visumod package

Module contents

This is the visualization module.

visumod.compare_pareto_fronts(pf_true, pf_calc)

Compare 2D Pareto fronts.

Parameters:
  • pf_true (np.array) – True Pareto front.
  • pf_calc (np.array) – Calculated Pareto front.
visumod.compare_surrogate(inputs, outputs, predict, iteration)

Plot the comparison of raw data and surrogate response.

Parameters:
  • inputs (np.array) – Input data.
  • outputs (np.array) – Output data.
  • predict (method) – Predict method of the surrogate.
  • iteration (int) – Iteration number.
visumod.correlation_heatmap(predict, dim_in)

Plot the correleation heatmap between variables.

Parameters:
  • predict (method) – Predict method of the surrogate.
  • dim_in (int) – Number of input dimensions.
visumod.plot_adaptive_candidates(candidates, data, iteration)

Plot candidates for adaptive sampling.

Parameters:
  • candidates (np.array) – Candidate samples.
  • data (np.array) – Combined adaptive sampling metric.
  • iteration (int) – Iteration number.
visumod.plot_raw(data, iteration, normalized=False)

Plot either a scatter, curve or surface plot.

Parameters:
  • data (np.array) – Raw data samples.
  • iteration (int) – Iteration number.
  • normalized (bool) – Whether the data is normalized.

Notes

Surface plot not used yet.

visumod.plot_training_history(history, train_in, train_out, test_in, test_out, predict, progress)

Plot the evolution of the training and testing error.

Parameters:
  • history (tensorflow.python.keras.callbacks.History/metamod.ANN_pt.TrainHistory) – Metrics history during the training.
  • train_in (np.array/torch.Tensor) – Training input data.
  • train_out (np.array/torch.Tensor) – Training output data.
  • test_in (np.array/torch.Tensor) – Testing input data.
  • test_out (np.array/torch.Tensor) – Testing output data.
  • predict (method) – Predict method of the surrogate.
  • progress (list) – Training progress status.
visumod.sample_size_convergence(metrics)

Plot the sample size convergence.

Parameters:metrics (dict) – Dictionary of convergence metrics.
visumod.surrogate_response(inputs, outputs, iteration)

Plot the surrogate response.

Parameters:
  • inputs (np.array) – Input data.
  • outputs (np.array) – Output to plot.
  • iteration (int) – Iteration number.
visumod.vis_design_space(data, iteration)

Visualize the design space in design coordinates.

Parameters:
  • res (pymoo.model.result.Result) – Results object.
  • iteration (int) – Iteration number.
visumod.vis_objective_space(data, iteration)

Visualize the design space in objective coordinates.

Parameters:
  • res (pymoo.model.result.Result) – Results object.
  • iteration (int) – Iteration number.
visumod.vis_objective_space_pcp(data, iteration)

Visualize the design space in objective coordinates with the parallel coordinates plot.

Parameters:
  • data (np.array) – Multidimensional Pareto front.
  • iteration (int) – Iteration number.