From 2d65a35739aa68e188248dda065199d8e34352fd Mon Sep 17 00:00:00 2001
From: Sait Cakmak <saitcakmak@fb.com>
Date: Thu, 12 Dec 2024 14:20:10 -0800
Subject: [PATCH] Replace lazy tensor with linear operator in RGPE tutorial
 (#2650)

Summary:
Lazy tensors have been deprecated for a long time and they were removed from GPyTorch today.

Pull Request resolved: https://github.com/pytorch/botorch/pull/2650

Reviewed By: Balandat

Differential Revision: D67156374

Pulled By: saitcakmak

fbshipit-source-id: 823c1edca133a509ddd719dc82d3109c172e6f74
---
 tutorials/meta_learning_with_rgpe.ipynb | 2206 ++++++++++++-----------
 1 file changed, 1118 insertions(+), 1088 deletions(-)

diff --git a/tutorials/meta_learning_with_rgpe.ipynb b/tutorials/meta_learning_with_rgpe.ipynb
index b183ec883c..1483b6909d 100644
--- a/tutorials/meta_learning_with_rgpe.ipynb
+++ b/tutorials/meta_learning_with_rgpe.ipynb
@@ -1,1134 +1,1164 @@
 {
-  "metadata": {
-    "kernelspec": {
-      "name": "python3",
-      "display_name": "python3",
-      "language": "python",
-      "isCinder": true
-    },
-    "language_info": {
-      "codemirror_mode": {
-        "name": "ipython",
-        "version": 3
-      },
-      "file_extension": ".py",
-      "mimetype": "text/x-python",
-      "name": "python",
-      "nbconvert_exporter": "python",
-      "pygments_lexer": "ipython3",
-      "version": "3.8.5"
-    }
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "62d6d3ed-36ff-4609-bc82-1451f8093bd9",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "## Meta-Learning with the Rank-Weighted GP Ensemble (RGPE)\n",
+    "\n",
+    "BoTorch is designed in to be model-agnostic and only requries that a model conform to a minimal interface. This tutorial walks through an example of implementing the rank-weighted Gaussian process ensemble (RGPE) [Feurer, Letham, Bakshy ICML 2018 AutoML Workshop] and using the RGPE in BoTorch to do meta-learning across related optimization tasks.\n",
+    "\n",
+    "* Original paper: https://arxiv.org/pdf/1802.02219.pdf"
+   ]
   },
-  "nbformat": 4,
-  "nbformat_minor": 2,
-  "cells": [
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "62d6d3ed-36ff-4609-bc82-1451f8093bd9",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "## Meta-Learning with the Rank-Weighted GP Ensemble (RGPE)\n",
-        "\n",
-        "BoTorch is designed in to be model-agnostic and only requries that a model conform to a minimal interface. This tutorial walks through an example of implementing the rank-weighted Gaussian process ensemble (RGPE) [Feurer, Letham, Bakshy ICML 2018 AutoML Workshop] and using the RGPE in BoTorch to do meta-learning across related optimization tasks.\n",
-        "\n",
-        "* Original paper: https://arxiv.org/pdf/1802.02219.pdf"
-      ]
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948296406,
+    "executionStopTime": 1724948298817,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "6fc093c3-2d62-49c1-a4cc-558193bcff8b",
-        "collapsed": false,
-        "requestMsgId": "6fc093c3-2d62-49c1-a4cc-558193bcff8b",
-        "customOutput": null,
-        "executionStartTime": 1724948296406,
-        "executionStopTime": 1724948298817,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 2277.4769549724
-      },
-      "source": [
-        "import os\n",
-        "import torch\n",
-        "import math\n",
-        "\n",
-        "\n",
-        "torch.manual_seed(29)\n",
-        "device = torch.device(\"cuda:2\" if torch.cuda.is_available() else \"cpu\")\n",
-        "dtype = torch.double\n",
-        "SMOKE_TEST = os.environ.get(\"SMOKE_TEST\")"
-      ],
-      "execution_count": 1,
-      "outputs": [
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "I0829 091817.060 _utils_internal.py:292] NCCL_DEBUG env var is set to None\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "I0829 091817.061 _utils_internal.py:310] NCCL_DEBUG is forced to WARN from None\n"
-          ]
-        }
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "2352cd6d-70b4-4426-9fdc-e0b8ac91aac5",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "### Toy Problem\n",
-        "* We consider optimizing the following 1-D synthetic function\n",
-        "$$f(x, s_i) = \\frac{1}{10}\\bigg(x-1\\bigg)\\bigg(\\sin(x+s_i)+\\frac{1}{10}\\bigg)$$\n",
-        "where\n",
-        "$$s_i = \\frac{(i+9)\\pi}{8}$$\n",
-        "is a task-dependent shift parameter and $i$ is the task index $i \\in [1, t]$.\n",
-        "\n",
-        "* In this tutorial, we will consider the scenario where we have collected data from 5 prior tasks (referred to as base tasks), which with a different task dependent shift parameter $s_i$.\n",
-        "\n",
-        "* The goal now is use meta-learning to improve sample efficiency when optimizing a 6th task."
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "babbfc73-97d4-491c-87e0-be07d1acc2d7",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "#### Toy Problem Setup\n",
-        "\n",
-        "First let's define a function for compute the shift parameter $s_i$ and set the shift amount for the target task."
-      ]
+    "language": "python",
+    "originalKey": "6fc093c3-2d62-49c1-a4cc-558193bcff8b",
+    "outputsInitialized": true,
+    "requestMsgId": "6fc093c3-2d62-49c1-a4cc-558193bcff8b",
+    "serverExecutionDuration": 2277.4769549724
+   },
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import torch\n",
+    "import math\n",
+    "\n",
+    "\n",
+    "torch.manual_seed(29)\n",
+    "device = torch.device(\"cuda:2\" if torch.cuda.is_available() else \"cpu\")\n",
+    "dtype = torch.double\n",
+    "SMOKE_TEST = os.environ.get(\"SMOKE_TEST\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "2352cd6d-70b4-4426-9fdc-e0b8ac91aac5",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "### Toy Problem\n",
+    "* We consider optimizing the following 1-D synthetic function\n",
+    "$$f(x, s_i) = \\frac{1}{10}\\bigg(x-1\\bigg)\\bigg(\\sin(x+s_i)+\\frac{1}{10}\\bigg)$$\n",
+    "where\n",
+    "$$s_i = \\frac{(i+9)\\pi}{8}$$\n",
+    "is a task-dependent shift parameter and $i$ is the task index $i \\in [1, t]$.\n",
+    "\n",
+    "* In this tutorial, we will consider the scenario where we have collected data from 5 prior tasks (referred to as base tasks), which with a different task dependent shift parameter $s_i$.\n",
+    "\n",
+    "* The goal now is use meta-learning to improve sample efficiency when optimizing a 6th task."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "babbfc73-97d4-491c-87e0-be07d1acc2d7",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "#### Toy Problem Setup\n",
+    "\n",
+    "First let's define a function for compute the shift parameter $s_i$ and set the shift amount for the target task."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948297830,
+    "executionStopTime": 1724948298839,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "4c0c2b47-6313-4450-bdec-366db7c00643",
-        "collapsed": false,
-        "requestMsgId": "4c0c2b47-6313-4450-bdec-366db7c00643",
-        "customOutput": null,
-        "executionStartTime": 1724948297830,
-        "executionStopTime": 1724948298839,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 1.4142158906907
-      },
-      "source": [
-        "NUM_BASE_TASKS = 5 if not SMOKE_TEST else 2\n",
-        "\n",
-        "\n",
-        "def task_shift(task):\n",
-        "    \"\"\"\n",
-        "    Fetch shift amount for task.\n",
-        "    \"\"\"\n",
-        "    return math.pi * task / 12.0\n",
-        "\n",
-        "\n",
-        "# set shift for target task\n",
-        "\n",
-        "TARGET_SHIFT = 0.0"
-      ],
-      "execution_count": 2,
-      "outputs": []
+    "language": "python",
+    "originalKey": "4c0c2b47-6313-4450-bdec-366db7c00643",
+    "outputsInitialized": true,
+    "requestMsgId": "4c0c2b47-6313-4450-bdec-366db7c00643",
+    "serverExecutionDuration": 1.4142158906907
+   },
+   "outputs": [],
+   "source": [
+    "NUM_BASE_TASKS = 5 if not SMOKE_TEST else 2\n",
+    "\n",
+    "\n",
+    "def task_shift(task):\n",
+    "    \"\"\"\n",
+    "    Fetch shift amount for task.\n",
+    "    \"\"\"\n",
+    "    return math.pi * task / 12.0\n",
+    "\n",
+    "\n",
+    "# set shift for target task\n",
+    "\n",
+    "TARGET_SHIFT = 0.0"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "d5650131-21e8-40d4-9003-d89b7431fb29",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "Then, let's define our function $f(x, s_i)$ and set bounds on $x$."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948298726,
+    "executionStopTime": 1724948298909,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "d5650131-21e8-40d4-9003-d89b7431fb29",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "Then, let's define our function $f(x, s_i)$ and set bounds on $x$."
