Mean directional accuracy

[lukaszkn]

I know this is dummy beginner question but how do I calculate mean directional accuracy for stack-rnn-arima-xgb or any deep learning experiments? E.g. instead of mean square error I want to calculate mean signum difference between current and previous value. Thanks for your help

[IISuperluminaLII]

write your custom loss function that calculates the metric you want given (y_pred, y_true). Assuming this is for forecasting market data....