[BUG] remove zero from pooling_sizes and downsample_frequencies (#1577)

This PR partly fixes #1571 for NHiTS.

If `max_prediction_length = 1`,  pooling_sizes and downsample_frequencies will contain zero, which causes divided by zero error. This PR replace zero by one.

pytorch_forecasting/models/nhits/__init__.py

@@ -154,8 +154,12 @@ def __init__(
         if pooling_sizes is None:
             pooling_sizes = np.exp2(np.round(np.linspace(0.49, np.log2(prediction_length / 2), n_stacks)))
             pooling_sizes = [int(x) for x in pooling_sizes[::-1]]
+            # remove zero from pooling_sizes
+            pooling_sizes = max(pooling_sizes, [1] * len(pooling_sizes))
         if downsample_frequencies is None:
             downsample_frequencies = [min(prediction_length, int(np.power(x, 1.5))) for x in pooling_sizes]
+            # remove zero from downsample_frequencies
+            downsample_frequencies = max(downsample_frequencies, [1] * len(downsample_frequencies))
 
         # set static hidden size
         if static_hidden_size is None:

tests/test_models/test_nhits.py

@@ -4,8 +4,11 @@
 import lightning.pytorch as pl
 from lightning.pytorch.callbacks import EarlyStopping
 from lightning.pytorch.loggers import TensorBoardLogger
+import numpy as np
+import pandas as pd
 import pytest
 
+from pytorch_forecasting.data.timeseries import TimeSeriesDataSet
 from pytorch_forecasting.metrics import MQF2DistributionLoss, QuantileLoss
 from pytorch_forecasting.metrics.distributions import ImplicitQuantileNetworkDistributionLoss
 from pytorch_forecasting.models import NHiTS
@@ -70,7 +73,12 @@ def _integration(dataloader, tmp_path, trainer_kwargs=None, **kwargs):
     finally:
         shutil.rmtree(tmp_path, ignore_errors=True)
 
-    net.predict(val_dataloader, fast_dev_run=True, return_index=True, return_decoder_lengths=True)
+    net.predict(
+        val_dataloader,
+        fast_dev_run=True,
+        return_index=True,
+        return_decoder_lengths=True,
+    )
 
 
 @pytest.mark.parametrize(
@@ -143,3 +151,48 @@ def test_interpretation(model, dataloaders_with_covariates):
     raw_predictions = model.predict(dataloaders_with_covariates["val"], mode="raw", return_x=True, fast_dev_run=True)
     model.plot_prediction(raw_predictions.x, raw_predictions.output, idx=0, add_loss_to_title=True)
     model.plot_interpretation(raw_predictions.x, raw_predictions.output, idx=0)
+
+
+# Bug when max_prediction_length=1 #1571
+@pytest.mark.parametrize("max_prediction_length", [1, 5])
+def test_prediction_length(max_prediction_length: int):
+    n_timeseries = 10
+    time_points = 10
+    data = pd.DataFrame(
+        data={
+            "target": np.random.rand(time_points * n_timeseries),
+            "time_varying_known_real_1": np.random.rand(time_points * n_timeseries),
+            "time_idx": np.tile(np.arange(time_points), n_timeseries),
+            "group_id": np.repeat(np.arange(n_timeseries), time_points),
+        }
+    )
+    training_dataset = TimeSeriesDataSet(
+        data=data,
+        time_idx="time_idx",
+        target="target",
+        group_ids=["group_id"],
+        time_varying_unknown_reals=["target"],
+        time_varying_known_reals=(["time_varying_known_real_1"]),
+        max_prediction_length=max_prediction_length,
+        max_encoder_length=3,
+    )
+    training_data_loader = training_dataset.to_dataloader(train=True)
+    forecaster = NHiTS.from_dataset(training_dataset, log_val_interval=1)
+    trainer = pl.Trainer(
+        accelerator="cpu",
+        max_epochs=3,
+        min_epochs=2,
+        limit_train_batches=10,
+    )
+    trainer.fit(
+        forecaster,
+        train_dataloaders=training_data_loader,
+    )
+    validation_dataset = TimeSeriesDataSet.from_dataset(training_dataset, data, stop_randomization=True, predict=True)
+    validation_data_loader = validation_dataset.to_dataloader(train=False)
+    forecaster.predict(
+        validation_data_loader,
+        fast_dev_run=True,
+        return_index=True,
+        return_decoder_lengths=True,
+    )