show_random_patches does not show image

[robmarkcole]

Image is not visible:

[samKhallaghi]

Hi Robin, That function is from an older release of the code which we decided to remove later on. I will add the correct one sometime this week.

[samKhallaghi]

You can use this code snippet to visualize random samples from your custom dataset. Notice that you call the function before passing the dataset into the pytorch dataloader.

[samKhallaghi]

import numbers import matplotlib.pyplot as plt import random import torch

def show_random_patches(dataset, sample_num, rgb_bands=(3, 2, 1)): """ Plots a user-defined number of image chips and the corresponding labels. """

if not (isinstance(rgb_bands, (tuple, list)) and len(rgb_bands) == 3 and 
        all(isinstance(band, numbers.Number) for band in rgb_bands) and
        1 <= sample_num <= len(dataset)):
    raise ValueError("'sample_num' or 'rgb_bands' are not properly defined")

# Sample indices for visualization
sample_indices = random.sample(range(len(dataset)), sample_num)

fig, axs = plt.subplots(nrows=sample_num, ncols=2, figsize=(16, sample_num * 16 / 2), squeeze=False)

for i, idx in enumerate(sample_indices):
    img, lbl, _ = dataset[idx]
    r, g, b = (img[band,:,:].cpu().view(1, *lbl.shape) for band in rgb_bands)

    axs[i, 0].set_title(f'Image Patch #{idx}')
    axs[i, 0].imshow(torch.cat([r, g, b], 0).permute(1, 2, 0))
    axs[i, 1].set_title(f'Label Patch #{idx}')
    axs[i, 1].imshow(lbl)

plt.show()