{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "2438a1d7-6564-4a4d-bb8a-5ae7f3eba552",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Visualizing Population Based Training (PBT) Hyperparameter Optimization\n",
    "\n",
    "**Assumptions:** The reader has a basic understanding of the [PBT algorithm](https://www.deepmind.com/blog/population-based-training-of-neural-networks) and wants to dive deeper and verify the underlying algorithm behavior with [Ray's PBT implementation](tune-scheduler-pbt). [This guide](pbt-guide-ref) provides resources for gaining some context.\n",
    "\n",
    "Population Based Training (PBT) is a powerful technique that combines parallel search with sequential optimization to efficiently find optimal hyperparameters. Unlike traditional hyperparameter tuning methods, PBT dynamically adjusts hyperparameters during training by having multiple training runs (\"trials\") that evolve together, periodically replacing poorly performing configurations with perturbations of better ones.\n",
    "\n",
    "This tutorial will go through a simple example that will help you develop a better understanding of what PBT is doing under the hood when using it to tune your algorithms.\n",
    "\n",
    "We will learn how to:\n",
    "\n",
    "1. **Set up checkpointing and loading for PBT** with the function trainable interface\n",
    "2. **Configure Tune and PBT scheduler parameters**\n",
    "3. **Visualize PBT algorithm behavior** to gain some intuition\n",
    "\n",
    "## Set up Toy the Example\n",
    "\n",
    "The toy example optimization problem we will use comes from the [PBT paper](https://arxiv.org/pdf/1711.09846.pdf) (see Figure 2 for more details). The goal is to find parameters that maximize an quadratic function, while only having access to an estimator that depends on a set of hyperparameters. A practical example of this is maximizing the (unknown) generalization capabilities of a model across all possible inputs with only access to the empirical loss of your model, which depends on hyperparameters in order to optimize.\n",
    "\n",
    "We'll start with some imports."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "49b2e7ba-532b-431e-aa81-1467cb2b4e70",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install -q -U \"ray[tune]\" matplotlib"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "efec7627-fd60-48e9-8214-0b4fbb8e4402",
   "metadata": {},
   "source": [
    "Note: this tutorial imports functions from {doc}`this helper file </tune/examples/pbt_visualization/pbt_visualization_utils>` named `pbt_visualization_utils.py`. These define plotting functions for the PBT training progress."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "90471b91",
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-02-24 16:21:26,622\tINFO util.py:154 -- Missing packages: ['ipywidgets']. Run `pip install -U ipywidgets`, then restart the notebook server for rich notebook output.\n",
      "2025-02-24 16:21:26,890\tINFO util.py:154 -- Missing packages: ['ipywidgets']. Run `pip install -U ipywidgets`, then restart the notebook server for rich notebook output.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import pickle\n",
    "import tempfile\n",
    "\n",
    "import ray\n",
    "from ray import tune\n",
    "from ray.tune.schedulers import PopulationBasedTraining\n",
    "from ray.tune.tune_config import TuneConfig\n",
    "from ray.tune.tuner import Tuner\n",
    "\n",
    "from pbt_visualization_utils import (\n",
    "    get_init_theta,\n",
    "    plot_parameter_history,\n",
    "    plot_Q_history,\n",
    "    make_animation,\n",
    ")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "a223d6a2-a7d5-40a1-8e12-2a5a1a0a0070",
   "metadata": {},
   "source": [
    "Concretely, we will use the definitions (with very minor modifications) provided in the [paper](https://arxiv.org/pdf/1711.09846.pdf) for the function we are trying to optimize, and the estimator we are given.\n",
    "\n",
    "Our goal is to maximize a quadratic function `Q`, but we only have access to a biased estimator `Qhat` that depends on hyperparameters. This simulates real-world scenarios where we want to optimize for true generalization performance but can only measure training performance, which is influenced by hyperparameters.\n",
    "\n",
    "\n",
    "Here is a list of the concepts we will use for the example, and what they might be analagous to in practice:\n",
    "\n",
    "| Symbol | In This Example | Real-World Analogy |\n",
    "|---------|-------------|-------------------|\n",
    "|`theta = [theta0, theta1]`| Model parameters, updated in each training step.|Neural network parameters|\n",
    "|`h = [h0, h1]`| The hyperparameters optimized by PBT. | Learning rate, batch size, etc.|\n",
    "|`Q(theta)`| **True reward function** we *want* to optimize, but is not directly use for training.|**True generalization**-- an theoretical and unobersvable in practice.|\n",
    "|`Qhat(theta \\| h)`| **Estimated reward function** we actually optimize against; depends on the hyperparameters as well as the model parameters.|**Empirical reward** in training.|\n",
    "|`grad_Qhat(theta \\| h)`| Gradient of the estimated reward function, used to update model parameters | Gradient descent step in training | \n",
    "\n",
    "Below are the implementations in code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a75e75db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initial parameter values: theta = [0.9 0.9]\n"
