@bstabler @e-lo @pedrocamargo
I've cut a new release of OMX, version 0.4.0b1, containing all of Pedro's team's modernization from #15. This version also should eliminate the problem identified in #11 by @jfbischoff.
It's on PyPi here: https://pypi.org/project/openmatrix/0.4.0b1/
The release was tested on the toy ActivitySim model and works great, but one of the larger ActivitySim implementations should test it now and provide feedback.
What's the best way to get this on the ActivitySim docket? I think SANDAG would be an ideal candidate. Theoretically it would involve updating the openmatrix python dependency to exactly 0.4.0b1 and running the full model, reporting back any problems. We will update with further point releases until this version of OMX is ready for full release.
There is a backlog of other ideas on improving OMX which can all come after this release is complete. =)
edit: I guess I could have posted this as a PR on activitysim instead of a plea here. Might do that if it makes more sense
@bstabler @e-lo @pedrocamargo
I've cut a new release of OMX, version 0.4.0b1, containing all of Pedro's team's modernization from #15. This version also should eliminate the problem identified in #11 by @jfbischoff.
It's on PyPi here: https://pypi.org/project/openmatrix/0.4.0b1/
The release was tested on the toy ActivitySim model and works great, but one of the larger ActivitySim implementations should test it now and provide feedback.
What's the best way to get this on the ActivitySim docket? I think SANDAG would be an ideal candidate. Theoretically it would involve updating the openmatrix python dependency to exactly
0.4.0b1and running the full model, reporting back any problems. We will update with further point releases until this version of OMX is ready for full release.There is a backlog of other ideas on improving OMX which can all come after this release is complete. =)
edit: I guess I could have posted this as a PR on activitysim instead of a plea here. Might do that if it makes more sense