yoch

Another (seems related) problem, is that if I set the whole rrule before current time, it's not scheduled at all. ```python dt = datetime.now() - timedelta(minutes=15) schedule = rrule('MINUTELY', dtstart=dt,...

@concreted: Thank you ! Your workaround works for me in the first case, but not for the second one (only the first run was fired).

@concreted : Yes, I think all these occurrences should fire almost immediately, or at least one at each beat. In fact, the more realistic case is when for some reason...

@ivallesp : did you tried to play with the `std_coeff` parameter ? Somoclu uses 0.5 by default, but SOM toolbox uses 1. Also, can you indicate more precisely what results...

@ivallesp In my latest experiments with my [sparse-som](https://github.com/yoch/sparse-som) tool, which seems to suffer from the same problem, I noticed that topographic error decrease nicely during the first iterations, but abruply...

Note that the code you provided uses **variance normalization**, which improve the results greatly. But the problem persists for me (I guess it's the same with somoclu, but I haven't...

@defnull This is definitively very annoying, especially for CORS gestion (for example, all errors are wrongly reported as CORS problems). Can you provide a fix for 0.12 version please ?

I've tested the `defnull-12-1125` branch, and I can confirm that the issue is fixed. My python version is 3.6 I also ran the tests with `python3 test/testall.py`, and here is...

I haven't created new tests yet, but here some examples: ```pycon >>> from dateutil.rrule import * >>> from datetime import datetime >>> from pprint import pprint >>> start_date = datetime(2014,...

Important remark: this SKIP doesn't cover all the possibles cases. The RFC give these cases : > 1. The start date of the recurrence is a leap day in the...