Closed Toku11 closed 1 week ago
Hi @Toku11 ,
I have tested the code snippets from APIs and the dtype should be int64 for the cases output_mode
with one_hot
multi_hot
and count
. The documentation needs to be changed to int64
instead of float32
. Thanks!
Documentation states we should expected float32 when using 'multi_hot', however int64 tensor is being returned
tf version: 2.16.1 <tf.Tensor: shape=(2, 5), dtype=int64, numpy= array([[0, 1, 0, 1, 1], [1, 0, 1, 0, 1]])>