-      ]
+    "language": "python",
+    "originalKey": "c1abc54e-b410-437d-beee-ce06d248706f",
+    "outputsInitialized": true,
+    "requestMsgId": "c1abc54e-b410-437d-beee-ce06d248706f",
+    "serverExecutionDuration": 15.071736183017
+   },
+   "outputs": [],
+   "source": [
+    "BOUNDS = torch.tensor([[-10.0], [10.0]], dtype=dtype, device=device)\n",
+    "\n",
+    "\n",
+    "def f(X, shift=TARGET_SHIFT):\n",
+    "    \"\"\"\n",
+    "    Torch-compatible objective function for the target_task\n",
+    "    \"\"\"\n",
+    "    f_X = X * torch.sin(X + math.pi + shift) + X / 10.0\n",
+    "    return f_X"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "d9896be8-4cd9-487e-9832-768e533635c4",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "#### Sample training data for prior base tasks"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "f27bbe94-c3cb-4d58-9d82-ebd60d4ebd59",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "We sample data from a Sobol sequence to help ensure numerical stability when using a small amount of 1-D data. Sobol sequences help prevent us from sampling a bunch of training points that are close together."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948300185,
+    "executionStopTime": 1724948301395,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "c1abc54e-b410-437d-beee-ce06d248706f",
-        "collapsed": false,
-        "requestMsgId": "c1abc54e-b410-437d-beee-ce06d248706f",
-        "customOutput": null,
-        "executionStartTime": 1724948298726,
-        "executionStopTime": 1724948298909,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 15.071736183017
-      },
-      "source": [
-        "BOUNDS = torch.tensor([[-10.0], [10.0]], dtype=dtype, device=device)\n",
-        "\n",
-        "\n",
-        "def f(X, shift=TARGET_SHIFT):\n",
-        "    \"\"\"\n",
-        "    Torch-compatible objective function for the target_task\n",
-        "    \"\"\"\n",
-        "    f_X = X * torch.sin(X + math.pi + shift) + X / 10.0\n",
-        "    return f_X"
-      ],
-      "execution_count": 3,
-      "outputs": []
+    "language": "python",
+    "originalKey": "75962b70-ca73-4ab4-97fd-e28de1abdd81",
+    "outputsInitialized": true,
+    "requestMsgId": "75962b70-ca73-4ab4-97fd-e28de1abdd81",
+    "serverExecutionDuration": 1088.0531340372
+   },
+   "outputs": [],
+   "source": [
+    "from botorch.utils.sampling import draw_sobol_samples\n",
+    "from botorch.utils.transforms import normalize, unnormalize\n",
+    "\n",
+    "\n",
+    "noise_std = 0.05\n",
+    "\n",
+    "# Sample data for each base task\n",
+    "data_by_task = {}\n",
+    "for task in range(NUM_BASE_TASKS):\n",
+    "    num_training_points = 20\n",
+    "    # draw points from a sobol sequence\n",
+    "    raw_x = draw_sobol_samples(\n",
+    "        bounds=BOUNDS,\n",
+    "        n=num_training_points,\n",
+    "        q=1,\n",
+    "        seed=task + 5397923,\n",
+    "    ).squeeze(1)\n",
+    "    # get observed values\n",
+    "    f_x = f(raw_x, task_shift(task + 1))\n",
+    "    train_y = f_x + noise_std * torch.randn_like(f_x)\n",
+    "    train_yvar = torch.full_like(train_y, noise_std**2)\n",
+    "    # store training data\n",
+    "    data_by_task[task] = {\n",
+    "        # scale x to [0, 1]\n",
+    "        \"train_x\": normalize(raw_x, bounds=BOUNDS),\n",
+    "        \"train_y\": train_y,\n",
+    "        \"train_yvar\": train_yvar,\n",
+    "    }"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "80336086-3253-4a0a-8875-e5db7440b362",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "#### Let's plot the base tasks and the target task function along with the observed points"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948301524,
+    "executionStopTime": 1724948303012,
+    "jupyter": {
+     "outputs_hidden": false
     },
+    "language": "python",
+    "originalKey": "aeff41b6-623a-4b10-a583-d47563b29700",
+    "outputsInitialized": true,
+    "requestMsgId": "aeff41b6-623a-4b10-a583-d47563b29700",
+    "serverExecutionDuration": 1299.3806430604
+   },
+   "outputs": [
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "d9896be8-4cd9-487e-9832-768e533635c4",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "#### Sample training data for prior base tasks"
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x800 with 1 Axes>"
       ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from matplotlib import pyplot as plt\n",
+    "\n",
+    "%matplotlib inline\n",
+    "\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 1, figsize=(12, 8))\n",
+    "x = torch.linspace(-10, 10, 51)\n",
+    "for task in data_by_task:\n",
+    "    # plot true function and observed values for base runs\n",
+    "    t = ax.plot(\n",
+    "        unnormalize(data_by_task[task][\"train_x\"], bounds=BOUNDS).cpu().numpy(),\n",
+    "        data_by_task[task][\"train_y\"].cpu().numpy(),\n",
+    "        \".\",\n",
+    "        markersize=10,\n",
+    "        label=f\"Observed task {task}\",\n",
+    "    )\n",
+    "    ax.plot(\n",
+    "        x.detach().numpy(),\n",
+    "        f(x, task_shift(task + 1)).cpu().numpy(),\n",
+    "        label=f\"Base task {task}\",\n",
+    "        color=t[0].get_color(),\n",
+    "    )\n",
+    "# plot true target function\n",
+    "ax.plot(\n",
+    "    x.detach().numpy(),\n",
+    "    f(x, TARGET_SHIFT).detach().numpy(),\n",
+    "    \"--\",\n",
+    "    label=\"Target task\",\n",
+    ")\n",
+    "ax.legend(loc=\"lower right\", fontsize=10)\n",
+    "plt.tight_layout()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "25d7a014-8982-40e0-8025-f12bba83dc45",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "### Fit base task models"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "cdf621a8-1057-4cf3-a129-668abb1f9453",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "First, let's define a helper function to fit a SingleTaskGP with an fixed observed noise level."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948316352,
+    "executionStopTime": 1724948316501,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "f27bbe94-c3cb-4d58-9d82-ebd60d4ebd59",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "We sample data from a Sobol sequence to help ensure numerical stability when using a small amount of 1-D data. Sobol sequences help prevent us from sampling a bunch of training points that are close together."
-      ]
+    "language": "python",
+    "originalKey": "5474e712-cefa-4a7d-b673-af10fcf83239",
+    "outputsInitialized": true,
+    "requestMsgId": "5474e712-cefa-4a7d-b673-af10fcf83239",
+    "serverExecutionDuration": 2.9860010836273
+   },
+   "outputs": [],
+   "source": [
+    "from gpytorch.mlls import ExactMarginalLogLikelihood\n",
+    "from botorch.models import SingleTaskGP\n",
+    "from botorch.fit import fit_gpytorch_mll\n",
+    "\n",
+    "\n",
+    "def get_fitted_model(train_X, train_Y, train_Yvar, state_dict=None):\n",
+    "    \"\"\"\n",
+    "    Get a single task GP. The model will be fit unless a state_dict with model\n",
+    "        hyperparameters is provided.\n",
+    "    \"\"\"\n",
+    "    model = SingleTaskGP(train_X=train_X, train_Y=train_Y, train_Yvar=train_Yvar)\n",
+    "    if state_dict is None:\n",
+    "        mll = ExactMarginalLogLikelihood(model.likelihood, model).to(train_X)\n",
+    "        fit_gpytorch_mll(mll)\n",
+    "    else:\n",
+    "        model.load_state_dict(state_dict)\n",
+    "    return model"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "fe570963-41f1-47a5-8e53-5cf08e6390ba",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "#### Now let's fit a SingleTaskGP for each base task"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948317183,
+    "executionStopTime": 1724948318815,
+    "jupyter": {
+     "outputs_hidden": false
     },
+    "language": "python",
+    "originalKey": "a7bd3664-5585-47e9-9743-2ee53d7a259f",
+    "outputsInitialized": true,
+    "requestMsgId": "a7bd3664-5585-47e9-9743-2ee53d7a259f",
+    "serverExecutionDuration": 1460.1275878958
+   },
+   "outputs": [
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "75962b70-ca73-4ab4-97fd-e28de1abdd81",
-        "collapsed": false,
-        "requestMsgId": "75962b70-ca73-4ab4-97fd-e28de1abdd81",
-        "customOutput": null,
-        "executionStartTime": 1724948300185,
-        "executionStopTime": 1724948301395,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 1088.0531340372
-      },
-      "source": [
-        "from botorch.utils.sampling import draw_sobol_samples\n",
-        "from botorch.utils.transforms import normalize, unnormalize\n",
-        "\n",
-        "\n",
-        "noise_std = 0.05\n",
-        "\n",
-        "# Sample data for each base task\n",
-        "data_by_task = {}\n",
-        "for task in range(NUM_BASE_TASKS):\n",
-        "    num_training_points = 20\n",
-        "    # draw points from a sobol sequence\n",
-        "    raw_x = draw_sobol_samples(\n",
-        "        bounds=BOUNDS,\n",
-        "        n=num_training_points,\n",
-        "        q=1,\n",
-        "        seed=task + 5397923,\n",
-        "    ).squeeze(1)\n",
-        "    # get observed values\n",
-        "    f_x = f(raw_x, task_shift(task + 1))\n",
-        "    train_y = f_x + noise_std * torch.randn_like(f_x)\n",
-        "    train_yvar = torch.full_like(train_y, noise_std**2)\n",
-        "    # store training data\n",
-        "    data_by_task[task] = {\n",
-        "        # scale x to [0, 1]\n",
-        "        \"train_x\": normalize(raw_x, bounds=BOUNDS),\n",
-        "        \"train_y\": train_y,\n",
-        "        \"train_yvar\": train_yvar,\n",
-        "    }"
-      ],
-      "execution_count": 4,
-      "outputs": []
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Fitting base model 0\n",
+      "Fitting base model 1\n",
+      "Fitting base model 2\n",
+      "Fitting base model 3\n",
+      "Fitting base model 4\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Fit base model\n",
+    "base_model_list = []\n",
+    "for task in range(NUM_BASE_TASKS):\n",
+    "    print(f\"Fitting base model {task}\")\n",
+    "    model = get_fitted_model(\n",
+    "        data_by_task[task][\"train_x\"],\n",
+    "        data_by_task[task][\"train_y\"],\n",
