index_col = 0 issue in spacebench/env.py

[fresleven]

whenever I use the healthd_dmgrcs_mortality_disc dataset, the index_col=0 on lines 298 and 300 of spacebench/env.py makes it so that it doesn't include treatment in the list of columns on data. FYI, there isn't any index column in the datasets, so I don't know why this is here? this causes an error on line 317.

it seems that the newest versions of the datasets on dataverse don't contain the indexes on the first column. could you please fix this?

[jckitch]

Hi @fresleven, thanks for bringing this up! This is indeed because of a mismatch between the code and the datasets on Dataverse. We're re-generating the datasets now, and should have the new ones up on Dataverse soon (within a week if not sooner). Will update here once that's complete.

[fresleven]

Thank you for the response! I just wanted to follow up since it's been a little over a week—do you have an estimate on when the new datasets will be available? Really appreciate your help!

[jckitch]

Hi @fresleven thank you for your patience! The new datasets have been run, just doing some quality checks on them, but they should be posted by the end of the week at the latest. Hope that's helpful.

[jckitch]

Just an update that the new datasets have been uploaded to Harvard dataverse, but we are solving a few remaining compatibility issues with env.py. Working on that currently and should be resolved soon.

[fresleven]

Hi, Thank you so much for the update and the new version! I also wanted to let you know that while I was plotting the PM2.5 synthetic data, I found that the PM2.5 and its components were misaligned. In the 106_pm25comps.ipynb in space-data/notebooks, there is this line: df["row"], df["col"] = zip( *[~src.transform * (x, y) for x, y in zip(df["x"], df["y"])] ) the df["row"] and df["col"] should be switched, since x corresponds to col and y corresponds to row. Just wanted to let you know about this, since it may change some of your results. Best, Ayush

…

On Sat, Jun 28, 2025 at 9:10 AM James Kitch ***@***.***> wrote: *jckitch* left a comment (NSAPH-Projects/space#160) <#160 (comment)> Just an update that the new datasets have been uploaded to Harvard dataverse, but we are solving a few remaining compatibility issues with env.py. Working on that currently and should be resolved soon. — Reply to this email directly, view it on GitHub <#160 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXQN3G2EQC4BEAAXNJFWYUL3F2O3VAVCNFSM6AAAAAB6QM5RVWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTAMJVGMYDSOJSHE> . You are receiving this because you were mentioned.Message ID: ***@***.***>