Update tf_explain_mnist.ipynb

sicara / tf-explain

Interpretability Methods for tf.keras models with Tensorflow 2.x

https://tf-explain.readthedocs.io

MIT License

1.02k stars 111 forks source link

Update tf_explain_mnist.ipynb #101

Closed twsl closed 4 years ago

twsl commented 4 years ago

Use np.argmax to decode One Hot Encoded Class Labels

RaphaelMeudec commented 4 years ago

@twsI Thanks for the cleaning! Could you also add this modification to the examples/callbacks/mnist.py? I believe it also uses an ugly construction of the one hot vector

twsl commented 4 years ago

@RaphaelMeudec yeah, you are correct, it uses the same way of checking. Fixed.

RaphaelMeudec commented 4 years ago

@twsI Thanks for the correction!