+    "        data_by_task[task][\"train_yvar\"],\n",
+    "    )\n",
+    "    base_model_list.append(model)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "6ea07cc4-db47-4d01-9840-220e9615b6a3",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "### Implement the RGPE\n",
+    "\n",
+    "The main idea of the RGPE is to estimate the target function as weighted sum of the target model and the base models:\n",
+    "$$\\bar f(\\mathbf x | \\mathcal D) =\n",
+    "\\sum_{i=1}^{t} w_if^i(\\mathbf x |\\mathcal D_i)$$\n",
+    "Importantly, the ensemble model is also a GP:\n",
+    "$$\\bar f(\\mathbf x | \\mathcal D) \\sim \\mathcal N\\bigg(\\sum_{i=1}^{t} w_i\\mu_i(\\mathbf x), \\sum_{i=1}^{t}w_i^2\\sigma_i^2\\bigg)$$\n",
+    "\n",
+    "The weights $w_i$ for model $i$ are based on the the ranking loss between a draw from the model's posterior and the targets. Specifically, the ranking loss for model $i$ is:\n",
+    "$$\\mathcal L(f^i, \\mathcal D_t) = \\sum_{j=1}^{n_t}\\sum_{k=1}^{n_t}\\mathbb 1\\bigg[\\bigg(f^i\\big(\\mathbf x^t_j\\big) < f^i\\big(\\mathbf x_k^t\\big)\\bigg)\\oplus \\big(y_j^t < y_k^t\\big)\\bigg]$$\n",
+    "where $\\oplus$ is exclusive-or.\n",
+    "\n",
+    "The loss for the target model is computing using leave-one-out cross-validation (LOOCV) and is given by:\n",
+    "$$\\mathcal L(f^t, \\mathcal D_t) = \\sum_{j=1}^{n_t}\\sum_{k=1}^{n_t}\\mathbb 1\\bigg[\\bigg(f^t_{-j}\\big(\\mathbf x^t_j\\big) < f^t_{-j}\\big(\\mathbf x_k^t\\big)\\bigg)\\oplus \\big(y_j^t < y_k^t\\big)\\bigg]$$\n",
+    "where $f^t_{-j}$ model fitted to all data from the target task except training example $j$.\n",
+    "\n",
+    "The weights are then computed as:\n",
+    "$$w_i = \\frac{1}{S}\\sum_{s=1}^S\\mathbb 1\\big(i = \\text{argmin}_{i'}l_{i', s}\\big)$$"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948320494,
+    "executionStopTime": 1724948320631,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "80336086-3253-4a0a-8875-e5db7440b362",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "#### Let's plot the base tasks and the target task function along with the observed points"
-      ]
+    "language": "python",
+    "originalKey": "58ed8284-8181-459c-9025-42532793929f",
+    "outputsInitialized": true,
+    "requestMsgId": "58ed8284-8181-459c-9025-42532793929f",
+    "serverExecutionDuration": 2.3661230225116
+   },
+   "outputs": [],
+   "source": [
+    "def roll_col(X, shift):\n",
+    "    \"\"\"\n",
+    "    Rotate columns to right by shift.\n",
+    "    \"\"\"\n",
+    "    return torch.cat((X[..., -shift:], X[..., :-shift]), dim=-1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948325542,
+    "executionStopTime": 1724948325683,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "aeff41b6-623a-4b10-a583-d47563b29700",
-        "collapsed": false,
-        "requestMsgId": "aeff41b6-623a-4b10-a583-d47563b29700",
-        "customOutput": null,
-        "executionStartTime": 1724948301524,
-        "executionStopTime": 1724948303012,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 1299.3806430604
-      },
-      "source": [
-        "from matplotlib import pyplot as plt\n",
-        "\n",
-        "%matplotlib inline\n",
-        "\n",
-        "\n",
-        "fig, ax = plt.subplots(1, 1, figsize=(12, 8))\n",
-        "x = torch.linspace(-10, 10, 51)\n",
-        "for task in data_by_task:\n",
-        "    # plot true function and observed values for base runs\n",
-        "    t = ax.plot(\n",
-        "        unnormalize(data_by_task[task][\"train_x\"], bounds=BOUNDS).cpu().numpy(),\n",
-        "        data_by_task[task][\"train_y\"].cpu().numpy(),\n",
-        "        \".\",\n",
-        "        markersize=10,\n",
-        "        label=f\"Observed task {task}\",\n",
-        "    )\n",
-        "    ax.plot(\n",
-        "        x.detach().numpy(),\n",
-        "        f(x, task_shift(task + 1)).cpu().numpy(),\n",
-        "        label=f\"Base task {task}\",\n",
-        "        color=t[0].get_color(),\n",
-        "    )\n",
-        "# plot true target function\n",
-        "ax.plot(\n",
-        "    x.detach().numpy(),\n",
-        "    f(x, TARGET_SHIFT).detach().numpy(),\n",
-        "    \"--\",\n",
-        "    label=\"Target task\",\n",
-        ")\n",
-        "ax.legend(loc=\"lower right\", fontsize=10)\n",
-        "plt.tight_layout()"
-      ],
-      "execution_count": 5,
-      "outputs": [
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "W0829 091822.520 font_manager.py:1403] findfont: Font family ['Liberation Sans', 'Noto Sans TC', 'Noto Sans SC', 'Noto Sans Thai', 'Noto Naskh Arabic UI', 'Noto Sans UI'] not found. Falling back to DejaVu Sans.\n"
-          ]
-        },
-        {
-          "output_type": "display_data",
-          "data": {
-            "text/plain": "<Figure size 1200x800 with 1 Axes>",
-            "image/png": "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"
-          },
-          "metadata": {}
-        }
-      ]
+    "language": "python",
+    "originalKey": "bde867e5-67e9-47f0-bae3-20112bebd51d",
+    "outputsInitialized": true,
+    "requestMsgId": "bde867e5-67e9-47f0-bae3-20112bebd51d",
+    "serverExecutionDuration": 3.575277980417
+   },
+   "outputs": [],
+   "source": [
+    "def compute_ranking_loss(f_samps, target_y):\n",
+    "    \"\"\"\n",
+    "    Compute ranking loss for each sample from the posterior over target points.\n",
+    "\n",
+    "    Args:\n",
+    "        f_samps: `n_samples x (n) x n`-dim tensor of samples\n",
+    "        target_y: `n x 1`-dim tensor of targets\n",
+    "    Returns:\n",
+    "        Tensor: `n_samples`-dim tensor containing the ranking loss across each sample\n",
+    "    \"\"\"\n",
+    "    n = target_y.shape[0]\n",
+    "    if f_samps.ndim == 3:\n",
+    "        # Compute ranking loss for target model\n",
+    "        # take cartesian product of target_y\n",
+    "        cartesian_y = torch.cartesian_prod(\n",
+    "            target_y.squeeze(-1),\n",
+    "            target_y.squeeze(-1),\n",
+    "        ).view(n, n, 2)\n",
+    "        # the diagonal of f_samps are the out-of-sample predictions\n",
+    "        # for each LOO model, compare the out of sample predictions to each in-sample prediction\n",
+    "        rank_loss = (\n",
+    "            (\n",
+    "                (f_samps.diagonal(dim1=1, dim2=2).unsqueeze(-1) < f_samps)\n",
+    "                ^ (cartesian_y[..., 0] < cartesian_y[..., 1])\n",
+    "            )\n",
+    "            .sum(dim=-1)\n",
+    "            .sum(dim=-1)\n",
+    "        )\n",
+    "    else:\n",
+    "        rank_loss = torch.zeros(\n",
+    "            f_samps.shape[0], dtype=torch.long, device=target_y.device\n",
+    "        )\n",
+    "        y_stack = target_y.squeeze(-1).expand(f_samps.shape)\n",
+    "        for i in range(1, target_y.shape[0]):\n",
+    "            rank_loss += (\n",
+    "                (roll_col(f_samps, i) < f_samps) ^ (roll_col(y_stack, i) < y_stack)\n",
+    "            ).sum(dim=-1)\n",
+    "    return rank_loss"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "158d11b4-020f-478c-9ec4-8655ae8c2aac",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "Define a function to:\n",
+    "1. Create a batch mode-gp LOOCV GP using the hyperparameters from `target_model`\n",
+    "2. Draw a joint sample across all points from the target task (in-sample and out-of-sample)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948361037,
+    "executionStopTime": 1724948361226,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "25d7a014-8982-40e0-8025-f12bba83dc45",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "### Fit base task models"
-      ]
+    "language": "python",
+    "originalKey": "91127a10-93d6-4cc1-9eaa-3ac49e7be678",
+    "outputsInitialized": true,
+    "requestMsgId": "91127a10-93d6-4cc1-9eaa-3ac49e7be678",
+    "serverExecutionDuration": 4.2148299980909
+   },
+   "outputs": [],
+   "source": [
+    "def get_target_model_loocv_sample_preds(\n",
+    "    train_x, train_y, train_yvar, target_model, num_samples\n",
+    "):\n",
+    "    \"\"\"\n",
+    "    Create a batch-mode LOOCV GP and draw a joint sample across all points from the target task.\n",
+    "\n",
+    "    Args:\n",
+    "        train_x: `n x d` tensor of training points\n",
+    "        train_y: `n x 1` tensor of training targets\n",
+    "        target_model: fitted target model\n",
+    "        num_samples: number of mc samples to draw\n",
+    "\n",
+    "    Return: `num_samples x n x n`-dim tensor of samples, where dim=1 represents the `n` LOO models,\n",
+    "        and dim=2 represents the `n` training points.\n",
+    "    \"\"\"\n",
+    "    batch_size = len(train_x)\n",
+    "    masks = torch.eye(len(train_x), dtype=torch.uint8, device=device).bool()\n",
+    "    train_x_cv = torch.stack([train_x[~m] for m in masks])\n",
+    "    train_y_cv = torch.stack([train_y[~m] for m in masks])\n",
+    "    train_yvar_cv = torch.stack([train_yvar[~m] for m in masks])\n",
+    "    state_dict = target_model.state_dict()\n",
+    "    # expand to batch size of batch_mode LOOCV model\n",
+    "    state_dict_expanded = {\n",
+    "        name: t.expand(batch_size, *[-1 for _ in range(t.ndim)])\n",
+    "        for name, t in state_dict.items()\n",
+    "    }\n",
+    "    model = get_fitted_model(\n",
+    "        train_x_cv, train_y_cv, train_yvar_cv, state_dict=state_dict_expanded\n",
+    "    )\n",
+    "    with torch.no_grad():\n",
+    "        posterior = model.posterior(train_x)\n",
+    "        # Since we have a batch mode gp and model.posterior always returns an output dimension,\n",
+    "        # the output from `posterior.sample()` here `num_samples x n x n x 1`, so let's squeeze\n",
+    "        # the last dimension.\n",
+    "        sampler = SobolQMCNormalSampler(sample_shape=torch.Size([num_samples]))\n",
+    "        return sampler(posterior).squeeze(-1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948370606,
+    "executionStopTime": 1724948370882,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "cdf621a8-1057-4cf3-a129-668abb1f9453",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "First, let's define a helper function to fit a SingleTaskGP with an fixed observed noise level."
