Sample

The Sample class is provides objects to store all kinds of information for a single sample.

It mandatory variables of a Samples are its index (id), an image and some information of the image like its shape and the number of channels. The Sample class can also store a optional segmentation with its number of classes, as well as a predicted segmentation from the model.

It is also possible to add additional custom information in the details dictionary. This feature can be exploited in later custom interfaces like in a Subfunction.

The Data IO class will automatically create Sample objects during a Pipeline run. It is also possible to obtain all created Sample objects by the following call:

sample_list = data_io.get_indiceslist()

Methods

Initialization

Sample(index, image, channels, classes)

Initialization function for creating a Sample object.

Arguments:

• index: Index (String) of a sample.
• image: NumPy array containing the image.
• channels: Number of channels of the image (dimension of last layer).
• classes: Number of classes of the segmentation.

Returns:
A Sample class object.

Example:

sample = data_io.sample_loader(index="case_00001", load_seg=True)

img = sample.img_data

---