     ]
    }
   ],
   "source": [
    "def Q(theta):\n",
    "    # equation for an elliptic paraboloid with a center at (0, 0, 1.2)\n",
    "    return 1.2 - (3 / 4 * theta[0] ** 2 + theta[1] ** 2)\n",
    "\n",
    "\n",
    "def Qhat(theta, h):\n",
    "    return 1.2 - (h[0] * theta[0] ** 2 + h[1] * theta[1] ** 2)\n",
    "\n",
    "\n",
    "def grad_Qhat(theta, h):\n",
    "    theta_grad = -2 * h * theta\n",
    "    theta_grad[0] *= 3 / 4\n",
    "    h_grad = -np.square(theta)\n",
    "    h_grad[0] *= 3 / 4\n",
    "    return {\"theta\": theta_grad, \"h\": h_grad}\n",
    "\n",
    "\n",
    "theta_0 = get_init_theta()\n",
    "print(f\"Initial parameter values: theta = {theta_0}\")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "0ee21632-9be6-4f80-ac80-c71696cb0f4f",
   "metadata": {},
   "source": [
    "## Defining the Function Trainable\n",
    "\n",
    "We will define the training loop:\n",
    "1. Load the hyperparameter configuration\n",
    "2. Initialize the model, **resuming from a checkpoint if one exists (this is important for PBT, since the scheduler will pause and resume trials frequently when trials get exploited).**\n",
    "3. Run the training loop and **checkpoint.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2d1a9fb5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_func(config):\n",
    "    # Load the hyperparam config passed in by the Tuner\n",
    "    h0 = config.get(\"h0\")\n",
    "    h1 = config.get(\"h1\")\n",
    "    h = np.array([h0, h1]).astype(float)\n",
    "\n",
    "    lr = config.get(\"lr\")\n",
    "    train_step = 1\n",
    "    checkpoint_interval = config.get(\"checkpoint_interval\", 1)\n",
    "\n",
    "    # Initialize the model parameters\n",
    "    theta = get_init_theta()\n",
    "\n",
    "    # Load a checkpoint if it exists\n",
    "    # This checkpoint could be a trial's own checkpoint to resume,\n",
    "    # or another trial's checkpoint placed by PBT that we will exploit\n",
    "    checkpoint = tune.get_checkpoint()\n",
    "    if checkpoint:\n",
    "        with checkpoint.as_directory() as checkpoint_dir:\n",
    "            with open(os.path.join(checkpoint_dir, \"checkpoint.pkl\"), \"rb\") as f:\n",
    "                checkpoint_dict = pickle.load(f)\n",
    "        # Load in model (theta)\n",
    "        theta = checkpoint_dict[\"theta\"]\n",
    "        last_step = checkpoint_dict[\"train_step\"]\n",
    "        train_step = last_step + 1\n",
    "\n",
    "    # Main training loop (trial stopping is configured later)\n",
    "    while True:\n",
    "        # Perform gradient ascent steps\n",
    "        param_grads = grad_Qhat(theta, h)\n",
    "        theta_grad = np.asarray(param_grads[\"theta\"])\n",
    "        theta = theta + lr * theta_grad\n",
    "\n",
    "        # Define which custom metrics we want in our trial result\n",
    "        result = {\n",
    "            \"Q\": Q(theta),\n",
    "            \"theta0\": theta[0],\n",
    "            \"theta1\": theta[1],\n",
    "            \"h0\": h0,\n",
    "            \"h1\": h1,\n",
    "            \"train_step\": train_step,\n",
    "        }\n",
    "\n",
    "        # Checkpoint every `checkpoint_interval` steps\n",
    "        should_checkpoint = train_step % checkpoint_interval == 0\n",
    "        with tempfile.TemporaryDirectory() as temp_checkpoint_dir:\n",
    "            checkpoint = None\n",
    "            if should_checkpoint:\n",
    "                checkpoint_dict = {\n",
    "                    \"h\": h,\n",
    "                    \"train_step\": train_step,\n",
    "                    \"theta\": theta,\n",
    "                }\n",
    "                with open(\n",
    "                    os.path.join(temp_checkpoint_dir, \"checkpoint.pkl\"), \"wb\"\n",
    "                ) as f:\n",
    "                    pickle.dump(checkpoint_dict, f)\n",
    "                checkpoint = tune.Checkpoint.from_directory(temp_checkpoint_dir)\n",
    "\n",
    "            # Report metric for this training iteration, and include the\n",
    "            # trial checkpoint that contains the current parameters if we\n",
    "            # saved it this train step\n",
    "            tune.report(result, checkpoint=checkpoint)\n",
    "\n",
    "        train_step += 1\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "5bdc96e0-b4bf-4a7a-9f15-e94de6f4d21b",
   "metadata": {},
   "source": [
    "```{note}\n",
    "Since PBT will keep restoring from latest checkpoints, it's important to save and load `train_step` correctly in a function trainable. **Make sure you increment the loaded `train_step` by one as shown above in `checkpoint_dict`.** This avoids repeating an iteration and causing the checkpoint and perturbation intervals to be out of sync.\n",
    "\n",
    "```"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "caa002e2-1d68-404c-84bd-99b8d8119dac",
   "metadata": {},
   "source": [
    "## Configure PBT and Tuner\n",
    "\n",
    "We start by initializing ray (shutting it down if a session existed previously)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f68445a3-958f-49a0-a9f9-03121c3c731c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-02-24 16:21:27,556\tINFO worker.py:1841 -- Started a local Ray instance.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div class=\"lm-Widget p-Widget lm-Panel p-Panel jp-Cell-outputWrapper\">\n",
       "    <div style=\"margin-left: 50px;display: flex;flex-direction: row;align-items: center\">\n",
