typos

pts/dataset/__init__.py

@@ -1 +1 @@
-from .loader import TransformedIterableDataset
+from .loader import TransformedIterableDataset

pts/dataset/repository/__init__.py

@@ -1 +1 @@
-from .datasets import dataset_recipes
+from .datasets import dataset_recipes

pts/feature/fourier_date_feature.py

@@ -47,6 +47,7 @@ class FourierDateFeatures(TimeFeature):
         steps = [x * 2.0 * np.pi / num_values for x in values]
         return np.vstack([np.cos(steps), np.sin(steps)])
 
+
 def fourier_time_features_from_frequency(freq_str: str) -> List[TimeFeature]:
     offset = to_offset(freq_str)
     multiple, granularity = offset.n, offset.name
@@ -66,4 +67,4 @@ def fourier_time_features_from_frequency(freq_str: str) -> List[TimeFeature]:
     feature_classes: List[TimeFeature] = [
         FourierDateFeatures(freq=freq) for freq in features[granularity]
     ]
-    return feature_classes
+    return feature_classes

pts/model/deepvar/deepvar_estimator.py

@@ -32,7 +32,7 @@ from pts import Trainer
 from pts.model.utils import get_module_forward_input_names
 from pts.feature import (
     fourier_time_features_from_frequency,
-    lags_for_fourier_time_features_from_frequency
+    lags_for_fourier_time_features_from_frequency,
 )
 from pts.model import PyTorchEstimator
 from pts.modules import LowRankMultivariateNormalOutput

pts/model/tempflow/tempflow_estimator.py

@@ -26,7 +26,7 @@ from gluonts.transform import (
 from pts import Trainer
 from pts.feature import (
     fourier_time_features_from_frequency,
-    lags_for_fourier_time_features_from_frequency
+    lags_for_fourier_time_features_from_frequency,
 )
 from pts.model.utils import get_module_forward_input_names
 from pts.model import PyTorchEstimator

pts/model/transformer/transformer_estimator.py

@@ -10,7 +10,7 @@ from gluonts.torch.modules.distribution_output import DistributionOutput
 from gluonts.torch.support.util import copy_parameters
 from gluonts.torch.model.predictor import PyTorchPredictor
 from gluonts.model.predictor import Predictor
-from gluonts.transform  import (
+from gluonts.transform import (
     Transformation,
     Chain,
     InstanceSplitter,
@@ -205,7 +205,7 @@ class TransformerEstimator(PyTorchEstimator):
     def create_predictor(
         self,
         transformation: Transformation,
-        trained_network:TransformerTrainingNetwork,
+        trained_network: TransformerTrainingNetwork,
         device: torch.device,
     ) -> Predictor:
 

pts/model/transformer_tempflow/transformer_tempflow_network.py

@@ -3,8 +3,7 @@ from typing import List, Optional, Tuple
 import torch
 import torch.nn as nn
 
-from gluonts.core.component import 
-
+from gluonts.core.component import validated
 from pts.modules import RealNVP, MAF, FlowOutput, MeanScaler, NOPScaler
 
 

pts/model/utils.py

@@ -35,7 +35,11 @@ def weighted_average(
     """
     if weights is not None:
         weighted_tensor = torch.where(weights != 0, x * weights, torch.zeros_like(x))
-        sum_weights = torch.clamp(weights.sum(dim=dim) if dim else weights.sum(), min=1.0)
-        return (weighted_tensor.sum(dim=dim) if dim else weighted_tensor.sum())/ sum_weights
+        sum_weights = torch.clamp(
+            weights.sum(dim=dim) if dim else weights.sum(), min=1.0
+        )
+        return (
+            weighted_tensor.sum(dim=dim) if dim else weighted_tensor.sum()
+        ) / sum_weights
     else:
         return x.mean(dim=dim)