uppercase instead of underscore

evidential_deep_learning/losses/continuous.py

@@ -62,7 +62,7 @@ def NIG_Reg(y, gamma, v, alpha, beta, omega=0.01, reduce=True, kl=False):
 
     return tf.reduce_mean(reg) if reduce else reg
 
-def evidential_regression(y_true, evidential_output, coeff=1.0):
+def EvidentialRegression(y_true, evidential_output, coeff=1.0):
     gamma, v, alpha, beta = tf.split(evidential_output, 4, axis=-1)
     loss_nll = NIG_NLL(y_true, gamma, v, alpha, beta)
     loss_reg = NIG_Reg(y_true, gamma, v, alpha, beta)

hello_world.py

@@ -19,13 +19,13 @@ def main():
     ])
 
     # Custom loss function to handle the custom regularizer coefficient
-    def evidential_regression_loss(true, pred):
-        return edl.losses.evidential_regression(true, pred, coeff=1e-2)
+    def EvidentialRegressionLoss(true, pred):
+        return edl.losses.EvidentialRegression(true, pred, coeff=1e-2)
 
     # Compile and fit the model!
     model.compile(
         optimizer=tf.keras.optimizers.Adam(5e-4),
-        loss=evidential_regression_loss)
+        loss=EvidentialRegressionLoss)
     model.fit(x_train, y_train, batch_size=100, epochs=500)
 
     # Predict and plot using the trained model