Why adding metrics

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

I remember using metrics=[‘accuracy’] in model.compile in last courses to get the accuracy. however it does not work in C4_W2_lab2. Would you please explain why it does not work here ?
Thanks,

2

C4_w2_Lab_2 refers to building a regression based time series model. Metrics like accuracy are meant for classification models. Use metrics like mae.

3

Thanks! One more question, why dense layer does not have and activation? If we don’t mention it does it mean the activation is linear ?

4

That is correct. A dense layer without activation defaults to linear.