-      ]
+    "language": "python",
+    "originalKey": "298fa009-eae9-4199-a682-df730e10c20e",
+    "outputsInitialized": true,
+    "requestMsgId": "298fa009-eae9-4199-a682-df730e10c20e",
+    "serverExecutionDuration": 3.5223178565502
+   },
+   "outputs": [],
+   "source": [
+    "def compute_rank_weights(train_x, train_y, base_models, target_model, num_samples):\n",
+    "    \"\"\"\n",
+    "    Compute ranking weights for each base model and the target model (using\n",
+    "        LOOCV for the target model). Note: This implementation does not currently\n",
+    "        address weight dilution, since we only have a small number of base models.\n",
+    "\n",
+    "    Args:\n",
+    "        train_x: `n x d` tensor of training points (for target task)\n",
+    "        train_y: `n` tensor of training targets (for target task)\n",
+    "        base_models: list of base models\n",
+    "        target_model: target model\n",
+    "        num_samples: number of mc samples\n",
+    "\n",
+    "    Returns:\n",
+    "        Tensor: `n_t`-dim tensor with the ranking weight for each model\n",
+    "    \"\"\"\n",
+    "    ranking_losses = []\n",
+    "    # compute ranking loss for each base model\n",
+    "    for task in range(len(base_models)):\n",
+    "        model = base_models[task]\n",
+    "        # compute posterior over training points for target task\n",
+    "        posterior = model.posterior(train_x)\n",
+    "        sampler = SobolQMCNormalSampler(sample_shape=torch.Size([num_samples]))\n",
+    "        base_f_samps = sampler(posterior).squeeze(-1).squeeze(-1)\n",
+    "        # compute and save ranking loss\n",
+    "        ranking_losses.append(compute_ranking_loss(base_f_samps, train_y))\n",
+    "    # compute ranking loss for target model using LOOCV\n",
+    "    # f_samps\n",
+    "    target_f_samps = get_target_model_loocv_sample_preds(\n",
+    "        train_x,\n",
+    "        train_y,\n",
+    "        train_yvar,\n",
+    "        target_model,\n",
+    "        num_samples,\n",
+    "    )\n",
+    "    ranking_losses.append(compute_ranking_loss(target_f_samps, train_y))\n",
+    "    ranking_loss_tensor = torch.stack(ranking_losses)\n",
+    "    # compute best model (minimum ranking loss) for each sample\n",
+    "    best_models = torch.argmin(ranking_loss_tensor, dim=0)\n",
+    "    # compute proportion of samples for which each model is best\n",
+    "    rank_weights = (\n",
+    "        best_models.bincount(minlength=len(ranking_losses)).type_as(train_x)\n",
+    "        / num_samples\n",
+    "    )\n",
+    "    return rank_weights"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948386869,
+    "executionStopTime": 1724948387021,
+    "jupyter": {
+     "outputs_hidden": false
     },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "5474e712-cefa-4a7d-b673-af10fcf83239",
-        "collapsed": false,
-        "requestMsgId": "5474e712-cefa-4a7d-b673-af10fcf83239",
-        "customOutput": null,
-        "executionStartTime": 1724948316352,
-        "executionStopTime": 1724948316501,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 2.9860010836273
-      },
-      "source": [
-        "from gpytorch.mlls import ExactMarginalLogLikelihood\n",
-        "from botorch.models import SingleTaskGP\n",
-        "from botorch.fit import fit_gpytorch_mll\n",
-        "\n",
-        "\n",
-        "def get_fitted_model(train_X, train_Y, train_Yvar, state_dict=None):\n",
-        "    \"\"\"\n",
-        "    Get a single task GP. The model will be fit unless a state_dict with model\n",
-        "        hyperparameters is provided.\n",
-        "    \"\"\"\n",
-        "    model = SingleTaskGP(train_X=train_X, train_Y=train_Y, train_Yvar=train_Yvar)\n",
-        "    if state_dict is None:\n",
-        "        mll = ExactMarginalLogLikelihood(model.likelihood, model).to(train_X)\n",
-        "        fit_gpytorch_mll(mll)\n",
-        "    else:\n",
-        "        model.load_state_dict(state_dict)\n",
-        "    return model"
-      ],
-      "execution_count": 8,
-      "outputs": []
+    "language": "python",
+    "originalKey": "5393f2d7-07c3-4e41-a68f-0cf92fb2aa8f",
+    "outputsInitialized": true,
+    "requestMsgId": "5393f2d7-07c3-4e41-a68f-0cf92fb2aa8f",
+    "serverExecutionDuration": 5.0720849540085
+   },
+   "outputs": [],
+   "source": [
+    "from botorch.models.gpytorch import GPyTorchModel\n",
+    "from gpytorch.models import GP\n",
+    "from gpytorch.distributions import MultivariateNormal\n",
+    "from gpytorch.likelihoods import LikelihoodList\n",
+    "from linear_operator.operators import PsdSumLinearOperator\n",
+    "from torch.nn import ModuleList\n",
+    "\n",
+    "\n",
+    "class RGPE(GP, GPyTorchModel):\n",
+    "    \"\"\"\n",
+    "    Rank-weighted GP ensemble. Note: this class inherits from GPyTorchModel which provides an\n",
+    "        interface for GPyTorch models in botorch.\n",
+    "    \"\"\"\n",
+    "\n",
+    "    _num_outputs = 1  # metadata for botorch\n",
+    "\n",
+    "    def __init__(self, models, weights):\n",
+    "        super().__init__()\n",
+    "        self.models = ModuleList(models)\n",
+    "        for m in models:\n",
+    "            if not hasattr(m, \"likelihood\"):\n",
+    "                raise ValueError(\n",
+    "                    \"RGPE currently only supports models that have a likelihood (e.g. ExactGPs)\"\n",
+    "                )\n",
+    "        self.likelihood = LikelihoodList(*[m.likelihood for m in models])\n",
+    "        self.weights = weights\n",
+    "        self.to(weights)\n",
+    "\n",
+    "    def forward(self, x):\n",
+    "        weighted_means = []\n",
+    "        weighted_covars = []\n",
+    "        # filter model with zero weights\n",
+    "        # weights on covariance matrices are weight**2\n",
+    "        non_zero_weight_indices = (self.weights**2 > 0).nonzero()\n",
+    "        non_zero_weights = self.weights[non_zero_weight_indices]\n",
+    "        # re-normalize\n",
+    "        non_zero_weights /= non_zero_weights.sum()\n",
+    "\n",
+    "        for non_zero_weight_idx in range(non_zero_weight_indices.shape[0]):\n",
+    "            raw_idx = non_zero_weight_indices[non_zero_weight_idx].item()\n",
+    "            model = self.models[raw_idx]\n",
+    "            posterior = model.posterior(x)\n",
+    "            # unstandardize predictions\n",
+    "            posterior_mean = posterior.mean.squeeze(-1)\n",
+    "            posterior_cov = posterior.mvn.lazy_covariance_matrix\n",
+    "            # apply weight\n",
+    "            weight = non_zero_weights[non_zero_weight_idx]\n",
+    "            weighted_means.append(weight * posterior_mean)\n",
+    "            weighted_covars.append(posterior_cov * weight**2)\n",
+    "        # set mean and covariance to be the rank-weighted sum the means and covariances of the\n",
+    "        # base models and target model\n",
+    "        mean_x = torch.stack(weighted_means).sum(dim=0)\n",
+    "        covar_x = PsdSumLinearOperator(*weighted_covars)\n",
+    "        return MultivariateNormal(mean_x, covar_x)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "49b770aa-d0c1-4e37-8366-ca1debde2f40",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "### Optimize target function using RGPE + qNEI"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948469002,
+    "executionStopTime": 1724948513622,
+    "jupyter": {
+     "outputs_hidden": false
     },
+    "language": "python",
+    "originalKey": "4670c8c2-c171-4e3e-87f6-0ca3543140df",
+    "outputsInitialized": true,
+    "requestMsgId": "4670c8c2-c171-4e3e-87f6-0ca3543140df",
+    "serverExecutionDuration": 44297.466597985
+   },
+   "outputs": [
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "fe570963-41f1-47a5-8e53-5cf08e6390ba",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "#### Now let's fit a SingleTaskGP for each base task"
-      ]
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 1 of 10\n",
+      "Trial 2 of 10\n",
+      "Trial 3 of 10\n",
+      "Trial 4 of 10\n",
+      "Trial 5 of 10\n",
+      "Trial 6 of 10\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "a7bd3664-5585-47e9-9743-2ee53d7a259f",
-        "collapsed": false,
-        "requestMsgId": "a7bd3664-5585-47e9-9743-2ee53d7a259f",
-        "customOutput": null,
-        "executionStartTime": 1724948317183,
-        "executionStopTime": 1724948318815,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 1460.1275878958
-      },
-      "source": [
-        "# Fit base model\n",
-        "base_model_list = []\n",
-        "for task in range(NUM_BASE_TASKS):\n",
-        "    print(f\"Fitting base model {task}\")\n",
-        "    model = get_fitted_model(\n",
-        "        data_by_task[task][\"train_x\"],\n",
-        "        data_by_task[task][\"train_y\"],\n",
-        "        data_by_task[task][\"train_yvar\"],\n",
-        "    )\n",
-        "    base_model_list.append(model)"
-      ],
-      "execution_count": 9,
-      "outputs": [
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Fitting base model 0\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Fitting base model 1\nFitting base model 2\nFitting base model 3\nFitting base model 4\n"
-          ]
-        }
-      ]
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed in `gen_candidates_scipy` with the following warning(s):\n",
+      "[OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.')]\n",
+      "Trying again with a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n",
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed on the second try, after generating a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n",
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed in `gen_candidates_scipy` with the following warning(s):\n",
