alan-turing-institute · sgreenbury · Jul 30, 2025 · Jul 24, 2025 · Jul 25, 2025 · Jul 26, 2025
diff --git a/autoemulate/experimental/emulators/__init__.py b/autoemulate/experimental/emulators/__init__.py
@@ -1,6 +1,6 @@
 from .base import Emulator
 from .ensemble import EnsembleMLP, EnsembleMLPDropout
-from .gaussian_process.exact import GaussianProcessExact, GaussianProcessExactCorrelated
+from .gaussian_process.exact import GaussianProcess, GaussianProcessCorrelated
 from .lightgbm import LightGBM
 
 # from .neural_processes.conditional_neural_process import CNPModule
@@ -11,8 +11,8 @@
 from .transformed.base import TransformedEmulator
 
 ALL_EMULATORS: list[type[Emulator]] = [
-    GaussianProcessExact,
-    GaussianProcessExactCorrelated,
+    GaussianProcess,
+    GaussianProcessCorrelated,
     LightGBM,
     # CNPModule,
     SupportVectorMachine,
@@ -57,8 +57,8 @@ def get_emulator_class(name: str) -> type[Emulator]:
     "MLP",
     "EnsembleMLP",
     "EnsembleMLPDropout",
-    "GaussianProcessExact",
-    "GaussianProcessExactCorrelated",
+    "GaussianProcess",
+    "GaussianProcessCorrelated",
     "LightGBM",
     "RadialBasisFunctions",
     "RandomForest",

diff --git a/autoemulate/experimental/emulators/base.py b/autoemulate/experimental/emulators/base.py
@@ -72,8 +72,8 @@ def model_name(cls) -> str:
     def short_name(cls) -> str:
         """
         Take the capital letters of the class name and return them as a lower case
-        string. For example, if the class name is `GaussianProcessExact`, this will
-        return `gpe`.
+        string. For example, if the class name is `GaussianProcess`, this will return
+        `gp`.
         """
         return "".join([c for c in cls.__name__ if c.isupper()]).lower()
 

diff --git a/autoemulate/experimental/emulators/gaussian_process/exact.py b/autoemulate/experimental/emulators/gaussian_process/exact.py
@@ -41,7 +41,7 @@
 from .mean import constant_mean, linear_mean, poly_mean, zero_mean
 
 
-class GaussianProcessExact(GaussianProcessEmulator, gpytorch.models.ExactGP):
+class GaussianProcess(GaussianProcessEmulator, gpytorch.models.ExactGP):
     """
     Gaussian Process Exact Emulator
 
@@ -75,7 +75,7 @@ def __init__(  # noqa: PLR0913 allow too many arguments since all currently requ
         **kwargs,
     ):
         """
-        Initialize the GaussianProcessExact emulator.
+        Initialize the GaussianProcess emulator.
 
         Parameters
         ----------
@@ -274,7 +274,7 @@ def _predict(self, x: TensorLike, with_grad: bool):
 
     @staticmethod
     def get_tune_config():
-        scheduler_config = GaussianProcessExact.scheduler_config()
+        scheduler_config = GaussianProcess.scheduler_config()
         return {
             "mean_module_fn": [
                 constant_mean,
@@ -300,11 +300,11 @@ def get_tune_config():
         }
 
 
-class GaussianProcessExactCorrelated(GaussianProcessExact):
+class GaussianProcessCorrelated(GaussianProcess):
     """
     Multioutput exact GP implementation with correlated task covariance.
 
-    This class extends the `GaussianProcessExact` to support correlated task covariance
+    This class extends the `GaussianProcess` to support correlated task covariance
     by using a `MultitaskKernel` with a rank-1 covariance factor and a `MultitaskMean`
     for the mean function.
 
@@ -331,7 +331,7 @@ def __init__(  # noqa: PLR0913 allow too many arguments since all currently requ
         **kwargs,
     ):
         """
-        Initialize the GaussianProcessExactCorrelated emulator.
+        Initialize the GaussianProcessCorrelated emulator.
 
