tft_qfc

# retrieve forecast series for chosen quantiles, 
# inverse-transform each series,
# insert them as columns in a new dataframe dfY
q50_RMSE = np.inf
q50_MAPE = np.inf
ts_q50 = None
pd.options.display.float_format = '{:,.2f}'.format
dfY = pd.DataFrame()
dfY["Actual"] = TimeSeries.pd_series(ts_test)


# helper function: get forecast values for selected quantile q and insert them in dataframe dfY
def predQ(ts_t, q):
    ts_tq = ts_t.quantile_timeseries(q)
    ts_q = scalerP.inverse_transform(ts_tq)
    s = TimeSeries.pd_series(ts_q)
    header = "Q" + format(int(q*100), "02d")
    dfY[header] = s
    if q==0.5:
        ts_q50 = ts_q
        q50_RMSE = rmse(ts_q50, ts_test)
        q50_MAPE = mape(ts_q50, ts_test) 
        print("RMSE:", f'{q50_RMSE:.2f}')
        print("MAPE:", f'{q50_MAPE:.2f}')
  
    
# call helper function predQ, once for every quantile
_ = [predQ(ts_tpred, q) for q in QUANTILES]

# move Q50 column to the left of the Actual column
col = dfY.pop("Q50")
dfY.insert(1, col.name, col)
dfY.iloc[np.r_[0:2, -2:0]]