.Model, X_test: Union[list, nextorch.utils.Matrix, torch.Tensor], n_dim: Optional[int] = 1, X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, x_index: Optional[int] = 0, Y_test: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_train: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_train: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, negate_Y: Optional[bool] = False, plot_real: Optional[bool] = False, Y_mean: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_std: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_names: Optional[str] = None, Y_name: Optional[str] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot response values at the given dimension defined by x_index Input X variables are in a unit scale and Input Y variables are in a real scale

Parameters

  • model ('botorch.models.model.Model'_) – A GP model

  • X_test (MatrixLike2d) – Test data points for plotting

  • n_dim (Optional[int], optional) – Dimensional of X, i.e., number of columns by default 1

  • X_ranges (Optional[MatrixLike2d], optional) – list of x ranges, by default None

  • x_index (Optional[int], optional) – index of the x variable, by default 0

  • Y_test (Optional[MatrixLike2d], optional) – Test Y data if the objective function is known, by default None

  • X_train (Optional[MatrixLike2d], optional) – Training X data points, by default None

  • Y_train (Optional[MatrixLike2d], optional) – Training Y data points, by default None

  • X_new (Optional[MatrixLike2d], optional) – The next X data point, i.e the infill points, by default None

  • Y_new (Optional[MatrixLike2d], optional) – The next Y data point, i.e the infill points, by default None

  • plot_real (Optional[bool], optional) – if true plot in the real scale for Y, by default False

  • Y_mean (MatrixLike2d) – The mean of initial Y set

  • Y_std (MatrixLike2d) – The std of initial Y set

  • X_names (Optional[str], optional) – Name of X varibales shown as x-label

  • Y_name (Optional[str], optional) – Name of Y varibale shown as y-label

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[str], optional) – Iteration number to add to the figure name by default ‘’

Raises
nextorch.plotting.response_1d_exp(Exp: nextorch.bo.Experiment, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, x_index: Optional[int] = 0, y_index: Optional[int] = 0, fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = None, fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = None, baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, plot_real: Optional[bool] = False, save_fig: Optional[bool] = False)

Plot reponse valus at the given dimension defined by x_index using Experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • X_test (MatrixLike2d) – Test X data points for plotting

  • Y_test (Optional[MatrixLike2d], optional) – Test Y data if the objective function is known, by default None

  • X_new (Optional[MatrixLike2d], optional) – The next X data point, i.e the infill points, by default None

  • Y_new (Optional[MatrixLike2d], optional) – The next Y data point, i.e the infill points, by default None

  • x_index (Optional[int], optional) – index of two x variables, by default 0

  • y_index (Optional[int], optional) – index of the y variables, by default 0

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default None

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default None

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (int, optional) – mesh size, by default 41

  • plot_real (Optional[bool], optional) – if true plot in the real scale for Y, by default False

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.response_heatmap(Y_real: Union[list, nextorch.utils.Matrix, torch.Tensor], Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, Y_name: Optional[str] = None, log_flag: Optional[bool] = False, n_dim: Optional[int] = 2, x_indices: Optional[List[int]] = [0, 1], X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_names: Optional[List[str]] = None, X_train: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Show a heat map for the response in a real scale

Parameters

  • Y_real (MatrixLike2d) – Response in a real scale

  • Y_real_range (ArrayLike1d) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • Y_name (Optional[str], optional) – Names of Y variable, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale, by default False

  • n_dim (Optional[int], optional) – Dimensional of X, i.e., number of columns by default 2

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • X_ranges (Optional[MatrixLike2d], optional) – list of x ranges, by default None

  • X_name (Optional[List(str)], optional) – Names of X varibale shown as x,y,z-labels by default None

  • X_train (Optional[MatrixLike2d], optional) – Data points used in training, by default None

  • X_new (Optional[MatrixLike2d], optional) – The next data point, i.e the infill points, by default None

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[str], optional) – Iteration number to add to the figure name by default ‘’’

nextorch.plotting.response_heatmap_err_exp(Exp: nextorch.bo.Experiment, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, save_fig: Optional[bool] = False)

Show a heat map for percentage error (objective - response)/objective in a real scale Using the experiment object :param Exp: Experiment object :type Exp: Experiment :param X_new: The next data point, i.e the infill points,

by default None

Parameters

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.response_heatmap_exp(Exp: nextorch.bo.Experiment, X_new: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], baseline: Optional[str] = 'left', mesh_size: Optional[int] = 41, show_samples: Optional[bool] = True, save_fig: Optional[bool] = False)