       "        <div class=\"jp-RenderedHTMLCommon\" style=\"display: flex; flex-direction: row;\">\n",
       "  <svg viewBox=\"0 0 567 224\" fill=\"none\" xmlns=\"http://www.w3.org/2000/svg\" style=\"height: 3em;\">\n",
       "    <g clip-path=\"url(#clip0_4338_178347)\">\n",
       "        <path d=\"M341.29 165.561H355.29L330.13 129.051C345.63 123.991 354.21 112.051 354.21 94.2307C354.21 71.3707 338.72 58.1807 311.88 58.1807H271V165.561H283.27V131.661H311.8C314.25 131.661 316.71 131.501 319.01 131.351L341.25 165.561H341.29ZM283.29 119.851V70.0007H311.82C331.3 70.0007 342.34 78.2907 342.34 94.5507C342.34 111.271 331.34 119.861 311.82 119.861L283.29 119.851ZM451.4 138.411L463.4 165.561H476.74L428.74 58.1807H416L367.83 165.561H380.83L392.83 138.411H451.4ZM446.19 126.601H398L422 72.1407L446.24 126.601H446.19ZM526.11 128.741L566.91 58.1807H554.35L519.99 114.181L485.17 58.1807H472.44L514.01 129.181V165.541H526.13V128.741H526.11Z\" fill=\"var(--jp-ui-font-color0)\"/>\n",
       "        <path d=\"M82.35 104.44C84.0187 97.8827 87.8248 92.0678 93.1671 87.9146C98.5094 83.7614 105.083 81.5067 111.85 81.5067C118.617 81.5067 125.191 83.7614 130.533 87.9146C135.875 92.0678 139.681 97.8827 141.35 104.44H163.75C164.476 101.562 165.622 98.8057 167.15 96.2605L127.45 56.5605C121.071 60.3522 113.526 61.6823 106.235 60.3005C98.9443 58.9187 92.4094 54.9203 87.8602 49.0574C83.3109 43.1946 81.0609 35.8714 81.5332 28.4656C82.0056 21.0599 85.1679 14.0819 90.4252 8.8446C95.6824 3.60726 102.672 0.471508 110.08 0.0272655C117.487 -0.416977 124.802 1.86091 130.647 6.4324C136.493 11.0039 140.467 17.5539 141.821 24.8501C143.175 32.1463 141.816 39.6859 138 46.0505L177.69 85.7505C182.31 82.9877 187.58 81.4995 192.962 81.4375C198.345 81.3755 203.648 82.742 208.33 85.3976C213.012 88.0532 216.907 91.9029 219.616 96.5544C222.326 101.206 223.753 106.492 223.753 111.875C223.753 117.258 222.326 122.545 219.616 127.197C216.907 131.848 213.012 135.698 208.33 138.353C203.648 141.009 198.345 142.375 192.962 142.313C187.58 142.251 182.31 140.763 177.69 138L138 177.7C141.808 184.071 143.155 191.614 141.79 198.91C140.424 206.205 136.44 212.75 130.585 217.313C124.731 221.875 117.412 224.141 110.004 223.683C102.596 223.226 95.6103 220.077 90.3621 214.828C85.1139 209.58 81.9647 202.595 81.5072 195.187C81.0497 187.779 83.3154 180.459 87.878 174.605C92.4405 168.751 98.9853 164.766 106.281 163.401C113.576 162.035 121.119 163.383 127.49 167.19L167.19 127.49C165.664 124.941 164.518 122.182 163.79 119.3H141.39C139.721 125.858 135.915 131.673 130.573 135.826C125.231 139.98 118.657 142.234 111.89 142.234C105.123 142.234 98.5494 139.98 93.2071 135.826C87.8648 131.673 84.0587 125.858 82.39 119.3H60C58.1878 126.495 53.8086 132.78 47.6863 136.971C41.5641 141.163 34.1211 142.972 26.7579 142.059C19.3947 141.146 12.6191 137.574 7.70605 132.014C2.79302 126.454 0.0813599 119.29 0.0813599 111.87C0.0813599 104.451 2.79302 97.2871 7.70605 91.7272C12.6191 86.1673 19.3947 82.5947 26.7579 81.6817C34.1211 80.7686 41.5641 82.5781 47.6863 86.7696C53.8086 90.9611 58.1878 97.2456 60 104.44H82.35ZM100.86 204.32C103.407 206.868 106.759 208.453 110.345 208.806C113.93 209.159 117.527 208.258 120.522 206.256C123.517 204.254 125.725 201.276 126.771 197.828C127.816 194.38 127.633 190.677 126.253 187.349C124.874 184.021 122.383 181.274 119.205 179.577C116.027 177.88 112.359 177.337 108.826 178.042C105.293 178.746 102.113 180.654 99.8291 183.44C97.5451 186.226 96.2979 189.718 96.3 193.32C96.2985 195.364 96.7006 197.388 97.4831 199.275C98.2656 201.163 99.4132 202.877 100.86 204.32ZM204.32 122.88C206.868 120.333 208.453 116.981 208.806 113.396C209.159 109.811 208.258 106.214 206.256 103.219C204.254 100.223 201.275 98.0151 197.827 96.97C194.38 95.9249 190.676 96.1077 187.348 97.4873C184.02 98.8669 181.274 101.358 179.577 104.536C177.879 107.714 177.337 111.382 178.041 114.915C178.746 118.448 180.653 121.627 183.439 123.911C186.226 126.195 189.717 127.443 193.32 127.44C195.364 127.443 197.388 127.042 199.275 126.259C201.163 125.476 202.878 124.328 204.32 122.88ZM122.88 19.4205C120.333 16.8729 116.981 15.2876 113.395 14.9347C109.81 14.5817 106.213 15.483 103.218 17.4849C100.223 19.4868 98.0146 22.4654 96.9696 25.9131C95.9245 29.3608 96.1073 33.0642 97.4869 36.3922C98.8665 39.7202 101.358 42.4668 104.535 44.1639C107.713 45.861 111.381 46.4036 114.914 45.6992C118.447 44.9949 121.627 43.0871 123.911 40.301C126.195 37.515 127.442 34.0231 127.44 30.4205C127.44 28.3772 127.038 26.3539 126.255 24.4664C125.473 22.5788 124.326 20.8642 122.88 19.4205ZM19.42 100.86C16.8725 103.408 15.2872 106.76 14.9342 110.345C14.5813 113.93 15.4826 117.527 17.4844 120.522C19.4863 123.518 22.4649 125.726 25.9127 126.771C29.3604 127.816 33.0638 127.633 36.3918 126.254C39.7198 124.874 42.4664 122.383 44.1635 119.205C45.8606 116.027 46.4032 112.359 45.6988 108.826C44.9944 105.293 43.0866 102.114 40.3006 99.8296C37.5145 97.5455 34.0227 96.2983 30.42 96.3005C26.2938 96.3018 22.337 97.9421 19.42 100.86ZM100.86 100.86C98.3125 103.408 96.7272 106.76 96.3742 110.345C96.0213 113.93 96.9226 117.527 98.9244 120.522C100.926 123.518 103.905 125.726 107.353 126.771C110.8 127.816 114.504 127.633 117.832 126.254C121.16 124.874 123.906 122.383 125.604 119.205C127.301 116.027 127.843 112.359 127.139 108.826C126.434 105.293 124.527 102.114 121.741 99.8296C118.955 97.5455 115.463 96.2983 111.86 96.3005C109.817 96.299 107.793 96.701 105.905 97.4835C104.018 98.2661 102.303 99.4136 100.86 100.86Z\" fill=\"#00AEEF\"/>\n",
