GridSearchCV and PyTorch with skorch shows error 'Invalid parameter lr'

I want to use sklearn's GridSearchCV in combination with PyTorch and use Skorch for compatibility. However, I receive an error telling me that lr is not a valid parameter. However, if I look up the available parameters lr is definitely in the list. You can find the full Traceback below.

net.get_params().keys()
        dict_keys(['module', 'criterion', 'optimizer', 'lr', 'max_epochs', 'batch_size',

And here is my code:

def create_NN_model(neurons):
    hidden_n = neurons
    model_NN = torch.nn.Sequential(
        torch.nn.Linear(X.shape[1], hidden_n),
        torch.nn.Dropout(0.5),
        torch.nn.ReLU(),
        torch.nn.Linear(hidden_n, hidden_n),
        torch.nn.Dropout(0.5),
        torch.nn.ReLU(),
        torch.nn.Linear(hidden_n, 1),
        )
    return model_NN

model_NN = create_NN_model(300)
net = NeuralNetRegressor(model_NN,
                         max_epochs=100,
                         lr=0.001,
                         verbose=1)


steps_NN = [('scaler', StandardScaler()), ('Net', net)]
pipeline_NN = Pipeline(steps_NN)
parameters_NN = {'lr': [0.001,0.005],'max_epochs': list(range(400,500))}

grid_NN = GridSearchCV(pipeline_NN, param_grid= parameters_NN, cv=5, verbose=1, scoring='neg_median_absolute_error')

grid_NN.fit(X_train, y_train)
print(grid_NN.score(X_test, y_test))
pred = grid_NN.best_estimator_.predict(X_test)

Traceback:

Traceback (most recent call last):
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\model_selection\_search.py", line 710, in fit
    self._run_search(evaluate_candidates)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\model_selection\_search.py", line 1151, in _run_search
    evaluate_candidates(ParameterGrid(self.param_grid))
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\model_selection\_search.py", line 689, in evaluate_candidates
    cv.split(X, y, groups)))
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\parallel.py", line 1004, in __call__
    if self.dispatch_one_batch(iterator):
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\parallel.py", line 835, in dispatch_one_batch
    self._dispatch(tasks)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\parallel.py", line 754, in _dispatch
    job = self._backend.apply_async(batch, callback=cb)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\_parallel_backends.py", line 209, in apply_async
    result = ImmediateResult(func)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\_parallel_backends.py", line 590, in __init__
    self.results = batch()
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\parallel.py", line 256, in __call__
    for func, args, kwargs in self.items]
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\joblib\parallel.py", line 256, in <listcomp>
    for func, args, kwargs in self.items]
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\model_selection\_validation.py", line 504, in _fit_and_score
    estimator = estimator.set_params(**cloned_parameters)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\pipeline.py", line 163, in set_params
    self._set_params('steps', **kwargs)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\utils\metaestimators.py", line 50, in _set_params
    super().set_params(**params)
  File "C:\Users\Maximilian\Anaconda3\envs\LisbonHousing\lib\site-packages\sklearn\base.py", line 236, in set_params
    (key, self))
ValueError: Invalid parameter lr for estimator Pipeline(memory=None,
         steps=[('scaler',
                 StandardScaler(copy=True, with_mean=True, with_std=True)),
                ('Net',
                 <class 'skorch.regressor.NeuralNetRegressor'>[uninitialized](
  module=Sequential(
    (0): Linear(in_features=85, out_features=300, bias=True)
    (1): Dropout(p=0.5)
    (2): ReLU()
    (3): Linear(in_features=300, out_features=300, bias=True)
    (4): Dropout(p=0.5)
    (5): ReLU()
    (6): Linear(in_features=300, out_features=1, bias=True)
  ),
))],
         verbose=False). Check the list of available parameters with `estimator.get_params().keys()`.