Do you have any script to plot predicted results over input data for forecasting task? Thanks

[Thanh-Binh]

No description provided.

[zhihanyue]

No. The output file is at training/DatasetName__RunName_Date_Time/out.pkl. You can easily load this file with pickle and plot the output.

[Thanh-Binh]

Thanks
What is the format of data?
After loading my data has the length of 5
What does it mean? Number of prediction time steps?

[zhihanyue]

Thanks
What is the format of data?
After loading my data has the length of 5
What does it mean? Number of prediction time steps?

It means the prediction horizon (e.g. 24,48,168,336,720). Check the code in /tasks/forecasting.py for more details.

        out_log[pred_len] = {
            'norm': test_pred,
            'raw': test_pred_inv,
            'norm_gt': test_labels,
            'raw_gt': test_labels_inv
        }

[Thanh-Binh]

I have just look at your source code.
After training I want to use the trained model, e.g. for prediction of time series x0,x1, x2, ... xN by prediction horizon of 10.
How can I call it in the loop?
pseudo-code look like:
for i = 1: N
currentData = x[i]
// get prediction data xpred at (i + prediction_horizon)
end

Thanks

[ayesha1093]

Hi,
Can you share the code if you were able to figure it out?