Data incomplete in Eastern-Massachusetts

[chkwon]

@jingzbu

In EMA_net.tntp, Free Flow Time, B, and Power are all zeroes now. These values need to be identified in the data file.

Also, Type needs to be integers, I believe.

[jingzbu]

@chkwon
The Free Flow Time are positive; please see the correct column. The cost function is specified in EMA_intro.pdf. The Type plays no essential role and I just put a zero value.

[chkwon]

I understand that the PDF file contains information, but computer codes usually don't read PDF files; only read TNTP data file. Anyway, you need to specify all values to follow the consistent format of TNTP.

I don't understand what role Type plays, but in all other TNTP data files, Type columns are integers like 1 or 9, instead of 0.00000. This format inconsistency makes reading and parsing data files rather difficult. If you want to put zero, please use 0, not 0.00000.

[jingzbu]

@chkwon
Have updated and created a pull request. You can use the typical BPR cost function for testing. The one estimated by inverseVI is a bit tedious and can be cited for purposes of writing a paper.

[chkwon]

Thanks!