mixer
mixer copied to clipboard
RuntimeWarning: divide by zero encountered in log10
Hello, thank you for developing mixer!
I've sucessfully run fit1, fit2 and the bivariate analysis. However, i've encuntered a RuntimeWarning such as this:
/home/nico/mixer/mixer/precimed/mixer/utils.py:894: RuntimeWarning: divide by zero encountered in log10
data_x=-np.log10(np.flip(np.cumsum(np.flip(data_weights[si])))) # step 2
/home/nico/.conda/envs/mixer/lib/python3.8/site-packages/scipy/interpolate/_interpolate.py:698: RuntimeWarning: invalid value encountered in subtract
slope = (y_hi - y_lo) / (x_hi - x_lo)[:, None]
INFO:root:log_message(Done)
(example from one bivariate log file)
When I try to generate the figures I run into this warning:
~/mixer/mixer/results/bivariate/mgn_msp$ python3 ~/mixer/mixer/precimed/mixer_figures.py two \
--json-fit ~/mixer/mixer/results/fit2_results/mgn_msp/fit2_mgn_msp_combined.json \
--json-test bivariate_mgn_msp_combined.json \
--out mgn_msp.outfigs \
--statistic mean std \
--trait1 MGN
--trait2 MSP
generate mgn_msp.outfigs.csv from 2 json files...
Done.
/home/nico/mixer/mixer/precimed/mixer/figures.py:264: RuntimeWarning: All-NaN slice encountered
plot_y[i, :] = plot_y[i, :] - np.nanmin(plot_y[i, :])
/home/nico/mixer/mixer/precimed/mixer/figures.py:266: RuntimeWarning: Mean of empty slice
plot_y = np.nanmean(plot_y, 0)
Generated mgn_msp.outfigs.png
--trait2: command not found
The figure is generated despite the warning (though it's not recognizing the trait2) and I'm confused as to what the issue is. Greatly appreciate any help :)