Grigorios Kalliatakis

No data provided for "input_1"

1

I've tried to replicate your `build_classification_model` using the following code ```python def build_default_classification_model(): """Instantiates the initial classification model """ input_shape = [IMG_DIM, IMG_DIM, NUM_BANDS] img_input = Input(shape=input_shape) # Block 1...

Negative loss & logits_variance_loss

4

Hi, I have created a Bayesian CNN classifier as described in this repo, but my model's loss is always negative as well as the logits_variance_loss (see screenshot below). Any idea...

Does anyone know how can I start with random initial coordinates for the first square patch?