Update on comments

Lightning-Universe · Oct 29, 2021 · 3cbd13e · 3cbd13e
1 parent 9c213d6
commit 3cbd13e
Show file tree

Hide file tree

Showing 4 changed files with 1 addition and 3 deletions.
diff --git a/flash/core/integrations/pytorch_forecasting/adapter.py b/flash/core/integrations/pytorch_forecasting/adapter.py
@@ -75,6 +75,7 @@ def from_task(
         **backbone_kwargs,
     ) -> Adapter:
         parameters = copy(parameters)
+        # Remove the single row of data from the parameters to reconstruct the `time_series_dataset`
         data = parameters.pop("data_sample")
         time_series_dataset = PatchTimeSeriesDataSet.from_parameters(parameters, data)
 

diff --git a/flash_examples/integrations/pytorch_forecasting/tabular_forecasting_interpretable.py b/flash_examples/integrations/pytorch_forecasting/tabular_forecasting_interpretable.py
@@ -28,7 +28,6 @@
 # Example based on this tutorial: https://pytorch-forecasting.readthedocs.io/en/latest/tutorials/ar.html
 # 1. Create the DataModule
 data = generate_ar_data(seasonality=10.0, timesteps=400, n_series=100, seed=42)
-data["static"] = 2
 data["date"] = pd.Timestamp("2020-01-01") + pd.to_timedelta(data.time_idx, "D")
 
 max_prediction_length = 20

diff --git a/flash_examples/tabular_forecasting.py b/flash_examples/tabular_forecasting.py
@@ -26,7 +26,6 @@
 # Example based on this tutorial: https://pytorch-forecasting.readthedocs.io/en/latest/tutorials/ar.html
 # 1. Create the DataModule
 data = generate_ar_data(seasonality=10.0, timesteps=400, n_series=100, seed=42)
-data["static"] = 2
 data["date"] = pd.Timestamp("2020-01-01") + pd.to_timedelta(data.time_idx, "D")
 
 max_prediction_length = 20

diff --git a/tests/tabular/forecasting/test_model.py b/tests/tabular/forecasting/test_model.py
@@ -29,7 +29,6 @@
 @pytest.fixture
 def sample_data():
     data = generate_ar_data(seasonality=10.0, timesteps=100, n_series=2, seed=42)
-    data["static"] = 2
     data["date"] = pd.Timestamp("2020-01-01") + pd.to_timedelta(data.time_idx, "D")
     max_prediction_length = 20
     training_cutoff = data["time_idx"].max() - max_prediction_length