       "    </g>\n",
       "    <defs>\n",
       "        <clipPath id=\"clip0_4338_178347\">\n",
       "            <rect width=\"566.93\" height=\"223.75\" fill=\"white\"/>\n",
       "        </clipPath>\n",
       "    </defs>\n",
       "  </svg>\n",
       "</div>\n",
       "\n",
       "        <table class=\"jp-RenderedHTMLCommon\" style=\"border-collapse: collapse;color: var(--jp-ui-font-color1);font-size: var(--jp-ui-font-size1);\">\n",
       "    <tr>\n",
       "        <td style=\"text-align: left\"><b>Python version:</b></td>\n",
       "        <td style=\"text-align: left\"><b>3.11.11</b></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"text-align: left\"><b>Ray version:</b></td>\n",
       "        <td style=\"text-align: left\"><b>2.42.1</b></td>\n",
       "    </tr>\n",
       "    \n",
       "</table>\n",
       "\n",
       "    </div>\n",
       "</div>\n"
      ],
      "text/plain": [
       "RayContext(dashboard_url='', python_version='3.11.11', ray_version='2.42.1', ray_commit='c2e38f7b75be223c0c033986472daada8622d64f')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "if ray.is_initialized():\n",
    "    ray.shutdown()\n",
    "ray.init()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "155ec478-4f5d-4614-90a5-1197897cbbcf",
   "metadata": {},
   "source": [
    "### Create the PBT scheduler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4e7d83d6-ecaf-4975-8b56-6c9cd5443d22",
   "metadata": {},
   "outputs": [],
   "source": [
    "perturbation_interval = 4\n",
    "\n",
    "pbt_scheduler = PopulationBasedTraining(\n",
    "    time_attr=\"training_iteration\",\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    metric=\"Q\",\n",
    "    mode=\"max\",\n",
    "    quantile_fraction=0.5,\n",
    "    resample_probability=0.5,\n",
    "    hyperparam_mutations={\n",
    "        \"lr\": tune.qloguniform(5e-3, 1e-1, 5e-4),\n",
    "        \"h0\": tune.uniform(0.0, 1.0),\n",
    "        \"h1\": tune.uniform(0.0, 1.0),\n",
    "    },\n",
    "    synch=True,\n",
    ")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "8143bd8d-b929-4e27-b965-cf852ba3b3d7",
   "metadata": {},
   "source": [
    "A few notes on the PBT config:\n",
    "- `time_attr=\"training_iteration\"` in combination with `perturbation_interval=4` will decide whether a trial should continue or exploit a different trial every 4 training iterations.\n",
    "- `metric=\"Q\"` and `mode=\"max\"` specify how trial performance is ranked. In this case, the high performing trials are the top 50% of trials (set by `quantile_fraction=0.5`) that report the highest `Q` metrics. Note that we could have set the metric/mode in `TuneConfig` instead.\n",
    "- `hyperparam_mutations` specifies that the learning rate `lr` and additional hyperparameters `h0`, `h1` should be perturbed by PBT and defines the resample distribution for each hyperparameter (where `resample_probability=0.5` means that resampling and mutation both happen with 50% probability).\n",
    "- `synch=True` means that PBT will run synchronously, which slows down the algorithm by introducing waits, but it produces more understandable visualizations for the purposes of this tutorial.\n",
    "    - In synchronous PBT, we wait until **all trials** reach the next `perturbation_interval` to decide which trials should continue and which trials should pause and start from the checkpoint of another trials. In the case of 2 trials, this means that every `perturbation_interval` will result in the worse performing trial exploiting the better performing trial.\n",
    "    - This is not always the case in asynchronous PBT, since trials report results and decide whether to continue or exploit **one by one**. This means that a trial could decide that it is a top-performer and decide to continue, since other trials haven't had the chance to report their better results yet. Therefore, we do not always see trials exploiting on every `perturbation_interval`."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "4efe72d7-873d-44c5-9e74-1cd2f41a5c22",
   "metadata": {},
   "source": [
    "### Create the Tuner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c7fa9c92-6ccc-4e8c-91ef-04b95af87a05",
   "metadata": {},
   "outputs": [],
   "source": [
    "tuner = Tuner(\n",
    "    train_func,\n",
    "    param_space={\n",
    "        \"lr\": 0.05,\n",
    "        \"h0\": tune.grid_search([0.0, 1.0]),\n",
    "        \"h1\": tune.sample_from(lambda spec: 1.0 - spec.config[\"h0\"]),\n",
    "        \"num_training_iterations\": 100,\n",
    "        # Match `checkpoint_interval` with `perturbation_interval`\n",
    "        \"checkpoint_interval\": perturbation_interval,\n",
    "    },\n",
    "    tune_config=TuneConfig(\n",
    "        num_samples=1,\n",
    "        # Set the PBT scheduler in this config\n",
    "        scheduler=pbt_scheduler,\n",
    "    ),\n",
    "    run_config=tune.RunConfig(\n",
    "        stop={\"training_iteration\": 100},\n",
    "        failure_config=tune.FailureConfig(max_failures=3),\n",
    "    ),\n",
    ")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "a7407fba-eb82-4cd6-a9cd-bb2adef451df",
   "metadata": {},
   "source": [
    "```{note}\n",
    "We recommend matching `checkpoint_interval` with `perturbation_interval` from the PBT config.\n",
    "This ensures that the PBT algorithm actually exploits the trials in the most recent iteration.\n",
    "\n",
