...
In general, we have a problem with the cadence: as far as I understood, the data are labeled w.r.t. cadences, i.e.: “24” ,”12”, and so on. In my experience, when I set the cadence to “24”, the downloaded data are always the same, regardless the starting time.
We have two solutions:
1) every time we download the data at 1h cadence and then we select the closest hour to the one set in “starting time”. I have some code for that, I used it to create the 24 datasets with 1 hour cadence;
2) the data labels are changed, in order to download the data according to both cadence and starting time.
Personally, I prefer the second option: the first one is a work-around that, once carved in the flarecast_engine, could potentially bring problems (I am thinking about the 12 minutes cadence). Moreover, downloading the data at 1 h cadence each time is computationally demanding.
What do you guys think?
dataset": { "cadence":"24h",...