Open gnarw opened 6 years ago
Hello, From previous of that error (here https://github.com/bulik/ldsc/issues/93 and here https://groups.google.com/d/msg/ldsc_users/9Mxw4vAB2l8/Muc_vtmYAwAJ) is appears to be caused by changes in pandas 0.21. The current solution is to revert pandas to e.g. version 0.20.1.
@tpoterba: Can we look at whether this is fixable to support pandas 0.21 without breaking backwards compatibility?
Cheers, Raymond
On Feb 13, 2018, at 6:32 AM, gnarw notifications@github.com wrote:
I´m getting errors from munge_sumstats I have recently updated some of the python packages, but I´m using python 2.7 The log with the error is not helping me finding the issue. (Attached file). iss.txt https://github.com/bulik/ldsc/files/1719701/iss.txt Any ideas ?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/bulik/ldsc/issues/104, or mute the thread https://github.com/notifications/unsubscribe-auth/AILEvcZpOU1Qfdjd4mAEhZgNhVnyVM95ks5tUXK6gaJpZM4SDnId.
I don't think this makes much sense, since it'll just break in the next version. I think the current plan is to release a conda environment with the known passing versions (which I think is pandas 0.17).
Reverting to pandas 0.18.1 solved this issue. I had to revert to 0.18.1, else there were also problems when estimating the genetic correlation.
I´m getting errors from munge_sumstats I have recently updated some of the python packages, but I´m using python 2.7 The log with the error is not helping me finding the issue. (Attached file). iss.txt
Any ideas ?