Categorical_crossentropy vs CategoricalCrossentropy

Good morning!
I’m writing this question according to the DLS Course 2 because the softmax regression has been discussed in the week 3. In the TensorFlow documentation there are two loss functions: categorical_crossentropy and CategoricalCrossentropy. I still have not figured out the sticker with that two terms. I wonder which loss function has been written in the course (for categorical_crossentropy or for CategoricalCrossentropy). Your support means a lot to me. Best wishes!

@przemyslaw
The two-loss functions that you mentioned are equivalent.
categorical_crossentropy is the function notation and CategoricalCrossentropy is the class notation.
The class notation enables you to pass configuration arguments to the constructor (e.g. loss_fn = CategoricalCrossentropy(from_logits=True) ).

Link to keras documentation: Losses

2 Likes