         Parameters
         ----------

diff --git a/autoemulate/experimental/exploratory/compare_transforms.ipynb b/autoemulate/experimental/exploratory/compare_transforms.ipynb
@@ -2,13 +2,13 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "702bb87d",
    "metadata": {},
    "outputs": [],
    "source": [
     "import torch\n",
-    "from autoemulate.experimental.emulators import GaussianProcessExact\n",
+    "from autoemulate.experimental.emulators import GaussianProcess\n",
     "from autoemulate.experimental.emulators.random_forest import RandomForest\n",
     "from autoemulate.experimental.emulators.transformed.base import TransformedEmulator\n",
     "from autoemulate.experimental.transforms import PCATransform, VAETransform, StandardizeTransform\n",
@@ -52,15 +52,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "d67372dd",
    "metadata": {},
    "outputs": [],
    "source": [
     "ae = AutoEmulate(\n",
     "    x,\n",
     "    y,\n",
-    "    models=[GaussianProcessExact, RandomForest],\n",
+    "    models=[GaussianProcess, RandomForest],\n",
     "    x_transforms_list=[[], [StandardizeTransform(), PCATransform(n_components=5)]],\n",
     "    y_transforms_list=[[], [StandardizeTransform(), PCATransform(n_components=1)]]\n",
     ")\n"

diff --git a/autoemulate/experimental/exploratory/hmc_refactor.ipynb b/autoemulate/experimental/exploratory/hmc_refactor.ipynb
@@ -16,9 +16,7 @@
     "import torch\n",
     "\n",
     "from autoemulate.experimental.simulations.projectile import Projectile, ProjectileMultioutput\n",
-    "from autoemulate.experimental.emulators.gaussian_process.exact import (\n",
-    "    GaussianProcessExact,\n",
-    ")"
+    "from autoemulate.experimental.emulators import GaussianProcess"
    ]
   },
   {
@@ -38,7 +36,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "gp = GaussianProcessExact(x, y)\n",
+    "gp = GaussianProcess(x, y)\n",
     "gp.fit(x, y)"
    ]
   },

diff --git a/autoemulate/experimental/exploratory/transforms.ipynb b/autoemulate/experimental/exploratory/transforms.ipynb
@@ -8,7 +8,7 @@
    "outputs": [],
    "source": [
     "import torch\n",
-    "from autoemulate.experimental.emulators import GaussianProcessExact\n",
+    "from autoemulate.experimental.emulators import GaussianProcess\n",
     "from autoemulate.experimental.emulators.random_forest import RandomForest\n",
     "from autoemulate.experimental.emulators.transformed.base import TransformedEmulator\n",
     "from autoemulate.experimental.transforms import PCATransform, VAETransform, StandardizeTransform\n",
@@ -85,7 +85,7 @@
     "    y=y,\n",
     "    x_transforms=[PCATransform(n_components=4), VAETransform(latent_dim=2)],\n",
     "    y_transforms=[StandardizeTransform(), PCATransform(n_components=1)],\n",
-    "    model=GaussianProcessExact,\n",
+    "    model=GaussianProcess,\n",
     "    epochs=100,\n",
     ")"
    ]
@@ -143,7 +143,7 @@
     "from autoemulate.experimental.model_selection import r2_metric\n",
     "\n",
     "\n",
-    "for model in [GaussianProcessExact, RandomForest]:\n",
+    "for model in [GaussianProcess, RandomForest]:\n",
     "    # Create transformed emulator with GP\n",
     "    emulator = TransformedEmulator(\n",
     "        x=x,\n",
@@ -223,7 +223,7 @@
     "    x, y = make_data(n_targets=5, n_samples=200)\n",
     "    t.fit(y)\n",
     "    z = t(y)\n",
-    "    gp = GaussianProcessExact(x, z, standardize_x=True, standardize_y=True)\n",
+    "    gp = GaussianProcess(x, z, standardize_x=True, standardize_y=True)\n",
     "    gp.fit(x, z)\n",
     "    z_pred = gp.predict(x[: x.shape[0] // 2])\n",
     "    assert isinstance(z_pred, GaussianLike)\n",
@@ -282,7 +282,7 @@
     "            y=y,\n",
     "            y_transforms=[t],\n",
     "            x_transforms=[],\n",
-    "            model=GaussianProcessExact,\n",
+    "            model=GaussianProcess,\n",
     "            epochs=50,\n",
     "            n_samples=n_samples,\n",
     "            full_covariance=True,\n",
@@ -324,7 +324,7 @@
     "            y=y,\n",
     "            x_transforms=[StandardizeTransform()],\n",
     "            y_transforms=[StandardizeTransform(), t],\n",
-    "            model=GaussianProcessExact,\n",
+    "            model=GaussianProcess,\n",
     "            epochs=50,\n",
     "            n_samples=n_samples,\n",
     "            full_covariance=True,\n",

