Replies: 2 comments 3 replies
-
@cwasicki does this correspond somewhat to what we discussed? And does this interface look usable? |
Beta Was this translation helpful? Give feedback.
2 replies
-
Also @llucax and @matthias-wende-frequenz, maybe you have some comments as well, I will use experience from this design when presenting the client interface to everyone on Monday. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Installing the client
For now, we have to add the api package as a dependency
pip install "frequenz-api-weather @ git+https://github.com/frequenz-floss/[email protected]"
or once there's a release,
pip install "frequenz-api-weather == 0.1.0"
Import
Initialization
Live data streaming method
The
stream_live_forecasts
method starts a task which,ReceiveLiveWeatherForecast
Forecast
objects and puts them into a channelSubsequent calls to
stream_live_forecasts
with the same argumentswould just reuse the or
Using the streamed forecasts
Objects on the stream are opaque, and we can reach the data only
through methods.
to_ndarray_vlf
for example would return a 3d numpy array, withdimensions in this order:
This method can take optional filters for each of these dimensions
Beta Was this translation helpful? Give feedback.
All reactions