Fix incorrect input routing for models (awslabs#3186)

Fixes awslabs#3185

*Description of changes:*

There is currently a bug where the model inputs may be routed incorrect
by the forecast generator. This effectively results in
`past_feat_dynamic_real` and `past_feat_dynamic_cat` being ignored by
the TFT model.

MWE:
```python
from unittest import mock
import numpy as np
import pandas as pd
from gluonts.torch.model.tft import TemporalFusionTransformerEstimator

freq = "D"
N = 50
data = [
    {"target": np.arange(N), "past_feat_dynamic_real": np.random.rand(1, N).astype("float32"), "start": pd.Period("2020-01-01", freq=freq)}
]

predictor = TemporalFusionTransformerEstimator(prediction_length=1, freq=freq, past_dynamic_dims=[1], trainer_kwargs={"max_epochs": 1}).train(data)

with mock.patch("gluonts.torch.model.tft.module.TemporalFusionTransformerModel._preprocess") as mock_fwd:
    try:
        fcst = list(predictor.predict(data))
    except:
        pass   
    call_kwargs = mock_fwd.call_args[1]

call_kwargs["feat_dynamic_cat"]  
# tensor([[[0.8073]]])
call_kwargs["past_feat_dynamic_real"]  
# None
```

The bug occurs because model inputs are passed as positional arguments
instead of keyword arguments.


By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice.


**Please tag this pr with at least one of these labels to make our
release process faster:** BREAKING, new feature, bug fix, other change,
dev setup

src/gluonts/model/forecast_generator.py

@@ -82,6 +82,15 @@ def make_distribution_forecast(distr, *args, **kwargs) -> Forecast:
     raise NotImplementedError
 
 
+def make_predictions(prediction_net, inputs: dict):
+    # MXNet predictors only support positional arguments
+    class_name = prediction_net.__class__.__module__
+    if class_name.startswith("gluonts.mx") or class_name.startswith("mxnet"):
+        return prediction_net(*inputs.values())
+    else:
+        return prediction_net(**inputs)
+
+
 class ForecastGenerator:
     """
     Classes used to bring the output of a network into a class.
@@ -115,7 +124,7 @@ def __call__(
     ) -> Iterator[Forecast]:
         for batch in inference_data_loader:
             inputs = select(input_names, batch, ignore_missing=True)
-            (outputs,), loc, scale = prediction_net(*inputs.values())
+            (outputs,), loc, scale = make_predictions(prediction_net, inputs)
             outputs = to_numpy(outputs)
             if scale is not None:
                 outputs = outputs * to_numpy(scale[..., None])
@@ -159,14 +168,16 @@ def __call__(
     ) -> Iterator[Forecast]:
         for batch in inference_data_loader:
             inputs = select(input_names, batch, ignore_missing=True)
-            outputs = to_numpy(prediction_net(*inputs.values()))
+            outputs = to_numpy(make_predictions(prediction_net, inputs))
             if output_transform is not None:
                 outputs = output_transform(batch, outputs)
             if num_samples:
                 num_collected_samples = outputs[0].shape[0]
                 collected_samples = [outputs]
                 while num_collected_samples < num_samples:
-                    outputs = to_numpy(prediction_net(*inputs.values()))
+                    outputs = to_numpy(
+                        make_predictions(prediction_net, inputs)
+                    )
                     if output_transform is not None:
                         outputs = output_transform(batch, outputs)
                     collected_samples.append(outputs)
@@ -209,7 +220,7 @@ def __call__(
     ) -> Iterator[Forecast]:
         for batch in inference_data_loader:
             inputs = select(input_names, batch, ignore_missing=True)
-            outputs = prediction_net(*inputs.values())
+            outputs = make_predictions(prediction_net, inputs)
 
             if output_transform:
                 log_once(OUTPUT_TRANSFORM_NOT_SUPPORTED_MSG)