    "If your `perturbation_interval` is large and want to checkpoint more frequently, set `perturbation_interval` to be a multiple of `checkpoint_interval`.\n",
    "```\n",
    "\n",
    "A few other notes on the Tuner config:\n",
    "- `param_space` specifies the *initial* `config` input to our training function. A `grid_search` over two values will launch two trials with a certain set of hyperparameters, and PBT will continue modifying them as training progresses.\n",
    "- The initial hyperparam settings for `h0` and `h1` are configured so that two trials will spawn, one with `h = [1, 0]` and the other with `h = [0, 1]`. This matches the paper experiment and will be used to compare against a `grid_search` baseline that removes the PBT scheduler."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "9221f992-48dc-4cf8-ba9e-3f080a741ee3",
   "metadata": {},
   "source": [
    "## Run the experiment\n",
    "\n",
    "We launch the trials by calling `Tuner.fit`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1559270f",
   "metadata": {
    "scrolled": true,
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2025-02-24 16:22:07</td></tr>\n",
       "<tr><td>Running for: </td><td>00:00:39.86        </td></tr>\n",
       "<tr><td>Memory:      </td><td>21.5/36.0 GiB      </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      PopulationBasedTraining: 24 checkpoints, 24 perturbs<br>Logical resource usage: 1.0/12 CPUs, 0/0 GPUs\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name            </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">     h0</th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">      Q</th><th style=\"text-align: right;\">    theta0</th><th style=\"text-align: right;\">    theta1</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>train_func_74757_00000</td><td>TERMINATED</td><td>127.0.0.1:23555</td><td style=\"text-align: right;\">0.89156</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0432718</td><td style=\"text-align: right;\">1.19993</td><td style=\"text-align: right;\">0.00573655</td><td style=\"text-align: right;\">0.00685687</td></tr>\n",
       "<tr><td>train_func_74757_00001</td><td>TERMINATED</td><td>127.0.0.1:23556</td><td style=\"text-align: right;\">1.11445</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0430496</td><td style=\"text-align: right;\">1.19995</td><td style=\"text-align: right;\">0.0038124 </td><td style=\"text-align: right;\">0.00615009</td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-02-24 16:21:28,081\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "2025-02-24 16:21:28,082\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "2025-02-24 16:21:29,018\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 0.243822) into trial 74757_00001 (score = 0.064403)\n",
      "\n",
      "2025-02-24 16:21:29,018\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.05 --- (resample) --> 0.017\n",
      "h0 : 0.0 --- (* 1.2) --> 0.0\n",
      "h1 : 1.0 --- (resample) --> 0.2659170728716209\n",
      "\n",
      "2025-02-24 16:21:29,795\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:30,572\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:30,579\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 0.442405) into trial 74757_00001 (score = 0.268257)\n",
      "\n",
      "2025-02-24 16:21:30,579\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.05 --- (resample) --> 0.0345\n",
      "h0 : 0.0 --- (resample) --> 0.9170235381005166\n",
      "h1 : 1.0 --- (resample) --> 0.6256279739131234\n",
      "\n",
      "2025-02-24 16:21:31,351\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:32,127\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:32,134\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 0.682806) into trial 74757_00000 (score = 0.527889)\n",
      "\n",
      "2025-02-24 16:21:32,134\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (resample) --> 0.0305\n",
      "h0 : 0.9170235381005166 --- (* 1.2) --> 1.1004282457206198\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.027475735413096558\n",
      "\n",
      "2025-02-24 16:21:32,921\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:33,706\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:33,713\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 0.846848) into trial 74757_00000 (score = 0.823588)\n",
      "\n",
      "2025-02-24 16:21:33,713\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (* 0.8) --> 0.027600000000000003\n",
      "h0 : 0.9170235381005166 --- (* 1.2) --> 1.1004282457206198\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.7558831532799641\n",
      "\n",
      "2025-02-24 16:21:34,498\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:35,346\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:35,353\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 0.958808) into trial 74757_00000 (score = 0.955140)\n",
      "\n",
      "2025-02-24 16:21:35,353\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (* 0.8) --> 0.027600000000000003\n",
      "h0 : 0.9170235381005166 --- (* 1.2) --> 1.1004282457206198\n",
      "h1 : 0.6256279739131234 --- (* 1.2) --> 0.750753568695748\n",
      "\n",
      "2025-02-24 16:21:36,193\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:36,979\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:36,986\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.035238) into trial 74757_00000 (score = 1.032648)\n",
      "\n",
      "2025-02-24 16:21:36,986\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (* 1.2) --> 0.0414\n",