+      "[BotorchWarning('Low-rank cholesky updates failed due NaNs or due to an ill-conditioned covariance matrix. Falling back to standard sampling.'), OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.'), BotorchWarning('Low-rank cholesky updates failed due NaNs or due to an ill-conditioned covariance matrix. Falling back to standard sampling.')]\n",
+      "Trying again with a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n"
+     ]
     },
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "6ea07cc4-db47-4d01-9840-220e9615b6a3",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "### Implement the RGPE\n",
-        "\n",
-        "The main idea of the RGPE is to estimate the target function as weighted sum of the target model and the base models:\n",
-        "$$\\bar f(\\mathbf x | \\mathcal D) =\n",
-        "\\sum_{i=1}^{t} w_if^i(\\mathbf x |\\mathcal D_i)$$\n",
-        "Importantly, the ensemble model is also a GP:\n",
-        "$$\\bar f(\\mathbf x | \\mathcal D) \\sim \\mathcal N\\bigg(\\sum_{i=1}^{t} w_i\\mu_i(\\mathbf x), \\sum_{i=1}^{t}w_i^2\\sigma_i^2\\bigg)$$\n",
-        "\n",
-        "The weights $w_i$ for model $i$ are based on the the ranking loss between a draw from the model's posterior and the targets. Specifically, the ranking loss for model $i$ is:\n",
-        "$$\\mathcal L(f^i, \\mathcal D_t) = \\sum_{j=1}^{n_t}\\sum_{k=1}^{n_t}\\mathbb 1\\bigg[\\bigg(f^i\\big(\\mathbf x^t_j\\big) < f^i\\big(\\mathbf x_k^t\\big)\\bigg)\\oplus \\big(y_j^t < y_k^t\\big)\\bigg]$$\n",
-        "where $\\oplus$ is exclusive-or.\n",
-        "\n",
-        "The loss for the target model is computing using leave-one-out cross-validation (LOOCV) and is given by:\n",
-        "$$\\mathcal L(f^t, \\mathcal D_t) = \\sum_{j=1}^{n_t}\\sum_{k=1}^{n_t}\\mathbb 1\\bigg[\\bigg(f^t_{-j}\\big(\\mathbf x^t_j\\big) < f^t_{-j}\\big(\\mathbf x_k^t\\big)\\bigg)\\oplus \\big(y_j^t < y_k^t\\big)\\bigg]$$\n",
-        "where $f^t_{-j}$ model fitted to all data from the target task except training example $j$.\n",
-        "\n",
-        "The weights are then computed as:\n",
-        "$$w_i = \\frac{1}{S}\\sum_{s=1}^S\\mathbb 1\\big(i = \\text{argmin}_{i'}l_{i', s}\\big)$$"
-      ]
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 7 of 10\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "58ed8284-8181-459c-9025-42532793929f",
-        "collapsed": false,
-        "requestMsgId": "58ed8284-8181-459c-9025-42532793929f",
-        "customOutput": null,
-        "executionStartTime": 1724948320494,
-        "executionStopTime": 1724948320631,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 2.3661230225116
-      },
-      "source": [
-        "def roll_col(X, shift):\n",
-        "    \"\"\"\n",
-        "    Rotate columns to right by shift.\n",
-        "    \"\"\"\n",
-        "    return torch.cat((X[..., -shift:], X[..., :-shift]), dim=-1)"
-      ],
-      "execution_count": 10,
-      "outputs": []
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed in `gen_candidates_scipy` with the following warning(s):\n",
+      "[OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.')]\n",
+      "Trying again with a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n",
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed on the second try, after generating a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "bde867e5-67e9-47f0-bae3-20112bebd51d",
-        "collapsed": false,
-        "requestMsgId": "bde867e5-67e9-47f0-bae3-20112bebd51d",
-        "customOutput": null,
-        "executionStartTime": 1724948325542,
-        "executionStopTime": 1724948325683,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 3.575277980417
-      },
-      "source": [
-        "def compute_ranking_loss(f_samps, target_y):\n",
-        "    \"\"\"\n",
-        "    Compute ranking loss for each sample from the posterior over target points.\n",
-        "\n",
-        "    Args:\n",
-        "        f_samps: `n_samples x (n) x n`-dim tensor of samples\n",
-        "        target_y: `n x 1`-dim tensor of targets\n",
-        "    Returns:\n",
-        "        Tensor: `n_samples`-dim tensor containing the ranking loss across each sample\n",
-        "    \"\"\"\n",
-        "    n = target_y.shape[0]\n",
-        "    if f_samps.ndim == 3:\n",
-        "        # Compute ranking loss for target model\n",
-        "        # take cartesian product of target_y\n",
-        "        cartesian_y = torch.cartesian_prod(\n",
-        "            target_y.squeeze(-1),\n",
-        "            target_y.squeeze(-1),\n",
-        "        ).view(n, n, 2)\n",
-        "        # the diagonal of f_samps are the out-of-sample predictions\n",
-        "        # for each LOO model, compare the out of sample predictions to each in-sample prediction\n",
-        "        rank_loss = (\n",
-        "            (\n",
-        "                (f_samps.diagonal(dim1=1, dim2=2).unsqueeze(-1) < f_samps)\n",
-        "                ^ (cartesian_y[..., 0] < cartesian_y[..., 1])\n",
-        "            )\n",
-        "            .sum(dim=-1)\n",
-        "            .sum(dim=-1)\n",
-        "        )\n",
-        "    else:\n",
-        "        rank_loss = torch.zeros(\n",
-        "            f_samps.shape[0], dtype=torch.long, device=target_y.device\n",
-        "        )\n",
-        "        y_stack = target_y.squeeze(-1).expand(f_samps.shape)\n",
-        "        for i in range(1, target_y.shape[0]):\n",
-        "            rank_loss += (\n",
-        "                (roll_col(f_samps, i) < f_samps) ^ (roll_col(y_stack, i) < y_stack)\n",
-        "            ).sum(dim=-1)\n",
-        "    return rank_loss"
-      ],
-      "execution_count": 11,
-      "outputs": []
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 8 of 10\n"
+     ]
     },
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "158d11b4-020f-478c-9ec4-8655ae8c2aac",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "Define a function to:\n",
-        "1. Create a batch mode-gp LOOCV GP using the hyperparameters from `target_model`\n",
-        "2. Draw a joint sample across all points from the target task (in-sample and out-of-sample)"
-      ]
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed in `gen_candidates_scipy` with the following warning(s):\n",
+      "[OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.')]\n",
+      "Trying again with a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n",
+      "/Users/saitcakmak/botorch/botorch/optim/optimize.py:648: RuntimeWarning: Optimization failed on the second try, after generating a new set of initial conditions.\n",
+      "  return _optimize_acqf_batch(opt_inputs=opt_inputs)\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "91127a10-93d6-4cc1-9eaa-3ac49e7be678",
-        "collapsed": false,
-        "requestMsgId": "91127a10-93d6-4cc1-9eaa-3ac49e7be678",
-        "customOutput": null,
-        "executionStartTime": 1724948361037,
-        "executionStopTime": 1724948361226,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 4.2148299980909
-      },
-      "source": [
-        "def get_target_model_loocv_sample_preds(\n",
-        "    train_x, train_y, train_yvar, target_model, num_samples\n",
-        "):\n",
-        "    \"\"\"\n",
-        "    Create a batch-mode LOOCV GP and draw a joint sample across all points from the target task.\n",
-        "\n",
-        "    Args:\n",
-        "        train_x: `n x d` tensor of training points\n",
-        "        train_y: `n x 1` tensor of training targets\n",
-        "        target_model: fitted target model\n",
-        "        num_samples: number of mc samples to draw\n",
-        "\n",
-        "    Return: `num_samples x n x n`-dim tensor of samples, where dim=1 represents the `n` LOO models,\n",
-        "        and dim=2 represents the `n` training points.\n",
-        "    \"\"\"\n",
-        "    batch_size = len(train_x)\n",
-        "    masks = torch.eye(len(train_x), dtype=torch.uint8, device=device).bool()\n",
-        "    train_x_cv = torch.stack([train_x[~m] for m in masks])\n",
-        "    train_y_cv = torch.stack([train_y[~m] for m in masks])\n",
-        "    train_yvar_cv = torch.stack([train_yvar[~m] for m in masks])\n",
-        "    state_dict = target_model.state_dict()\n",
-        "    # expand to batch size of batch_mode LOOCV model\n",
-        "    state_dict_expanded = {\n",
-        "        name: t.expand(batch_size, *[-1 for _ in range(t.ndim)])\n",
-        "        for name, t in state_dict.items()\n",
-        "    }\n",
-        "    model = get_fitted_model(\n",
-        "        train_x_cv, train_y_cv, train_yvar_cv, state_dict=state_dict_expanded\n",
-        "    )\n",
-        "    with torch.no_grad():\n",
-        "        posterior = model.posterior(train_x)\n",
-        "        # Since we have a batch mode gp and model.posterior always returns an output dimension,\n",
-        "        # the output from `posterior.sample()` here `num_samples x n x n x 1`, so let's squeeze\n",
-        "        # the last dimension.\n",
-        "        sampler = SobolQMCNormalSampler(sample_shape=torch.Size([num_samples]))\n",
-        "        return sampler(posterior).squeeze(-1)"
-      ],
-      "execution_count": 12,
-      "outputs": []
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 9 of 10\n",
+      "Trial 10 of 10\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "298fa009-eae9-4199-a682-df730e10c20e",
-        "collapsed": false,
-        "requestMsgId": "298fa009-eae9-4199-a682-df730e10c20e",
-        "customOutput": null,
-        "executionStartTime": 1724948370606,
-        "executionStopTime": 1724948370882,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 3.5223178565502
-      },
-      "source": [
