CerebNet.utils.misc

CerebNet.utils.misc.load_classwise_weights(cfg)

Loading class-wise median frequency weights.

CerebNet.utils.misc.plot_confusion_matrix(cm, classes, title='Confusion matrix', cmap=<matplotlib.colors.LinearSegmentedColormap object>, figsize=(20, 20), file_save_name=None)

This function prints and plots the confusion matrix.

Parameters:

cmnp.ndarray

Confusion matrix.

classeslist

List of classes.

titlestr, default=”Confusion matrix”

Title of the confusion matrix.

cmapplt.cm, default=matplotlib.pyplot.cm.Blues

Color map.

figsizetuple, default=(20, 20)

Figure size.

file_save_namestr, optional

File save name.

Returns:

figplt.Figure

Figure object.

CerebNet.utils.misc.plot_predictions(images_batch, labels_batch, batch_output, plt_title)

Function to plot predictions from validation set. :param images_batch: :param labels_batch: :param batch_output: :param plt_title: :param file_save_name: :return:

CerebNet.utils.misc.set_summary_path(cfg)

Set last experiment number(EXPR_NUM) and updates the summary path accordingly.

Parameters:

cfgyacs.config.CfgNode

Configuration node.

CerebNet.utils.misc.update_results_dir(cfg)

It will update the results path by finding the last experiment number.

Parameters:

cfgyacs.config.CfgNode

Configuration node.

CerebNet.utils.misc.update_split_path(cfg)

Updating path with respect to the split number

Parameters:

cfgyacs.config.CfgNode

Configuration node.

CerebNet.utils.misc.visualize_batch(img, label, idx)

For deubg :param batch_dict: :return: