Using gfs_quarter_deg_hourly_to_hourly_mean

[alisboainegi]

I am trying to use the forecast of solarforecastorbiter.

Hello!

When I run:
models.gfs_quarter_deg_hourly_to_hourly_mean(latitude, longitude, elevation, init_time, start, end, interval_label)

Where:

• latitude: - 41
• longitude: -8
• elevation: 104
• init_time: pd.Timestamp(datetime.date.today(), tz=self.tz)
• start: pd.Timestamp(datetime.date.today(), tz=self.tz)
• end: start + pd.Timedelta(days=2)
• interval_label: beggining

I get this error:
FileNotFoundError: gfs_0p25\2021\11\19\00\gfs_0p25.nc does not exist

Could you help me understand my mistake?

[wholmgren]

Hi @alisboainegi. Have you downloaded the NWP data? The functions in models.py assume you already have that data. The easiest way to download data is to use the solararbiter command line tool with the fetchnwp argument. That's easy to do if you install the package using the conda forge package on linux, but harder on windows due to dependency issues. Or use the docker image.

Then it's a matter of pointing functions in models.py to the data you've downloaded. Those functions contain a keyword argument load_forecast. That must be a function that can load the forecast data in a specific format. By default that's solarforecastarbiter.io.nwp.load_forecast. You can wrap that function while defining a new base_path keyword argument or call io.nwp.set_base_path to change the directory.