fix renaming for test fails

Author: dario-coscia

docs/source/_rst/loss/ntk_weighting.rst

@@ -4,6 +4,6 @@ NeuralTangentKernelWeighting
 
 .. automodule:: pina.loss.ntk_weighting
 
-.. autoclass:: NeuralTangetKernelWeighting
+.. autoclass:: NeuralTangentKernelWeighting
    :members:
    :show-inheritance:

pina/loss/__init__.py

@@ -8,12 +8,12 @@
     "PowerLoss",
     "WeightingInterface",
     "ScalarWeighting",
-    "NeuralTangetKernelWeighting",
+    "NeuralTangentKernelWeighting",
 ]
 
 from .loss_interface import LossInterface
 from .power_loss import PowerLoss
 from .lp_loss import LpLoss
 from .weighting_interface import WeightingInterface
 from .scalar_weighting import ScalarWeighting
-from .ntk_weighting import NeuralTangetKernelWeighting
+from .ntk_weighting import NeuralTangentKernelWeighting

tests/test_weighting/test_ntk_weighting.py

@@ -3,7 +3,7 @@
 from pina.solver import PINN
 from pina.model import FeedForward
 from pina.problem.zoo import Poisson2DSquareProblem
-from pina.loss import NeuralTangetKernelWeighting
+from pina.loss import NeuralTangentKernelWeighting
 
 problem = Poisson2DSquareProblem()
 condition_names = problem.conditions.keys()
@@ -21,13 +21,13 @@
     ],
 )
 def test_constructor(model, alpha):
-    NeuralTangetKernelWeighting(model=model, alpha=alpha)
+    NeuralTangentKernelWeighting(model=model, alpha=alpha)
 
 
 @pytest.mark.parametrize("model", [0.5])
 def test_wrong_constructor1(model):
     with pytest.raises(ValueError):
-        NeuralTangetKernelWeighting(model)
+        NeuralTangentKernelWeighting(model)
 
 
 @pytest.mark.parametrize(
@@ -43,7 +43,7 @@ def test_wrong_constructor1(model):
 )
 def test_wrong_constructor2(model, alpha):
     with pytest.raises(ValueError):
-        NeuralTangetKernelWeighting(model, alpha)
+        NeuralTangentKernelWeighting(model, alpha)
 
 
 @pytest.mark.parametrize(
@@ -58,7 +58,7 @@ def test_wrong_constructor2(model, alpha):
     ],
 )
 def test_train_aggregation(model, alpha):
-    weighting = NeuralTangetKernelWeighting(model=model, alpha=alpha)
+    weighting = NeuralTangentKernelWeighting(model=model, alpha=alpha)
     problem.discretise_domain(50)
     solver = PINN(problem=problem, model=model, weighting=weighting)
     trainer = Trainer(solver=solver, max_epochs=5, accelerator="cpu")