diff --git a/lib/scholar/preprocessing.ex b/lib/scholar/preprocessing.ex
index 67c38b83..77d2b84b 100644
--- a/lib/scholar/preprocessing.ex
+++ b/lib/scholar/preprocessing.ex
@@ -133,14 +133,8 @@ defmodule Scholar.Preprocessing do
       >
   """
   deftransform standard_scale(tensor, opts \\ []) do
-    standard_scale_n(tensor, NimbleOptions.validate!(opts, @general_schema))
-  end
-
-  defnp standard_scale_n(tensor, opts) do
-    std = Nx.standard_deviation(tensor, axes: opts[:axes], keep_axes: true)
-    mean_reduced = Nx.mean(tensor, axes: opts[:axes], keep_axes: true)
-    mean_reduced = Nx.select(std == 0, 0.0, mean_reduced)
-    (tensor - mean_reduced) / Nx.select(std == 0, 1.0, std)
+    opts = NimbleOptions.validate!(opts, @general_schema)
+    Scholar.Preprocessing.StandardScaler.fit_transform(tensor, opts)
   end
 
   @doc """
diff --git a/lib/scholar/scaler/standard_scaler.ex b/lib/scholar/preprocessing/standard_scaler.ex
similarity index 67%
rename from lib/scholar/scaler/standard_scaler.ex
rename to lib/scholar/preprocessing/standard_scaler.ex
index 676a84e9..9a028bd3 100644
--- a/lib/scholar/scaler/standard_scaler.ex
+++ b/lib/scholar/preprocessing/standard_scaler.ex
@@ -1,4 +1,4 @@
-defmodule Scholar.Scaler.StandardScaler do
+defmodule Scholar.Preprocessing.StandardScaler do
   import Nx.Defn
 
   defstruct [:deviation, :mean]
@@ -17,20 +17,27 @@ defmodule Scholar.Scaler.StandardScaler do
 
   deftransform fit(tensor, opts \\ []) do
     NimbleOptions.validate!(opts, @opts_schema)
+    {std, mean} = fit_n(tensor, opts)
 
+    %__MODULE__{deviation: std, mean: mean}
+  end
+
+  defnp fit_n(tensor, opts) do
     std = Nx.standard_deviation(tensor, axes: opts[:axes], keep_axes: true)
     mean_reduced = Nx.mean(tensor, axes: opts[:axes], keep_axes: true)
     mean_reduced = Nx.select(Nx.equal(std, 0), 0.0, mean_reduced)
-    %__MODULE__{deviation: std, mean: mean_reduced}
+
+    {std, mean_reduced}
   end
 
-  deftransform transform(tensor, %__MODULE__{deviation: std, mean: mean}) do
+  deftransform transform(%__MODULE__{deviation: std, mean: mean}, tensor) do
     scale(tensor, std, mean)
   end
 
-  deftransform fit_transform(tensor, opts \\ []) do
-    scaler = __MODULE__.fit(tensor, opts)
-    __MODULE__.transform(tensor, scaler)
+  defn fit_transform(tensor, opts \\ []) do
+    tensor
+    |> fit(opts)
+    |> transform(tensor)
   end
 
   defnp scale(tensor, std, mean) do
diff --git a/test/scholar/scaler/standard_scaler_test.exs b/test/scholar/preprocessing/standard_scaler_test.exs
similarity index 94%
rename from test/scholar/scaler/standard_scaler_test.exs
rename to test/scholar/preprocessing/standard_scaler_test.exs
index b788e36d..4470efcf 100644
--- a/test/scholar/scaler/standard_scaler_test.exs
+++ b/test/scholar/preprocessing/standard_scaler_test.exs
@@ -1,6 +1,6 @@
 defmodule StandardScalerTest do
   use Scholar.Case, async: true
-  alias Scholar.Scaler.StandardScaler
+  alias Scholar.Preprocessing.StandardScaler
 
   describe "fit_transform/2" do
     test "applies standard scaling to data" do