diff --git a/docs/experimental/tutorials/emulation/03_reaction_diffusion_experimental.ipynb b/docs/experimental/tutorials/emulation/03_reaction_diffusion_experimental.ipynb
@@ -148,7 +148,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from autoemulate.experimental.emulators.gaussian_process.exact import GaussianProcessExact\n",
+    "from autoemulate.experimental.emulators.gaussian_process.exact import GaussianProcess\n",
     "from autoemulate.experimental.emulators.gaussian_process.kernel import rbf_plus_constant\n",
     "from autoemulate.experimental.emulators.gaussian_process.mean import constant_mean\n",
     "from autoemulate.experimental.emulators.transformed.base import TransformedEmulator\n",
@@ -162,7 +162,7 @@
     "em = TransformedEmulator(\n",
     "    x,\n",
     "    y,\n",
-    "    model= GaussianProcessExact,\n",
+    "    model= GaussianProcess,\n",
     "    x_transforms=[StandardizeTransform()],\n",
     "    y_transforms=[\n",
     "        StandardizeTransform(),\n",

diff --git a/docs/experimental/tutorials/simulator/02_active_learning.ipynb b/docs/experimental/tutorials/simulator/02_active_learning.ipynb
@@ -69,7 +69,7 @@
     "# Import core classes from the source code.\n",
     "from autoemulate.experimental.learners import stream\n",
     "from autoemulate.experimental.simulations.base import Simulator\n",
-    "from autoemulate.experimental.emulators.gaussian_process.exact import GaussianProcessExact\n",
+    "from autoemulate.experimental.emulators import GaussianProcess\n",
     "from autoemulate.experimental.types import GaussianLike\n",
     "\n",
     "warnings.filterwarnings('ignore')\n",
@@ -205,7 +205,7 @@
     "x_train = simulator.sample_inputs(25).sort(dim=0).values\n",
     "y_train = simulator.forward_batch(x_train)\n",
     "\n",
-    "emulator = GaussianProcessExact(x_train, y_train)\n",
+    "emulator = GaussianProcess(x_train, y_train)\n",
     "emulator.fit(x_train, y_train)\n",
     "\n",
     "# Test emulator\n",
@@ -259,7 +259,7 @@
     "\n",
     "x_train = simulator.sample_inputs(5)\n",
     "y_train = simulator.forward_batch(x_train)\n",
-    "emulator = GaussianProcessExact(x_train, y_train)\n",
+    "emulator = GaussianProcess(x_train, y_train)\n",
     "\n",
     "# Learner itself!\n",
     "learner = stream.Random(\n",
@@ -458,33 +458,33 @@
     "    x_train = simulator.sample_inputs(n_initial_samples)\n",
     "    y_train = simulator.forward_batch(x_train)\n",
     "    yield stream.Random(\n",
-    "        simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "        simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "        x_train=x_train, y_train=y_train,\n",
     "        p_query=0.25, show_progress=show_progress\n",
     "    )\n",
     "    if not adaptive_only:\n",
     "        yield stream.Distance(\n",
-    "            simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "            simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "            x_train=x_train, y_train=y_train,\n",
     "            threshold=0.5, show_progress=show_progress\n",
     "        )\n",
     "        yield stream.A_Optimal(\n",
-    "            simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "            simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "            x_train=x_train, y_train=y_train,\n",
     "            threshold=1.0, show_progress=show_progress\n",
     "        )\n",
     "        yield stream.D_Optimal(\n",
-    "            simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "            simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "            x_train=x_train, y_train=y_train,\n",
     "            threshold=-4.2, show_progress=show_progress\n",
     "        )\n",
     "        yield stream.E_Optimal(\n",
-    "            simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "            simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "            x_train=x_train, y_train=y_train,\n",
     "            threshold=1.0, show_progress=show_progress\n",
     "        )\n",
     "    yield stream.Adaptive_Distance(\n",
-    "        simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "        simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "        x_train=x_train, y_train=y_train,\n",
     "        threshold=0.5, Kp=1.0, Ki=1.0, Kd=1.0,\n",
     "        key=\"rate\", target=0.25,\n",
@@ -493,7 +493,7 @@
     "        window_size=10, show_progress=show_progress\n",
     "    )\n",
     "    yield stream.Adaptive_A_Optimal(\n",
-    "        simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "        simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "        x_train=x_train, y_train=y_train,\n",
     "        threshold=1e-1, Kp=2.0, Ki=1.0, Kd=2.0,\n",
     "        key=\"rate\", target=0.25,\n",
@@ -502,7 +502,7 @@
     "        window_size=10, show_progress=show_progress\n",
     "    )\n",
     "    yield stream.Adaptive_D_Optimal(\n",
-    "        simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "        simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "        x_train=x_train, y_train=y_train,\n",
     "        threshold=-4.0, Kp=2.0, Ki=1.0, Kd=2.0,\n",
     "        key=\"rate\", target=0.25,\n",
@@ -511,7 +511,7 @@
     "        window_size=10, show_progress=show_progress\n",
     "    )\n",
     "    yield stream.Adaptive_E_Optimal(\n",
-    "        simulator=simulator, emulator=GaussianProcessExact(x_train, y_train),\n",
+    "        simulator=simulator, emulator=GaussianProcess(x_train, y_train),\n",
     "        x_train=x_train, y_train=y_train,\n",
     "        threshold=0.75 if isinstance(simulator, Sin) else 1000, \n",
     "        Kp=2.0, Ki=1.0, Kd=2.0,\n",
@@ -776,7 +776,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.7"
+   "version": "3.12.11"
   }
  },
  "nbformat": 4,

