ACCLAB / DABEST-python

Data Analysis with Bootstrapped ESTimation
https://acclab.github.io/DABEST-python/
Apache License 2.0
341 stars 47 forks source link

Does not run with more recent versions of dependencies #60

Closed kebarr closed 4 years ago

kebarr commented 5 years ago

I have tried to run the test code from the README with the example dataset and get an error (TypeError: must be real number, not list) on the line:

iris_dabest.mean_diff.plot(), and occurs due to this line in the gapped_lines function:

--> 162 quantiles = data.groupby(x)[y].quantile([0.25, 0.75])\ 163 .unstack()\ 164 .reindex(index=group_order)

I am using Python3.7 with the following modules:

scipy 1.3.0 seaborn 0.9.0 pandas 0.25.0 numpy 1.17.0 matplotlib 3.1.1

which I notice are more recent versions of the dependencies. I installed dabest using pip. I have also built the cloned version to test and get the same error.

I have forked a copy of the code to see if there is a simple fix.

josesho commented 5 years ago

Thanks for this notice; I do need to track if dabest plays well with the latest update to numpy and matplotlib. Did you close this because this has been resolved?

kebarr commented 5 years ago

I closed it because I found a similar issue during later debugging, and the resolution for that worked here- using pandas 0.24. I have not changed any of the other dependencies so so far it works with later versions of NumPy and matplotlib.

josesho commented 4 years ago

Closed with #89