In Keras, the first object returned in the score list is loss.

score = model.evaluate(X_test, y_test,verbose=1)

[1] [0.025217213829228164, 0.99487179487179489]

Is this the same thing as variance?


1 Answer 1


No. Loss measures the error between your predicted values and true values in a given train set. Whereas variance mesures how your model performs (i.e. how the loss changes) on different training sets. Also, variance is an atribute of a given model whereas loss depends on your dataset.

In other words, we can also say that loss on different sets is used to measure variance of a given model.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.