diff --git a/docs/experimental/tutorials/simulator/03_history_matching.ipynb b/docs/experimental/tutorials/simulator/03_history_matching.ipynb
@@ -2,13 +2,11 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from autoemulate.experimental.emulators.gaussian_process.exact import (\n",
-    "    GaussianProcessExact,\n",
-    ")\n",
+    "from autoemulate.experimental.emulators import GaussianProcess\n",
     "from autoemulate.experimental.compare import AutoEmulate\n",
     "from autoemulate.experimental.simulations.epidemic import Epidemic\n",
     "from autoemulate.experimental.calibration.history_matching import HistoryMatchingWorkflow\n",
@@ -73,7 +71,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "metadata": {},
    "outputs": [
     {
@@ -85,7 +83,7 @@
     }
    ],
    "source": [
-    "ae = AutoEmulate(x, y, models=[GaussianProcessExact])"
+    "ae = AutoEmulate(x, y, models=[GaussianProcess])"
    ]
   },
   {

diff --git a/docs/experimental/tutorials/tasks/02_history_matching.ipynb b/docs/experimental/tutorials/tasks/02_history_matching.ipynb
@@ -2,13 +2,11 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from autoemulate.experimental.emulators.gaussian_process.exact import (\n",
-    "    GaussianProcessExact,\n",
-    ")\n",
+    "from autoemulate.experimental.emulators import GaussianProcess\n",
     "from autoemulate.experimental.compare import AutoEmulate\n",
     "from autoemulate.experimental.simulations.projectile import Projectile\n",
     "from autoemulate.experimental.calibration.history_matching import HistoryMatching"
@@ -71,7 +69,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
    "outputs": [
     {
@@ -83,7 +81,7 @@
     }
    ],
    "source": [
-    "ae = AutoEmulate(x, y, models=[GaussianProcessExact])"
+    "ae = AutoEmulate(x, y, models=[GaussianProcess])"
    ]
   },
   {