-        "def compute_rank_weights(train_x, train_y, base_models, target_model, num_samples):\n",
-        "    \"\"\"\n",
-        "    Compute ranking weights for each base model and the target model (using\n",
-        "        LOOCV for the target model). Note: This implementation does not currently\n",
-        "        address weight dilution, since we only have a small number of base models.\n",
-        "\n",
-        "    Args:\n",
-        "        train_x: `n x d` tensor of training points (for target task)\n",
-        "        train_y: `n` tensor of training targets (for target task)\n",
-        "        base_models: list of base models\n",
-        "        target_model: target model\n",
-        "        num_samples: number of mc samples\n",
-        "\n",
-        "    Returns:\n",
-        "        Tensor: `n_t`-dim tensor with the ranking weight for each model\n",
-        "    \"\"\"\n",
-        "    ranking_losses = []\n",
-        "    # compute ranking loss for each base model\n",
-        "    for task in range(len(base_models)):\n",
-        "        model = base_models[task]\n",
-        "        # compute posterior over training points for target task\n",
-        "        posterior = model.posterior(train_x)\n",
-        "        sampler = SobolQMCNormalSampler(sample_shape=torch.Size([num_samples]))\n",
-        "        base_f_samps = sampler(posterior).squeeze(-1).squeeze(-1)\n",
-        "        # compute and save ranking loss\n",
-        "        ranking_losses.append(compute_ranking_loss(base_f_samps, train_y))\n",
-        "    # compute ranking loss for target model using LOOCV\n",
-        "    # f_samps\n",
-        "    target_f_samps = get_target_model_loocv_sample_preds(\n",
-        "        train_x,\n",
-        "        train_y,\n",
-        "        train_yvar,\n",
-        "        target_model,\n",
-        "        num_samples,\n",
-        "    )\n",
-        "    ranking_losses.append(compute_ranking_loss(target_f_samps, train_y))\n",
-        "    ranking_loss_tensor = torch.stack(ranking_losses)\n",
-        "    # compute best model (minimum ranking loss) for each sample\n",
-        "    best_models = torch.argmin(ranking_loss_tensor, dim=0)\n",
-        "    # compute proportion of samples for which each model is best\n",
-        "    rank_weights = (\n",
-        "        best_models.bincount(minlength=len(ranking_losses)).type_as(train_x)\n",
-        "        / num_samples\n",
-        "    )\n",
-        "    return rank_weights"
-      ],
-      "execution_count": 13,
-      "outputs": []
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 8 of 10\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "5393f2d7-07c3-4e41-a68f-0cf92fb2aa8f",
-        "collapsed": false,
-        "requestMsgId": "5393f2d7-07c3-4e41-a68f-0cf92fb2aa8f",
-        "customOutput": null,
-        "executionStartTime": 1724948386869,
-        "executionStopTime": 1724948387021,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 5.0720849540085
-      },
-      "source": [
-        "from botorch.models.gpytorch import GPyTorchModel\n",
-        "from gpytorch.models import GP\n",
-        "from gpytorch.distributions import MultivariateNormal\n",
-        "from gpytorch.lazy import PsdSumLazyTensor\n",
-        "from gpytorch.likelihoods import LikelihoodList\n",
-        "from torch.nn import ModuleList\n",
-        "\n",
-        "\n",
-        "class RGPE(GP, GPyTorchModel):\n",
-        "    \"\"\"\n",
-        "    Rank-weighted GP ensemble. Note: this class inherits from GPyTorchModel which provides an\n",
-        "        interface for GPyTorch models in botorch.\n",
-        "    \"\"\"\n",
-        "\n",
-        "    _num_outputs = 1  # metadata for botorch\n",
-        "\n",
-        "    def __init__(self, models, weights):\n",
-        "        super().__init__()\n",
-        "        self.models = ModuleList(models)\n",
-        "        for m in models:\n",
-        "            if not hasattr(m, \"likelihood\"):\n",
-        "                raise ValueError(\n",
-        "                    \"RGPE currently only supports models that have a likelihood (e.g. ExactGPs)\"\n",
-        "                )\n",
-        "        self.likelihood = LikelihoodList(*[m.likelihood for m in models])\n",
-        "        self.weights = weights\n",
-        "        self.to(weights)\n",
-        "\n",
-        "    def forward(self, x):\n",
-        "        weighted_means = []\n",
-        "        weighted_covars = []\n",
-        "        # filter model with zero weights\n",
-        "        # weights on covariance matrices are weight**2\n",
-        "        non_zero_weight_indices = (self.weights**2 > 0).nonzero()\n",
-        "        non_zero_weights = self.weights[non_zero_weight_indices]\n",
-        "        # re-normalize\n",
-        "        non_zero_weights /= non_zero_weights.sum()\n",
-        "\n",
-        "        for non_zero_weight_idx in range(non_zero_weight_indices.shape[0]):\n",
-        "            raw_idx = non_zero_weight_indices[non_zero_weight_idx].item()\n",
-        "            model = self.models[raw_idx]\n",
-        "            posterior = model.posterior(x)\n",
-        "            # unstandardize predictions\n",
-        "            posterior_mean = posterior.mean.squeeze(-1)\n",
-        "            posterior_cov = posterior.mvn.lazy_covariance_matrix\n",
-        "            # apply weight\n",
-        "            weight = non_zero_weights[non_zero_weight_idx]\n",
-        "            weighted_means.append(weight * posterior_mean)\n",
-        "            weighted_covars.append(posterior_cov * weight**2)\n",
-        "        # set mean and covariance to be the rank-weighted sum the means and covariances of the\n",
-        "        # base models and target model\n",
-        "        mean_x = torch.stack(weighted_means).sum(dim=0)\n",
-        "        covar_x = PsdSumLazyTensor(*weighted_covars)\n",
-        "        return MultivariateNormal(mean_x, covar_x)"
-      ],
-      "execution_count": 14,
-      "outputs": []
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 9 of 10\n"
+     ]
     },
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "49b770aa-d0c1-4e37-8366-ca1debde2f40",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "### Optimize target function using RGPE + qNEI"
-      ]
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[W 240829 09:21:46 optimize:564] Optimization failed in `gen_candidates_scipy` with the following warning(s):\n",
+      "    [OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.')]\n",
+      "    Trying again with a new set of initial conditions.\n"
+     ]
     },
     {
-      "cell_type": "code",
-      "metadata": {
-        "scrolled": false,
-        "originalKey": "4670c8c2-c171-4e3e-87f6-0ca3543140df",
-        "collapsed": false,
-        "requestMsgId": "4670c8c2-c171-4e3e-87f6-0ca3543140df",
-        "customOutput": null,
-        "executionStartTime": 1724948469002,
-        "executionStopTime": 1724948513622,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 44297.466597985
-      },
-      "source": [
-        "# suppress GPyTorch warnings about adding jitter\n",
-        "import warnings\n",
-        "\n",
-        "from botorch.acquisition.logei import qLogNoisyExpectedImprovement\n",
-        "from botorch.optim.optimize import optimize_acqf\n",
-        "from botorch.sampling.normal import SobolQMCNormalSampler\n",
-        "\n",
-        "\n",
-        "warnings.filterwarnings(\"ignore\", \"^.*jitter.*\", category=RuntimeWarning)\n",
-        "\n",
-        "\n",
-        "best_rgpe_all = []\n",
-        "best_random_all = []\n",
-        "best_vanilla_nei_all = []\n",
-        "N_BATCH = 10 if not SMOKE_TEST else 2\n",
-        "NUM_POSTERIOR_SAMPLES = 256 if not SMOKE_TEST else 16\n",
-        "RANDOM_INITIALIZATION_SIZE = 3\n",
-        "N_TRIALS = 10 if not SMOKE_TEST else 2\n",
-        "MC_SAMPLES = 512 if not SMOKE_TEST else 32\n",
-        "N_RESTART_CANDIDATES = 512 if not SMOKE_TEST else 8\n",
-        "N_RESTARTS = 10 if not SMOKE_TEST else 2\n",
-        "Q_BATCH_SIZE = 1\n",
-        "\n",
-        "\n",
-        "# Average over multiple trials\n",
-        "for trial in range(N_TRIALS):\n",
-        "    print(f\"Trial {trial + 1} of {N_TRIALS}\")\n",
-        "    best_rgpe = []\n",
-        "    best_random = []\n",
-        "    best_vanilla_nei = []\n",
-        "    # Initial random observations\n",
-        "    raw_x = draw_sobol_samples(\n",
-        "        bounds=BOUNDS, n=RANDOM_INITIALIZATION_SIZE, q=1, seed=trial\n",
-        "    ).squeeze(1)\n",
-        "    train_x = normalize(raw_x, bounds=BOUNDS)\n",
-        "    train_y_noiseless = f(raw_x)\n",
-        "    train_y = train_y_noiseless + noise_std * torch.randn_like(train_y_noiseless)\n",
-        "    train_yvar = torch.full_like(train_y, noise_std**2)\n",
-        "    vanilla_nei_train_x = train_x.clone()\n",
-        "    vanilla_nei_train_y = train_y.clone()\n",
-        "    vanilla_nei_train_yvar = train_yvar.clone()\n",
-        "    # keep track of the best observed point at each iteration\n",
-        "    best_value = train_y.max().item()\n",
-        "    best_rgpe.append(best_value)\n",
-        "    best_random.append(best_value)\n",
-        "    vanilla_nei_best_value = best_value\n",
-        "    best_vanilla_nei.append(vanilla_nei_best_value)\n",
-        "\n",
-        "    # Run N_BATCH rounds of BayesOpt after the initial random batch\n",
-        "    for iteration in range(N_BATCH):\n",
-        "        target_model = get_fitted_model(train_x, train_y, train_yvar)\n",
-        "        model_list = base_model_list + [target_model]\n",
-        "        rank_weights = compute_rank_weights(\n",
-        "            train_x,\n",
-        "            train_y,\n",
