Fix imports in MBM tutorials (#2732)

Summary:
`ax.modelbridge.registry.Models` has been renamed to `Generators`. This diff updates the tutorials to utilize the new name.

Pull Request resolved: #2732

Reviewed By: esantorella

Differential Revision: D69210338

Pulled By: saitcakmak

fbshipit-source-id: d417f3904828b597b801be780c3b29eb5a432bbe

Author: saitcakmak

tutorials/custom_acquisition/custom_acquisition.ipynb

@@ -580,20 +580,20 @@
       "outputs": [],
       "source": [
         "from ax.modelbridge.generation_strategy import GenerationStep, GenerationStrategy\n",
-        "from ax.modelbridge.registry import Models\n",
+        "from ax.modelbridge.registry import Generators\n",
         "\n",
         "\n",
         "gs = GenerationStrategy(\n",
         "    steps=[\n",
         "        # Quasi-random initialization step\n",
         "        GenerationStep(\n",
-        "            model=Models.SOBOL,\n",
+        "            model=Generators.SOBOL,\n",
         "            num_trials=5,  # How many trials should be produced from this generation step\n",
         "            model_kwargs={\"seed\": 999},  # Any kwargs you want passed into the model\n",
         "        ),\n",
         "        # Bayesian optimization step using the custom acquisition function\n",
         "        GenerationStep(\n",
-        "            model=Models.BOTORCH_MODULAR,\n",
+        "            model=Generators.BOTORCH_MODULAR,\n",
         "            num_trials=-1,  # No limitation on how many trials should be produced from this step\n",
         "            # For `BOTORCH_MODULAR`, we pass in kwargs to specify what surrogate or acquisition function to use.\n",
         "            # `acquisition_options` specifies the set of additional arguments to pass into the input constructor.\n",

tutorials/custom_botorch_model_in_ax/custom_botorch_model_in_ax.ipynb

@@ -157,11 +157,10 @@
     "showInput": false
    },
    "source": [
-    "### Instantiate a `BoTorchModel` in Ax\n",
+    "### Instantiate a `BoTorchGenerator` in Ax\n",
     "\n",
-    "A `BoTorchModel` in Ax encapsulates both the surrogate -- which `Ax` calls a `Surrogate` and BoTorch calls a `Model` -- and an acquisition function. Here, we will only specify the custom surrogate and let Ax choose the default acquisition function.\n",
+    "A `BoTorchGenerator` in Ax encapsulates both the surrogate -- which `Ax` calls a `Surrogate` and BoTorch calls a `Model` -- and an acquisition function. Here, we will only specify the custom surrogate and let Ax choose the default acquisition function.\n",
     "\n",
-    "Most models should work with the base `Surrogate` in Ax, except for BoTorch `ModelListGP`, which works with `ListSurrogate`.\n",
     "Note that the `Model` (e.g., the `SimpleCustomGP`) must implement `construct_inputs`, as this is used to construct the inputs required for instantiating a `Model` instance from the experiment data."
    ]
   },
@@ -188,11 +187,11 @@
    },
    "outputs": [],
    "source": [
-    "from ax.models.torch.botorch_modular.model import BoTorchModel\n",
+    "from ax.models.torch.botorch_modular.model import BoTorchGenerator\n",
     "from ax.models.torch.botorch_modular.surrogate import Surrogate, SurrogateSpec\n",
     "from ax.models.torch.botorch_modular.utils import ModelConfig\n",
     "\n",
-    "ax_model = BoTorchModel(\n",
+    "ax_model = BoTorchGenerator(\n",
     "    surrogate=Surrogate(\n",
     "        surrogate_spec=SurrogateSpec(\n",
     "            model_configs=[\n",
@@ -230,11 +229,11 @@
    "source": [
     "### Combine with a `ModelBridge`\n",
     "\n",
-    "`Model`s in Ax require a `ModelBridge` to interface with `Experiment`s. A `ModelBridge` takes the inputs supplied by the `Experiment` and converts them to the inputs expected by the `Model`. For a `BoTorchModel`, we use `TorchModelBridge`. The Modular BoTorch interface creates the `BoTorchModel` and the `TorchModelBridge` in a single step, as follows:\n",
+    "`Model`s in Ax require a `ModelBridge` to interface with `Experiment`s. A `ModelBridge` takes the inputs supplied by the `Experiment` and converts them to the inputs expected by the `Model`. For a `BoTorchGenerator`, we use `TorchModelBridge`. The Modular BoTorch interface creates the `BoTorchGenerator` and the `TorchModelBridge` in a single step, as follows:\n",
     "\n",
     "```\n",
-    "from ax.modelbridge.registry import Models\n",
-    "model_bridge = Models.BOTORCH_MODULAR(\n",
+    "from ax.modelbridge.registry import Generators\n",
+    "model_bridge = Generators.BOTORCH_MODULAR(\n",
     "    experiment=experiment,\n",
     "    data=data,\n",
     "    surrogate=Surrogate(SimpleCustomGP),\n",
@@ -308,19 +307,19 @@
    "outputs": [],
    "source": [
     "from ax.modelbridge.generation_strategy import GenerationStep, GenerationStrategy\n",
-    "from ax.modelbridge.registry import Models\n",
+    "from ax.modelbridge.registry import Generators\n",
     "\n",
     "\n",
     "gs = GenerationStrategy(\n",
     "    steps=[\n",
     "        # Quasi-random initialization step\n",
     "        GenerationStep(\n",
-    "            model=Models.SOBOL,\n",
+    "            model=Generators.SOBOL,\n",
     "            num_trials=5,  # How many trials should be produced from this generation step\n",
     "        ),\n",
     "        # Bayesian optimization step using the custom acquisition function\n",
     "        GenerationStep(\n",
-    "            model=Models.BOTORCH_MODULAR,\n",
+    "            model=Generators.BOTORCH_MODULAR,\n",
     "            num_trials=-1,  # No limitation on how many trials should be produced from this step\n",
     "            # For `BOTORCH_MODULAR`, we pass in kwargs to specify what surrogate or acquisition function to use.\n",
     "            model_kwargs={\n",
@@ -1809,10 +1808,10 @@
    },
    "outputs": [],
    "source": [
-    "from ax.modelbridge.registry import Models\n",
+    "from ax.modelbridge.registry import Generators\n",
     "\n",
     "\n",
-    "sobol = Models.SOBOL(exp.search_space)\n",
+    "sobol = Generators.SOBOL(exp.search_space)\n",
     "\n",
     "for i in range(5):\n",
     "    trial = exp.new_trial(generator_run=sobol.gen(1))\n",
@@ -1878,7 +1877,7 @@
    "source": [
     "with fast_smoke_test():\n",
     "    for i in range(NUM_EVALS - 5):\n",
-    "        model_bridge = Models.BOTORCH_MODULAR(\n",
+    "        model_bridge = Generators.BOTORCH_MODULAR(\n",
     "            experiment=exp,\n",
     "            data=exp.fetch_data(),\n",
     "            surrogate_spec=SurrogateSpec(\n",