Where can I define what variable should be predicted?

[Larosen]

I found the following code and it works fine so far.
https://github.com/keras-team/keras-io/blob/master/examples/timeseries/timeseries_weather_forecasting.py
With Output: https://keras.io/examples/timeseries/timeseries_weather_forecasting/

In line 214 and 216 I determine which values it should use when trancing the model. So e.g. wind speed, humidity, temperature and rain amount.

From line 355 on, the predictions and the actual values can be displayed.

But where do I see which values he predicted there in the graph? The heading only says "Single Step Prediction" without any reference to the variable.
Was the temperature or the humidity or randomly always something else predicted here?

I would have to predict a certain value for a job, which is included in the variables of the training model.

Could someone please point me to the correct position?

[dhantule]

Hi @Larosen, thanks for reporting this.

The model is predicting temperature, T(degC) , in the prediction block

x[0][:, 1] – shows the historical temperature values used as input
y[0] – true temperature value 12 hours later
model.predict(x)[0] – model’s predicted temperature 12 hours later

[github-actions]

Are you satisfied with the resolution of your issue?
Yes
No