-        "            base_model_list,\n",
-        "            target_model,\n",
-        "            NUM_POSTERIOR_SAMPLES,\n",
-        "        )\n",
-        "\n",
-        "        # create model and acquisition function\n",
-        "        rgpe_model = RGPE(model_list, rank_weights)\n",
-        "        sampler_qnei = SobolQMCNormalSampler(sample_shape=torch.Size([MC_SAMPLES]))\n",
-        "        qNEI = qLogNoisyExpectedImprovement(\n",
-        "            model=rgpe_model,\n",
-        "            X_baseline=train_x,\n",
-        "            sampler=sampler_qnei,\n",
-        "            prune_baseline=False,\n",
-        "        )\n",
-        "\n",
-        "        # optimize\n",
-        "        candidate, _ = optimize_acqf(\n",
-        "            acq_function=qNEI,\n",
-        "            bounds=torch.tensor([[0.0], [1.0]], dtype=dtype, device=device),\n",
-        "            q=Q_BATCH_SIZE,\n",
-        "            num_restarts=N_RESTARTS,\n",
-        "            raw_samples=N_RESTART_CANDIDATES,\n",
-        "        )\n",
-        "\n",
-        "        # fetch the new values\n",
-        "        new_x = candidate.detach()\n",
-        "        new_y_noiseless = f(unnormalize(new_x, bounds=BOUNDS))\n",
-        "        new_y = new_y_noiseless + noise_std * torch.randn_like(new_y_noiseless)\n",
-        "        new_yvar = torch.full_like(new_y, noise_std**2)\n",
-        "\n",
-        "        # update training points\n",
-        "        train_x = torch.cat((train_x, new_x))\n",
-        "        train_y = torch.cat((train_y, new_y))\n",
-        "        train_yvar = torch.cat((train_yvar, new_yvar))\n",
-        "        random_candidate = torch.rand(1, dtype=dtype, device=device)\n",
-        "        next_random_noiseless = f(unnormalize(random_candidate, bounds=BOUNDS))\n",
-        "        next_random = next_random_noiseless + noise_std * torch.randn_like(\n",
-        "            next_random_noiseless\n",
-        "        )\n",
-        "        next_random_best = next_random.max().item()\n",
-        "        best_random.append(max(best_random[-1], next_random_best))\n",
-        "\n",
-        "        # get the new best observed value\n",
-        "        best_value = train_y.max().item()\n",
-        "        best_rgpe.append(best_value)\n",
-        "\n",
-        "        # Run Vanilla NEI for comparison\n",
-        "        vanilla_nei_model = get_fitted_model(\n",
-        "            vanilla_nei_train_x,\n",
-        "            vanilla_nei_train_y,\n",
-        "            vanilla_nei_train_yvar,\n",
-        "        )\n",
-        "        vanilla_nei_sampler = SobolQMCNormalSampler(\n",
-        "            sample_shape=torch.Size([MC_SAMPLES])\n",
-        "        )\n",
-        "        vanilla_qNEI = qLogNoisyExpectedImprovement(\n",
-        "            model=vanilla_nei_model,\n",
-        "            X_baseline=vanilla_nei_train_x,\n",
-        "            sampler=vanilla_nei_sampler,\n",
-        "        )\n",
-        "        vanilla_nei_candidate, _ = optimize_acqf(\n",
-        "            acq_function=vanilla_qNEI,\n",
-        "            bounds=torch.tensor([[0.0], [1.0]], dtype=dtype, device=device),\n",
-        "            q=Q_BATCH_SIZE,\n",
-        "            num_restarts=N_RESTARTS,\n",
-        "            raw_samples=N_RESTART_CANDIDATES,\n",
-        "        )\n",
-        "        # fetch the new values\n",
-        "        vanilla_nei_new_x = vanilla_nei_candidate.detach()\n",
-        "        vanilla_nei_new_y_noiseless = f(unnormalize(vanilla_nei_new_x, bounds=BOUNDS))\n",
-        "        vanilla_nei_new_y = vanilla_nei_new_y_noiseless + noise_std * torch.randn_like(\n",
-        "            new_y_noiseless\n",
-        "        )\n",
-        "        vanilla_nei_new_yvar = torch.full_like(vanilla_nei_new_y, noise_std**2)\n",
-        "\n",
-        "        # update training points\n",
-        "        vanilla_nei_train_x = torch.cat([vanilla_nei_train_x, vanilla_nei_new_x])\n",
-        "        vanilla_nei_train_y = torch.cat([vanilla_nei_train_y, vanilla_nei_new_y])\n",
-        "        vanilla_nei_train_yvar = torch.cat(\n",
-        "            [vanilla_nei_train_yvar, vanilla_nei_new_yvar]\n",
-        "        )\n",
-        "\n",
-        "        # get the new best observed value\n",
-        "        vanilla_nei_best_value = vanilla_nei_train_y.max().item()\n",
-        "        best_vanilla_nei.append(vanilla_nei_best_value)\n",
-        "\n",
-        "    best_rgpe_all.append(best_rgpe)\n",
-        "    best_random_all.append(best_random)\n",
-        "    best_vanilla_nei_all.append(best_vanilla_nei)"
-      ],
-      "execution_count": 18,
-      "outputs": [
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 1 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 2 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 3 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 4 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 5 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "[W 240829 09:21:28 optimize:564] Optimization failed in `gen_candidates_scipy` with the following warning(s):\n    [OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.')]\n    Trying again with a new set of initial conditions.\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "[W 240829 09:21:28 optimize:564] Optimization failed on the second try, after generating a new set of initial conditions.\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 6 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 7 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 8 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 9 of 10\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "[W 240829 09:21:46 optimize:564] Optimization failed in `gen_candidates_scipy` with the following warning(s):\n    [OptimizationWarning('Optimization failed within `scipy.optimize.minimize` with status 2 and message ABNORMAL_TERMINATION_IN_LNSRCH.')]\n    Trying again with a new set of initial conditions.\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stderr",
-          "text": [
-            "[W 240829 09:21:46 optimize:564] Optimization failed on the second try, after generating a new set of initial conditions.\n"
-          ]
-        },
-        {
-          "output_type": "stream",
-          "name": "stdout",
-          "text": [
-            "Trial 10 of 10\n"
-          ]
-        }
-      ]
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[W 240829 09:21:46 optimize:564] Optimization failed on the second try, after generating a new set of initial conditions.\n"
+     ]
     },
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "originalKey": "3dbf06b6-28ec-4b73-99f3-a9327b074159",
-        "showInput": false,
-        "outputsInitialized": false,
-        "language": "markdown"
-      },
-      "source": [
-        "#### Plot best observed value vs iteration"
-      ]
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Trial 10 of 10\n"
+     ]
+    }
+   ],
+   "source": [
+    "# suppress GPyTorch warnings about adding jitter\n",
+    "import warnings\n",
+    "\n",
+    "from botorch.acquisition.logei import qLogNoisyExpectedImprovement\n",
+    "from botorch.optim.optimize import optimize_acqf\n",
+    "from botorch.sampling.normal import SobolQMCNormalSampler\n",
+    "\n",
+    "\n",
+    "warnings.filterwarnings(\"ignore\", \"^.*jitter.*\", category=RuntimeWarning)\n",
+    "\n",
+    "\n",
+    "best_rgpe_all = []\n",
+    "best_random_all = []\n",
+    "best_vanilla_nei_all = []\n",
+    "N_BATCH = 10 if not SMOKE_TEST else 2\n",
+    "NUM_POSTERIOR_SAMPLES = 256 if not SMOKE_TEST else 16\n",
+    "RANDOM_INITIALIZATION_SIZE = 3\n",
+    "N_TRIALS = 10 if not SMOKE_TEST else 2\n",
+    "MC_SAMPLES = 512 if not SMOKE_TEST else 32\n",
+    "N_RESTART_CANDIDATES = 512 if not SMOKE_TEST else 8\n",
+    "N_RESTARTS = 10 if not SMOKE_TEST else 2\n",
+    "Q_BATCH_SIZE = 1\n",
+    "\n",
+    "\n",
+    "# Average over multiple trials\n",
+    "for trial in range(N_TRIALS):\n",
+    "    print(f\"Trial {trial + 1} of {N_TRIALS}\")\n",
+    "    best_rgpe = []\n",
+    "    best_random = []\n",
+    "    best_vanilla_nei = []\n",
+    "    # Initial random observations\n",
+    "    raw_x = draw_sobol_samples(\n",
+    "        bounds=BOUNDS, n=RANDOM_INITIALIZATION_SIZE, q=1, seed=trial\n",
+    "    ).squeeze(1)\n",
+    "    train_x = normalize(raw_x, bounds=BOUNDS)\n",
+    "    train_y_noiseless = f(raw_x)\n",
+    "    train_y = train_y_noiseless + noise_std * torch.randn_like(train_y_noiseless)\n",
+    "    train_yvar = torch.full_like(train_y, noise_std**2)\n",
+    "    vanilla_nei_train_x = train_x.clone()\n",
+    "    vanilla_nei_train_y = train_y.clone()\n",
+    "    vanilla_nei_train_yvar = train_yvar.clone()\n",
+    "    # keep track of the best observed point at each iteration\n",
+    "    best_value = train_y.max().item()\n",
+    "    best_rgpe.append(best_value)\n",
+    "    best_random.append(best_value)\n",
+    "    vanilla_nei_best_value = best_value\n",
+    "    best_vanilla_nei.append(vanilla_nei_best_value)\n",
+    "\n",
+    "    # Run N_BATCH rounds of BayesOpt after the initial random batch\n",
+    "    for iteration in range(N_BATCH):\n",
+    "        target_model = get_fitted_model(train_x, train_y, train_yvar)\n",
+    "        model_list = base_model_list + [target_model]\n",
+    "        rank_weights = compute_rank_weights(\n",
+    "            train_x,\n",
+    "            train_y,\n",
+    "            base_model_list,\n",
+    "            target_model,\n",
+    "            NUM_POSTERIOR_SAMPLES,\n",
+    "        )\n",
+    "\n",
+    "        # create model and acquisition function\n",
+    "        rgpe_model = RGPE(model_list, rank_weights)\n",
