Quiz question: How do you set the learning rate of the SGD optimizer?

Course Q&A TensorFlow Developer Professional Sequences, Time Series and Prediction
1

There are four options in this question.
According to the documentation on tensorflow.org, the the correct argument is learning_rate, which can be set with a floating-point value.

In the quiz, the options are:

  • You can’t set it

  • Use the RateOfLearning property

  • Use the lr property

  • Use the Rate property

None of these appear to be correct. Is this intentional?

1 Like
2

There is a valid but legacy option in the choices. That said, I agree with you that this quiz question could rely on current values.

3
  • Rename lr to learning_rate for all optimizers.
1 Like
4

Yeah, probably this question should be updated then. I have to admit, I don’t check the legacy documentation, because it can easily confuse me.

1 Like
5

I feel your pain, but it’s a reality of life in this world. You will frequently run into cases where one package or another has evolved in a non-backwards-compatible way that breaks some legacy code. It’s a good skill to have in the toolbox to be able to sleuth these out.

6

I hope the issue is sorted sir

7

I think this should be appear in the latest update of course 4.