Question about Train, Test loss, and Test Metric

[dapyrate]

Hi,
I ran Poisson on L-shaped domain code using DeepXDE.
As you can see for step 0, 1000, 2000, and ... I get a list of loss Train and Test loss numbers of length 2.
For example for step 0 I get [1.26e+00, 2.07e-01] for train loss and [1.26+00, 2.07e-01] for test loss. Should I not get a scalar instead of a list of numbers for train and test loss? What is the first and second component in the list representing exactly? Are they representing boundary conditions loss? (How can I find out which component represents which boundary condition?)
In addition, why is test metric [] (an empty list)? What is test metric exactly representing? I also could not find any explanation in the documentation.

Thanks in advance

[lululxvi]

See FAQ.