Show a heat map for the response in a real scale Using the experiment object :param Exp: Experiment object :type Exp: Experiment :param X_new: The next data point, i.e the infill points,

by default None

Parameters

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • show_samples (Optional[bool], optional) – if true show the sample points by default True

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.response_scatter_exp(Exp: nextorch.bo.Experiment, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, Y_name: Optional[str] = None, n_dim: Optional[int] = 3, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1, 2], X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_names: Optional[List[str]] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot a response surface in 3-dimensional space

Parameters

  • Y_real_range (ArrayLike1d) – Ranges of the response, [lb, rb]

  • Y_name (Optional[str], optional) – Name of Y variable, by default None

  • n_dim (Optional[int], optional) – Dimensional of X, i.e., number of columns

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • X_ranges (MatrixLike2d, optional) – list of x ranges, by default None

  • X_names (Optional[List(str)], optional) – Names of X varibale shown as x,y,z-labels

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[str], optional) – Iteration number to add to the figure name by default ‘’’

nextorch.plotting.response_surface(X1_test: Union[list, nextorch.utils.Matrix, torch.Tensor], X2_test: Union[list, nextorch.utils.Matrix, torch.Tensor], Y_real: Union[list, nextorch.utils.Matrix, torch.Tensor], Y_real_lower: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_real_upper: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, Y_name: Optional[str] = None, n_dim: Optional[int] = 2, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, X_names: Optional[List[str]] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None, i_iter: Optional[Union[str, int]] = '')

Plot a response surface in 3-dimensional space

Parameters

  • X1_test (MatrixLike2d) – [description]

  • X2_test (MatrixLike2d) – [description]

  • Y_real (MatrixLike2d) – Response in a real scale

  • Y_real_lower (Optional[MatrixLike2d], optional) – Model predicted lower bound in a real scale, by default None

  • Y_real_upper (Optional[MatrixLike2d], optional) – Model predicted lower bound in a real scale, , by default None

  • Y_real_range (ArrayLike1d) – Ranges of the response, [lb, rb]

  • Y_name (Optional[str], optional) – Name of Y variable, by default None

  • n_dim (Optional[int], optional) – Dimensional of X, i.e., number of columns

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • X_ranges (Optional[MatrixLike2d], optional) – list of x ranges, by default None

  • X_name (Optional[List(str)], optional) – Names of X varibale shown as x,y,z-labels

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

  • i_iter (Optional[str], optional) – Iteration number to add to the figure name by default ‘’’

nextorch.plotting.response_surface_exp(Exp: nextorch.bo.Experiment, Y_real_range: Optional[Union[list, nextorch.utils.Array, torch.Tensor]] = None, log_flag: Optional[bool] = False, x_indices: Optional[List[int]] = [0, 1], fixed_values: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], fixed_values_real: Optional[Union[list, nextorch.utils.Array, torch.Tensor, float]] = [], baseline: Optional[str] = 'left', show_confidence: Optional[bool] = False, mesh_size: Optional[int] = 41, save_fig: Optional[bool] = False)

Show a 3-dimensional response surface in a real scale Using the experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • Y_real_range (Optional[ArrayLike1d], optional) – Ranges of the response, [lb, rb] to show on the plot, by default None

  • log_flag (Optional[bool], optional) – flag to plot in a log scale

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • fixed_values (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a unit scale, by default []

  • fixed_values_real (Optional[Union[ArrayLike1d, float]], optional) – fixed values in other dimensions, in a real scale, by default []

  • baseline (Optional[str], optional) – the choice of baseline, must be left, right or center

  • mesh_size (Optional[int], optional) – mesh size, by default 41

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.sampling_2d(Xs: Union[list, nextorch.utils.Matrix, torch.Tensor, List[Union[list, nextorch.utils.Matrix, torch.Tensor]]], X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, x_indices: Optional[List[int]] = [0, 1], X_names: Optional[List[str]] = None, design_names: Optional[Union[str, List[str]]] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None)

Plot sampling plan(s) in 2 dimensional space X must be 2 dimensional

Parameters

  • Xs (Union[MatrixLike2d, List[MatrixLike2d]]) – The set of sampling plans, Can be a list of matrices or one matrix

