graphorge.gnn_base_model.train.training_plots.plot_loss_convergence_test

plot_loss_convergence_test(testing_loss, training_loss=None, loss_type=None, is_log_loss=False, loss_scale='linear', filename='loss_convergence_test', save_dir=None, is_save_fig=False, is_stdout_display=False, is_latex=False)[source]

Plot testing and training loss for different training data set sizes.

Parameters:

  • testing_loss (numpyp.ndarray(2d)) – Testing loss data array where i-th row is associated with the i-th testing process for a given training data set size and the corresponding data is stored as folows: testing_loss[i, 0] is the size of the dataset used to train the model, testing_loss[i, 1:] is the testing loss for each trained model (e.g., different training data sets in k-fold cross-validation). Missing loss values should be stored as None.

  • training_loss (numpy.ndarray(2d), default=None) – Training loss data array where i-th row is associated with the i-th training process for given training data set size and the corresponding data is stored as folows: training_loss[i, 0] is the size of the data set used to train the model, training_loss[i, 1:] is the training loss for each trained model (e.g., different training data sets in k-fold cross-validation). Missing loss values should be stored as None.

  • loss_type (str, default=None) – Loss type. If provided, then loss type is added to the y-axis label.

  • is_log_loss (bool, default=False) – Applies logarithm to loss values if True, keeps original loss values otherwise.

  • loss_scale ({'linear', 'log'}, default='linear') – Loss axis scale type.

  • filename (str, default='training_loss_history') – Figure name.

  • save_dir (str, default=None) – Directory where figure is saved. If None, then figure is saved in current working directory.

  • is_save_fig (bool, default=False) – Save figure.

  • is_stdout_display (bool, default=False) – True if displaying figure to standard output device, False otherwise.

  • is_latex (bool, default=False) – If True, then render all strings in LaTeX. If LaTex is not available, then this option is silently set to False and all input strings are processed to remove $(…)$ enclosure.