Updated docstring

Alexander März · Alexander März · commit 40652e731486 · 2023-06-15T16:10:26.000+02:00
diff --git a/lightgbmlss/distributions/Beta.py b/lightgbmlss/distributions/Beta.py
@@ -23,7 +23,7 @@ class Beta:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/Cauchy.py b/lightgbmlss/distributions/Cauchy.py
@@ -22,7 +22,7 @@ class Cauchy:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/Gamma.py b/lightgbmlss/distributions/Gamma.py
@@ -23,7 +23,7 @@ class Gamma:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/Gaussian.py b/lightgbmlss/distributions/Gaussian.py
@@ -22,7 +22,7 @@ class Gaussian:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/Gumbel.py b/lightgbmlss/distributions/Gumbel.py
@@ -22,7 +22,7 @@ class Gumbel:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/Laplace.py b/lightgbmlss/distributions/Laplace.py
@@ -22,7 +22,7 @@ class Laplace:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/LogNormal.py b/lightgbmlss/distributions/LogNormal.py
@@ -22,7 +22,7 @@ class LogNormal:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/NegativeBinomial.py b/lightgbmlss/distributions/NegativeBinomial.py
@@ -25,7 +25,7 @@ class NegativeBinomial:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn_total_count: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp", "softplus" or "relu".
     response_fn_probs: str
         When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
diff --git a/lightgbmlss/distributions/Poisson.py b/lightgbmlss/distributions/Poisson.py
@@ -20,7 +20,7 @@ class Poisson:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp", "softplus" or "relu".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/StudentT.py b/lightgbmlss/distributions/StudentT.py
@@ -25,7 +25,7 @@ class StudentT:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/Weibull.py b/lightgbmlss/distributions/Weibull.py
@@ -23,7 +23,7 @@ class Weibull:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/ZABeta.py b/lightgbmlss/distributions/ZABeta.py
@@ -27,7 +27,7 @@ class ZABeta:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/ZAGamma.py b/lightgbmlss/distributions/ZAGamma.py
@@ -27,7 +27,7 @@ class ZAGamma:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/ZALN.py b/lightgbmlss/distributions/ZALN.py
@@ -27,7 +27,7 @@ class ZALN:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp" or "softplus".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
diff --git a/lightgbmlss/distributions/ZINB.py b/lightgbmlss/distributions/ZINB.py
@@ -25,7 +25,7 @@ class ZINB:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn_total_count: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp", "softplus" or "relu".
     response_fn_probs: str
         When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
diff --git a/lightgbmlss/distributions/ZIPoisson.py b/lightgbmlss/distributions/ZIPoisson.py
@@ -23,7 +23,7 @@ class ZIPoisson:
     stabilization: str
         Stabilization method for the Gradient and Hessian. Options are "None", "MAD", "L2".
     response_fn: str
-        When a custom objective and metric are provided, XGBoost doesn't know its response and link function. Hence,
+        When a custom objective and metric are provided, LightGBM doesn't know its response and link function. Hence,
         the user is responsible for specifying the transformations. Options are "exp", "softplus" or "relu".
     loss_fn: str
         Loss function. Options are "nll" (negative log-likelihood) or "crps" (continuous ranked probability score).