  • X_ranges (Optional[MatrixLike2d], optional) – list of x ranges, by default None

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • X_name (Optional[str], optional) – Names of X varibale shown as x,y-labels

  • design_names (Optional[List[str]], optional) – Names of the designs, by default None

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

nextorch.plotting.sampling_2d_exp(Exp: nextorch.bo.Experiment, x_indices: Optional[List[int]] = [0, 1], design_names: Optional[Union[str, List[str]]] = None, save_fig: Optional[bool] = False)

Plot sampling plan(s) in 2 dimensional space X must be 2 dimensional Using the experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • x_indices (Optional[List[int]], optional) – indices of two x variables, by default [0, 1]

  • design_names (Optional[List[str]], optional) – Names of the designs, by default None

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.sampling_3d(Xs: Union[list, nextorch.utils.Matrix, torch.Tensor, List[Union[list, nextorch.utils.Matrix, torch.Tensor]]], X_ranges: Optional[Union[list, nextorch.utils.Matrix, torch.Tensor]] = None, x_indices: Optional[List[int]] = [0, 1, 2], X_names: Optional[List[str]] = None, slice_axis: Optional[Union[str, int]] = None, slice_value: Optional[float] = None, slice_value_real: Optional[float] = None, design_names: Optional[Union[str, List[str]]] = None, save_fig: Optional[bool] = False, save_path: Optional[str] = None)

Plot sampling plan(s) in 3 dimensional space X must be 3 dimensional

Parameters

  • Xs (Union[MatrixLike2d, List[MatrixLike2d]]) – The set of sampling plans in a unit scale, Can be a list of matrices or one matrix

  • X_ranges (Optional[MatrixLike2d], optional) – list of x ranges, by default None

  • x_indices (Optional[List[int]], optional) – indices of three x variables, by default [0, 1, 2]

  • X_name (Optional[List(str)], optional) – Names of X varibale shown as x,y,z-labels

  • slice_axis (Optional[Union[str, int]], optional) – axis where a 2d slice is made, by default None

  • slice_value (Optional[float], optional) – value on the axis where a 2d slide is made, in a unit scale, by default None

  • slice_value_real (Optional[float], optional) – value on the axis where a 2d slide is made, in a real scale, by default None

  • design_names (Optional[List[str]], optional) – Names of the designs, by default None

  • save_fig (Optional[bool], optional) – if true save the plot by default False

  • save_path (Optional[str], optional) – Path where the figure is being saved by default the current directory

Raises

  • ValueError – if input axis is defined but the value is not given

  • ValueError – if input axis name is not x, y or z, or 0, 1, 2

nextorch.plotting.sampling_3d_exp(Exp: nextorch.bo.Experiment, x_indices: Optional[List[int]] = [0, 1, 2], slice_axis: Optional[str] = None, slice_value: Optional[float] = None, slice_value_real: Optional[float] = None, design_names: Optional[Union[str, List[str]]] = None, save_fig: Optional[bool] = False)

Plot sampling plan(s) in 3 dimensional space X must be 3 dimensional Using the experiment object

Parameters

  • Exp (Experiment) – Experiment object

  • x_indices (Optional[List[int]], optional) – indices of three x variables, by default [0, 1, 2]

  • slice_axis (Optional[str], optional) – axis where a 2d slice is made, by default None

  • slice_value (Optional[float], optional) – value on the axis where a 2d slide is made, by default None

  • slice_value_real (Optional[float], optional) – value on the axis where a 2d slide is made, in a real scale, by default None

  • design_names (Optional[List[str]], optional) – Names of the designs, by default None

  • save_fig (Optional[bool], optional) – if true save the plot by default False

nextorch.plotting.set_axis_values(xi_range: Union[list, nextorch.utils.Array, torch.Tensor], n_sections: Optional[int] = 2, decimals: Optional[int] = 2) → Union[list, nextorch.utils.Array, torch.Tensor]

Divide xi_range into n_sections

Parameters

  • xi_range (ArrayLike1d) – range of x, [left bound, right bound]

  • n_sections (Optional[int], optional) – number of sections, by default 2

  • decimals (Optional[int], optional) – number of decimal places to keep, by default 2

Returns

axis_values – axis values with rounding up Number of values is n_sections + 1

Return type

ArrayLike1d