Strange Performance of the Code

[Umaruchain]

Hi, thanks for your great work!
I have downloaded your code directly, however, I obtained the following performance:
Average CPC of test tiles: 0.4445 stdev: 0.1901.
It is very different from the performance reported by the paper, right?

[MassimilianoLuca]

Hi! Would you be willing to share additional information on the data you used? Which OD matrices did you use?

Also, in the main paper we report the performances for each population decile, which are sometimes lower wrt 0.4445 and sometimes higher. However, if you look at the average performance of the model -- depending on the country you are referring to -- I would say your performances are in line with the ones of the paper