I'm using keras with last layer as softmax which returns a probability (0-1.0) float values. Suppose I have 4 classes, how can I get the index of the class that has the highest probability? Is there a keras, numpy or scikit-learn function to do this?
pred = model.predict(....)
# pred = [[0.9, 0.0, 0.0, 0.1], --------> [0, 1, 1]
# [0.1, 0.8, 0.1, 0.0], change to
# [0.1, 0.8, 0.1, 0.0]]
The reason I want to change from array of 0-1 float to integer is because I want to use scikit-learn's confusion matrix to visualise the accuracy and it only accepts integer labels.
