1x1 convolution

3x3 convolution with padding

3x3 convolution with padding

Assemble list of images + mask files fine - modes: train/val/test/trainval cv:0,1,2 coarse - modes: train/val cv:n

x.shape : (None, 1, len_src), returns 2D batch of a mono power-spectrogram

[K] Convert (linear) amplitude to decibel (log10(x)). x: Keras *batch* tensor or variable. It has to be batch because of sample-wise `K.max()`.

Call this function in your script after you have finished setting all cfg values that are necessary (e.g., merging a config from a file, merging

Return matrix of 0/1 predictions, same shape as probabilities.

Calculate weights of classes based on the training crop

Calculate weights of the classes based on training crop

Colorize a segmentation mask.

3x3 convolution with padding

3x3 convolution with padding

Custom Softmax

Handle partial loading when some tensors don't match up in size. Because we want to use models that were trained off a different number o

Initialize Pretrain Model Information, Dowload weights, load weights, set variables