      "h0 : 0.9170235381005166 --- (resample) --> 0.42270740484472435\n",
      "h1 : 0.6256279739131234 --- (* 0.8) --> 0.5005023791304988\n",
      "\n",
      "2025-02-24 16:21:37,808\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/result.json\n",
      "2025-02-24 16:21:38,675\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.087423) into trial 74757_00000 (score = 1.070314)\n",
      "\n",
      "2025-02-24 16:21:38,675\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (resample) --> 0.013000000000000001\n",
      "h0 : 0.9170235381005166 --- (resample) --> 0.2667247790077112\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.7464010779997918\n",
      "\n",
      "2025-02-24 16:21:40,273\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.123062) into trial 74757_00000 (score = 1.094701)\n",
      "\n",
      "2025-02-24 16:21:40,274\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (resample) --> 0.035\n",
      "h0 : 0.9170235381005166 --- (resample) --> 0.6700641473724329\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.09369892963876703\n",
      "\n",
      "2025-02-24 16:21:42,000\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.147406) into trial 74757_00000 (score = 1.138657)\n",
      "\n",
      "2025-02-24 16:21:42,000\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (* 0.8) --> 0.027600000000000003\n",
      "h0 : 0.9170235381005166 --- (* 1.2) --> 1.1004282457206198\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.4113637620174102\n",
      "\n",
      "2025-02-24 16:21:43,617\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.164039) into trial 74757_00000 (score = 1.161962)\n",
      "\n",
      "2025-02-24 16:21:43,618\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (* 0.8) --> 0.027600000000000003\n",
      "h0 : 0.9170235381005166 --- (resample) --> 0.22455715637303986\n",
      "h1 : 0.6256279739131234 --- (* 1.2) --> 0.750753568695748\n",
      "\n",
      "2025-02-24 16:21:45,229\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.175406) into trial 74757_00000 (score = 1.168546)\n",
      "\n",
      "2025-02-24 16:21:45,229\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (resample) --> 0.0075\n",
      "h0 : 0.9170235381005166 --- (* 0.8) --> 0.7336188304804133\n",
      "h1 : 0.6256279739131234 --- (* 1.2) --> 0.750753568695748\n",
      "\n",
      "2025-02-24 16:21:46,822\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.183176) into trial 74757_00000 (score = 1.177124)\n",
      "\n",
      "2025-02-24 16:21:46,823\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (resample) --> 0.016\n",
      "h0 : 0.9170235381005166 --- (resample) --> 0.9850746699152328\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.6345079222898454\n",
      "\n",
      "2025-02-24 16:21:48,411\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.188488) into trial 74757_00000 (score = 1.186006)\n",
      "\n",
      "2025-02-24 16:21:48,411\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0345 --- (resample) --> 0.0545\n",
      "h0 : 0.9170235381005166 --- (resample) --> 0.644936448785508\n",
      "h1 : 0.6256279739131234 --- (resample) --> 0.47452815582611396\n",
      "\n",
      "2025-02-24 16:21:49,978\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.192519) into trial 74757_00001 (score = 1.192121)\n",
      "\n",
      "2025-02-24 16:21:49,978\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (resample) --> 0.006500000000000001\n",
      "h0 : 0.644936448785508 --- (* 0.8) --> 0.5159491590284064\n",
      "h1 : 0.47452815582611396 --- (resample) --> 0.20892073190112748\n",
      "\n",
      "2025-02-24 16:21:51,547\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.195139) into trial 74757_00001 (score = 1.192779)\n",
      "\n",
      "2025-02-24 16:21:51,548\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (resample) --> 0.0405\n",
      "h0 : 0.644936448785508 --- (* 0.8) --> 0.5159491590284064\n",
      "h1 : 0.47452815582611396 --- (* 0.8) --> 0.3796225246608912\n",
      "\n",
      "2025-02-24 16:21:53,193\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.196841) into trial 74757_00001 (score = 1.196227)\n",
      "\n",
      "2025-02-24 16:21:53,194\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (resample) --> 0.043000000000000003\n",
      "h0 : 0.644936448785508 --- (resample) --> 0.8612751379606769\n",
      "h1 : 0.47452815582611396 --- (resample) --> 0.008234170890763504\n",
      "\n",
      "2025-02-24 16:21:54,799\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.197947) into trial 74757_00001 (score = 1.197688)\n",
      "\n",
      "2025-02-24 16:21:54,799\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (* 1.2) --> 0.0654\n",
      "h0 : 0.644936448785508 --- (resample) --> 0.2636264337170955\n",
      "h1 : 0.47452815582611396 --- (* 0.8) --> 0.3796225246608912\n",
      "\n",
      "2025-02-24 16:21:56,428\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.198666) into trial 74757_00001 (score = 1.198417)\n",
      "\n",
      "2025-02-24 16:21:56,429\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (resample) --> 0.0445\n",
      "h0 : 0.644936448785508 --- (* 0.8) --> 0.5159491590284064\n",
      "h1 : 0.47452815582611396 --- (resample) --> 0.4078642041684053\n",
      "\n",
      "2025-02-24 16:21:58,033\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.199133) into trial 74757_00001 (score = 1.198996)\n",
      "\n",
      "2025-02-24 16:21:58,033\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (resample) --> 0.0085\n",
