[TFT] TFT.feature_importances() returns error when there is no futr_exog_list

[jeanboy44]

What happened + What you expected to happen

What happened:

• When I call feature_importances(), this error occurs.
• It happens when futr_exog_list is None or []

What I expected.
.feature_importances() method returns expected features

I assume that this error occurs because of line 504 in tft module.

Versions / Dependencies

neuralforecast==1.7.5

Reproduction script

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
from neuralforecast import NeuralForecast
from neuralforecast.models import TFT
from neuralforecast.losses.pytorch import DistributionLoss
from neuralforecast.utils import AirPassengersPanel, AirPassengersStatic

AirPassengersPanel['month']=AirPassengersPanel.ds.dt.month
Y_train_df = AirPassengersPanel[
    AirPassengersPanel.ds < AirPassengersPanel["ds"].values[-24]
]  # 120 train
Y_test_df = AirPassengersPanel[
    (AirPassengersPanel.ds >= AirPassengersPanel["ds"].values[-24])
    & (AirPassengersPanel.ds < AirPassengersPanel["ds"].values[-12])
].reset_index(
    drop=True
)  # 12 test
nf = NeuralForecast(
    models=[
        TFT(
            h=12,
            input_size=48,
            hidden_size=20,
            loss=DistributionLoss(distribution="StudentT", level=[80, 90]),
            learning_rate=0.005,
            stat_exog_list=[],
            futr_exog_list=[],
            hist_exog_list=['trend'],
            max_steps=300,
            val_check_steps=10,
            early_stop_patience_steps=10,
            scaler_type="robust",
            windows_batch_size=None,
            enable_progress_bar=True,
        ),
    ],
    freq="M",
)
nf.fit(df=Y_train_df, static_df=AirPassengersStatic, val_size=12)
Y_hat_df = nf.predict(futr_df=Y_test_df)

# Plot quantile predictions
Y_hat_df = Y_hat_df.reset_index(drop=False).drop(columns=["unique_id", "ds"])
plot_df = pd.concat([Y_test_df, Y_hat_df], axis=1)
plot_df = pd.concat([Y_train_df, plot_df])

plot_df = plot_df[plot_df.unique_id == "Airline1"].drop("unique_id", axis=1)
plt.plot(plot_df["ds"], plot_df["y"], c="black", label="True")
plt.plot(plot_df["ds"], plot_df["TFT"], c="purple", label="mean")
plt.plot(plot_df["ds"], plot_df["TFT-median"], c="blue", label="median")
plt.fill_between(
    x=plot_df["ds"][-12:],
    y1=plot_df["TFT-lo-90"][-12:].values,
    y2=plot_df["TFT-hi-90"][-12:].values,
    alpha=0.4,
    label="level 90",
)
plt.legend()
plt.grid()
plt.plot()

Y_hat_df = nf.predict(futr_df=Y_test_df)
feature_importances = nf.models[0].feature_importances()

Issue Severity

High: It blocks me from completing my task.

[elephaint]

Thanks, I've fixed the error and filed a PR for it