araujolma / SOAR

7 stars 5 forks source link

Rejections screw up the plots' frequencies #96

Open araujolma opened 5 years ago

araujolma commented 5 years ago

Although the rejected grad steps still theoretically count as grad steps, whenever there are multiple rejections, the interf module seems to stop checking the conditions for new plots. Sometimes this means hundreds of grad steps and no plot to update the situation of the convergence process.

Let's fix this!