      "h0 : 0.644936448785508 --- (resample) --> 0.21841880940819025\n",
      "h1 : 0.47452815582611396 --- (* 0.8) --> 0.3796225246608912\n",
      "\n",
      "2025-02-24 16:21:59,690\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.199437) into trial 74757_00001 (score = 1.199159)\n",
      "\n",
      "2025-02-24 16:21:59,690\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.0545 --- (* 1.2) --> 0.0654\n",
      "h0 : 0.644936448785508 --- (* 1.2) --> 0.7739237385426097\n",
      "h1 : 0.47452815582611396 --- (resample) --> 0.15770319740458727\n",
      "\n",
      "2025-02-24 16:22:01,361\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.199651) into trial 74757_00000 (score = 1.199634)\n",
      "\n",
      "2025-02-24 16:22:01,362\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.0654 --- (* 0.8) --> 0.052320000000000005\n",
      "h0 : 0.7739237385426097 --- (* 1.2) --> 0.9287084862511316\n",
      "h1 : 0.15770319740458727 --- (resample) --> 0.4279796053289977\n",
      "\n",
      "2025-02-24 16:22:03,081\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.199790) into trial 74757_00001 (score = 1.199772)\n",
      "\n",
      "2025-02-24 16:22:03,082\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.052320000000000005 --- (* 0.8) --> 0.041856000000000004\n",
      "h0 : 0.9287084862511316 --- (resample) --> 0.579167003721271\n",
      "h1 : 0.4279796053289977 --- (* 1.2) --> 0.5135755263947972\n",
      "\n",
      "2025-02-24 16:22:04,698\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00000 (score = 1.199872) into trial 74757_00001 (score = 1.199847)\n",
      "\n",
      "2025-02-24 16:22:04,699\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00001:\n",
      "lr : 0.052320000000000005 --- (* 1.2) --> 0.062784\n",
      "h0 : 0.9287084862511316 --- (* 1.2) --> 1.1144501835013578\n",
      "h1 : 0.4279796053289977 --- (resample) --> 0.25894972559062557\n",
      "\n",
      "2025-02-24 16:22:06,309\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 74757_00001 (score = 1.199924) into trial 74757_00000 (score = 1.199920)\n",
      "\n",
      "2025-02-24 16:22:06,310\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial74757_00000:\n",
      "lr : 0.062784 --- (resample) --> 0.006500000000000001\n",
      "h0 : 1.1144501835013578 --- (* 0.8) --> 0.8915601468010863\n",
      "h1 : 0.25894972559062557 --- (resample) --> 0.4494584110928429\n",
      "\n",
      "2025-02-24 16:22:07,944\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28' in 0.0049s.\n",
      "2025-02-24 16:22:07,946\tINFO tune.py:1041 -- Total run time: 39.88 seconds (39.86 seconds for the tuning loop).\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m(train_func pid=23370)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000000)\n",
      "\u001b[36m(train_func pid=23377)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000000)\n",
      "\u001b[36m(train_func pid=23397)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000004)\u001b[32m [repeated 8x across cluster] (Ray deduplicates logs by default. Set RAY_DEDUP_LOGS=0 to disable log deduplication, or see https://docs.ray.io/en/master/ray-observability/user-guides/configure-logging.html#log-deduplication for more options.)\u001b[0m\n",
      "\u001b[36m(train_func pid=23398)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000001)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23428)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000005)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23428)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000004)\u001b[32m [repeated 6x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23453)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000011)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23453)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000008)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23478)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000014)\u001b[32m [repeated 6x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23479)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000013)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23509)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000018)\u001b[32m [repeated 8x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23509)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000017)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23530)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00000_0_h0=0.0000_2025-02-24_16-21-28/checkpoint_000021)\u001b[32m [repeated 6x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23530)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000011)\u001b[32m [repeated 6x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23556)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000012)\n",
      "\u001b[36m(train_func pid=23556)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-21-28/train_func_74757_00001_1_h0=1.0000_2025-02-24_16-21-28/checkpoint_000013)\n"
     ]
    }
   ],
   "source": [
    "pbt_results = tuner.fit()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "6e0249f7-9e3e-4f2a-9cf7-b0d6c128fc46",
   "metadata": {},
   "source": [
    "## Visualize results\n",
    "\n",
    "Using some helper functions {doc}`from here </tune/examples/pbt_visualization/pbt_visualization_utils>`, we can create some visuals to help us understand the training progression of PBT."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c2af6574",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1300x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 2, figsize=(13, 6), gridspec_kw=dict(width_ratios=[1.5, 1]))\n",