+    "        sampler_qnei = SobolQMCNormalSampler(sample_shape=torch.Size([MC_SAMPLES]))\n",
+    "        qNEI = qLogNoisyExpectedImprovement(\n",
+    "            model=rgpe_model,\n",
+    "            X_baseline=train_x,\n",
+    "            sampler=sampler_qnei,\n",
+    "            prune_baseline=False,\n",
+    "        )\n",
+    "\n",
+    "        # optimize\n",
+    "        candidate, _ = optimize_acqf(\n",
+    "            acq_function=qNEI,\n",
+    "            bounds=torch.tensor([[0.0], [1.0]], dtype=dtype, device=device),\n",
+    "            q=Q_BATCH_SIZE,\n",
+    "            num_restarts=N_RESTARTS,\n",
+    "            raw_samples=N_RESTART_CANDIDATES,\n",
+    "        )\n",
+    "\n",
+    "        # fetch the new values\n",
+    "        new_x = candidate.detach()\n",
+    "        new_y_noiseless = f(unnormalize(new_x, bounds=BOUNDS))\n",
+    "        new_y = new_y_noiseless + noise_std * torch.randn_like(new_y_noiseless)\n",
+    "        new_yvar = torch.full_like(new_y, noise_std**2)\n",
+    "\n",
+    "        # update training points\n",
+    "        train_x = torch.cat((train_x, new_x))\n",
+    "        train_y = torch.cat((train_y, new_y))\n",
+    "        train_yvar = torch.cat((train_yvar, new_yvar))\n",
+    "        random_candidate = torch.rand(1, dtype=dtype, device=device)\n",
+    "        next_random_noiseless = f(unnormalize(random_candidate, bounds=BOUNDS))\n",
+    "        next_random = next_random_noiseless + noise_std * torch.randn_like(\n",
+    "            next_random_noiseless\n",
+    "        )\n",
+    "        next_random_best = next_random.max().item()\n",
+    "        best_random.append(max(best_random[-1], next_random_best))\n",
+    "\n",
+    "        # get the new best observed value\n",
+    "        best_value = train_y.max().item()\n",
+    "        best_rgpe.append(best_value)\n",
+    "\n",
+    "        # Run Vanilla NEI for comparison\n",
+    "        vanilla_nei_model = get_fitted_model(\n",
+    "            vanilla_nei_train_x,\n",
+    "            vanilla_nei_train_y,\n",
+    "            vanilla_nei_train_yvar,\n",
+    "        )\n",
+    "        vanilla_nei_sampler = SobolQMCNormalSampler(\n",
+    "            sample_shape=torch.Size([MC_SAMPLES])\n",
+    "        )\n",
+    "        vanilla_qNEI = qLogNoisyExpectedImprovement(\n",
+    "            model=vanilla_nei_model,\n",
+    "            X_baseline=vanilla_nei_train_x,\n",
+    "            sampler=vanilla_nei_sampler,\n",
+    "        )\n",
+    "        vanilla_nei_candidate, _ = optimize_acqf(\n",
+    "            acq_function=vanilla_qNEI,\n",
+    "            bounds=torch.tensor([[0.0], [1.0]], dtype=dtype, device=device),\n",
+    "            q=Q_BATCH_SIZE,\n",
+    "            num_restarts=N_RESTARTS,\n",
+    "            raw_samples=N_RESTART_CANDIDATES,\n",
+    "        )\n",
+    "        # fetch the new values\n",
+    "        vanilla_nei_new_x = vanilla_nei_candidate.detach()\n",
+    "        vanilla_nei_new_y_noiseless = f(unnormalize(vanilla_nei_new_x, bounds=BOUNDS))\n",
+    "        vanilla_nei_new_y = vanilla_nei_new_y_noiseless + noise_std * torch.randn_like(\n",
+    "            new_y_noiseless\n",
+    "        )\n",
+    "        vanilla_nei_new_yvar = torch.full_like(vanilla_nei_new_y, noise_std**2)\n",
+    "\n",
+    "        # update training points\n",
+    "        vanilla_nei_train_x = torch.cat([vanilla_nei_train_x, vanilla_nei_new_x])\n",
+    "        vanilla_nei_train_y = torch.cat([vanilla_nei_train_y, vanilla_nei_new_y])\n",
+    "        vanilla_nei_train_yvar = torch.cat(\n",
+    "            [vanilla_nei_train_yvar, vanilla_nei_new_yvar]\n",
+    "        )\n",
+    "\n",
+    "        # get the new best observed value\n",
+    "        vanilla_nei_best_value = vanilla_nei_train_y.max().item()\n",
+    "        best_vanilla_nei.append(vanilla_nei_best_value)\n",
+    "\n",
+    "    best_rgpe_all.append(best_rgpe)\n",
+    "    best_random_all.append(best_random)\n",
+    "    best_vanilla_nei_all.append(best_vanilla_nei)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "language": "markdown",
+    "originalKey": "3dbf06b6-28ec-4b73-99f3-a9327b074159",
+    "outputsInitialized": false,
+    "showInput": false
+   },
+   "source": [
+    "#### Plot best observed value vs iteration"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {
+    "collapsed": false,
+    "customOutput": null,
+    "executionStartTime": 1724948509190,
+    "executionStopTime": 1724948514271,
+    "jupyter": {
+     "outputs_hidden": false
     },
+    "language": "python",
+    "originalKey": "77d08a71-caf0-444c-994c-94a0f63efc42",
+    "outputsInitialized": true,
+    "requestMsgId": "77d08a71-caf0-444c-994c-94a0f63efc42",
+    "serverExecutionDuration": 412.50938293524
+   },
+   "outputs": [
     {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "77d08a71-caf0-444c-994c-94a0f63efc42",
-        "collapsed": false,
-        "requestMsgId": "77d08a71-caf0-444c-994c-94a0f63efc42",
-        "customOutput": null,
-        "executionStartTime": 1724948509190,
-        "executionStopTime": 1724948514271,
-        "outputsInitialized": true,
-        "language": "python",
-        "serverExecutionDuration": 412.50938293524
-      },
-      "source": [
-        "import numpy as np\n",
-        "\n",
-        "\n",
-        "best_rgpe_all = np.array(best_rgpe_all)\n",
-        "best_random_all = np.array(best_random_all)\n",
-        "best_vanilla_nei_all = np.array(best_vanilla_nei_all)\n",
-        "\n",
-        "x = range(RANDOM_INITIALIZATION_SIZE, RANDOM_INITIALIZATION_SIZE + N_BATCH + 1)\n",
-        "\n",
-        "fig, ax = plt.subplots(1, 1, figsize=(10, 6))\n",
-        "# Plot RGPE - LogNEI\n",
-        "ax.errorbar(\n",
-        "    x,\n",
-        "    best_rgpe_all.mean(axis=0),\n",
-        "    yerr=1.96 * best_rgpe_all.std(axis=0) / math.sqrt(N_TRIALS),\n",
-        "    label=\"RGPE - LogNEI\",\n",
-        "    linewidth=3,\n",
-        "    capsize=5,\n",
-        "    capthick=3,\n",
-        ")\n",
-        "# Plot SingleTaskGP - LogNEI\n",
-        "ax.errorbar(\n",
-        "    x,\n",
-        "    best_vanilla_nei_all.mean(axis=0),\n",
-        "    yerr=1.96 * best_vanilla_nei_all.std(axis=0) / math.sqrt(N_TRIALS),\n",
-        "    label=\"SingleTaskGP - LogNEI\",\n",
-        "    linewidth=3,\n",
-        "    capsize=5,\n",
-        "    capthick=3,\n",
-        ")\n",
-        "# Plot Random\n",
-        "ax.errorbar(\n",
-        "    x,\n",
-        "    best_random_all.mean(axis=0),\n",
-        "    yerr=1.96 * best_random_all.std(axis=0) / math.sqrt(N_TRIALS),\n",
-        "    label=\"Random\",\n",
-        "    linewidth=3,\n",
-        "    capsize=5,\n",
-        "    capthick=3,\n",
-        ")\n",
-        "ax.set_ylim(bottom=0)\n",
-        "ax.set_xlabel(\"Iteration\", fontsize=12)\n",
-        "ax.set_ylabel(\"Best Observed Value\", fontsize=12)\n",
-        "ax.set_title(\"Best Observed Value by Iteration\", fontsize=12)\n",
-        "ax.legend(loc=\"lower right\", fontsize=10)\n",
-        "plt.tight_layout()"
-      ],
-      "execution_count": 19,
-      "outputs": [
-        {
-          "output_type": "display_data",
-          "data": {
-            "text/plain": "<Figure size 1000x600 with 1 Axes>",
-            "image/png": "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"
-          },
-          "metadata": {}
-        }
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
       ]
-    },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "originalKey": "bc7cf5ae-bdf2-465c-b918-a69c8f2e3e8f",
-        "showInput": true,
-        "customInput": null,
-        "language": "python"
-      },
-      "source": [
-        ""
-      ],
-      "execution_count": null,
-      "outputs": []
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
-  ]
+   ],
+   "source": [
+    "import numpy as np\n",
+    "\n",
+    "\n",
+    "best_rgpe_all = np.array(best_rgpe_all)\n",
+    "best_random_all = np.array(best_random_all)\n",
+    "best_vanilla_nei_all = np.array(best_vanilla_nei_all)\n",
+    "\n",
+    "x = range(RANDOM_INITIALIZATION_SIZE, RANDOM_INITIALIZATION_SIZE + N_BATCH + 1)\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 1, figsize=(10, 6))\n",
+    "# Plot RGPE - LogNEI\n",
+    "ax.errorbar(\n",
+    "    x,\n",
+    "    best_rgpe_all.mean(axis=0),\n",
+    "    yerr=1.96 * best_rgpe_all.std(axis=0) / math.sqrt(N_TRIALS),\n",
+    "    label=\"RGPE - LogNEI\",\n",
+    "    linewidth=3,\n",
+    "    capsize=5,\n",
+    "    capthick=3,\n",
+    ")\n",
+    "# Plot SingleTaskGP - LogNEI\n",
+    "ax.errorbar(\n",
+    "    x,\n",
+    "    best_vanilla_nei_all.mean(axis=0),\n",
+    "    yerr=1.96 * best_vanilla_nei_all.std(axis=0) / math.sqrt(N_TRIALS),\n",
+    "    label=\"SingleTaskGP - LogNEI\",\n",
+    "    linewidth=3,\n",
+    "    capsize=5,\n",
+    "    capthick=3,\n",
+    ")\n",
+    "# Plot Random\n",
+    "ax.errorbar(\n",
+    "    x,\n",
+    "    best_random_all.mean(axis=0),\n",
+    "    yerr=1.96 * best_random_all.std(axis=0) / math.sqrt(N_TRIALS),\n",
+    "    label=\"Random\",\n",
+    "    linewidth=3,\n",
+    "    capsize=5,\n",
+    "    capthick=3,\n",
+    ")\n",
+    "ax.set_ylim(bottom=0)\n",
+    "ax.set_xlabel(\"Iteration\", fontsize=12)\n",
+    "ax.set_ylabel(\"Best Observed Value\", fontsize=12)\n",
+    "ax.set_title(\"Best Observed Value by Iteration\", fontsize=12)\n",
+    "ax.legend(loc=\"lower right\", fontsize=10)\n",
+    "plt.tight_layout()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "customInput": null,
+    "language": "python",
+    "originalKey": "bc7cf5ae-bdf2-465c-b918-a69c8f2e3e8f",
+    "showInput": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.14"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
 }