Previous topic

fynance.models.rolling._RollingBasis.__call__

Next topic

fynance.models.rolling.RollMultiLayerPerceptron

fynance.models.rolling._RollingBasis.run

_RollingBasis.run(backtest_plot=True, backtest_kpi=True, figsize=(9, 6), func=<ufunc 'sign'>)

Run neural network model and backtest predictions.

Parameters:
backtest_plot : bool, optional

If True, display plot of backtest performances.

backtest_kpi : bool, optional

If True, display kpi of backtest performances.

figsize : tuple of int, optional

Size of the figure to plot loss and performances.

func : callable, optional

Function to apply on the prediction, default is np.sign function. If func is None, then func = lambda x: x so the raw values of the prediction are used to compute returns.