    "\n",
    "colors = [\"red\", \"black\"]\n",
    "labels = [\"h = [1, 0]\", \"h = [0, 1]\"]\n",
    "\n",
    "plot_parameter_history(\n",
    "    pbt_results,\n",
    "    colors,\n",
    "    labels,\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    fig=fig,\n",
    "    ax=axs[0],\n",
    ")\n",
    "plot_Q_history(pbt_results, colors, labels, ax=axs[1])\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "c3d59716-4292-484b-af69-2894fd452505",
   "metadata": {},
   "source": [
    "The plot on the right shows the true function value `Q(theta)` as training progresses for both trials. Both trials reach the maximum value of `1.2`. This demonstrates PBT's ability to find optimal solutions regardless of the initial hyperparameter configuration.\n",
    "\n",
    "Here's how to understand the plot on the left:\n",
    "- The plot on the left shows the parameter values `(theta0, theta1)` on every training iteration, for both trials. As the training iteration increases, the size of the point gets smaller.\n",
    "- We see the iteration shown as a label next to points at every `perturbation_interval` training iterations. Let's zoom into the transition from iteration 4 to 5 for both the trials.\n",
    "    - We see that a trial either **continues** (see how iteration 4 to 5 for the red trial just continues training) or **exploits and perturbs the other trial and then performs a train step** (see how iteration 4 to 5 for the black trial jumps to the parameter value of the red trial).\n",
    "    - The gradient direction also changes at this step for the red trial due to the hyperparameters changing from the exploit and explore steps of PBT. Remember that the gradient of the estimator `Qhat` depends on the hyperparameters `(h0, h1)`.\n",
    "    - The varying size of jumps between training iterations shows that the learning rate is also changing, since we included `lr` in the set of hyperparameters to mutate."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "af6aa641-7f1e-41d1-8e77-c389ad198dab",
   "metadata": {},
   "source": [
    "### Animate the training progress"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "79513127-7705-4e04-947d-29cc9da4b259",
   "metadata": {},
   "outputs": [],
   "source": [
    "make_animation(\n",
    "    pbt_results,\n",
    "    colors,\n",
    "    labels,\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    filename=\"pbt.gif\",\n",
    ")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "6adf9e5b-e26f-4495-a28b-8282feb48a40",
   "metadata": {},
   "source": [
    "We can also animate the training progress to see what's happening to the model parameters at each step. The animation shows:\n",
    "\n",
    "1. How parameters move through space during training\n",
    "2. When exploitation occurs (jumps in parameter space)\n",
    "3. How gradient directions change after hyperparameter perturbation\n",
    "4. Both trials eventually converging to the optimal parameter region\n",
    "\n",
    "![PBT Visualization GIF](pbt.gif)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "ac66d5f0",
   "metadata": {},
   "source": [
    "## Grid Search Comparison\n",
    "\n",
    "The paper includes a comparison to a grid search of 2 trials, using the same initial hyperparameter configurations (`h = [1, 0], h = [0, 1]`) as the PBT experiment. The only difference in the code below is removing the PBT scheduler from the `TuneConfig`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1765efa3",
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2025-02-24 16:22:18</td></tr>\n",
       "<tr><td>Running for: </td><td>00:00:01.24        </td></tr>\n",
       "<tr><td>Memory:      </td><td>21.5/36.0 GiB      </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      Using FIFO scheduling algorithm.<br>Logical resource usage: 1.0/12 CPUs, 0/0 GPUs\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name            </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">  h0</th><th style=\"text-align: right;\">   lr</th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">       Q</th><th style=\"text-align: right;\">     theta0</th><th style=\"text-align: right;\">   theta1</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>train_func_91d06_00000</td><td>TERMINATED</td><td>127.0.0.1:23610</td><td style=\"text-align: right;\">   0</td><td style=\"text-align: right;\">0.015</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.068691 </td><td style=\"text-align: right;\">0.590668</td><td style=\"text-align: right;\">0.9        </td><td style=\"text-align: right;\">0.0427973</td></tr>\n",
       "<tr><td>train_func_91d06_00001</td><td>TERMINATED</td><td>127.0.0.1:23609</td><td style=\"text-align: right;\">   1</td><td style=\"text-align: right;\">0.045</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0659969</td><td style=\"text-align: right;\">0.389999</td><td style=\"text-align: right;\">0.000830093</td><td style=\"text-align: right;\">0.9      </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-02-24 16:22:17,325\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "2025-02-24 16:22:17,326\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000000)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000001)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000002)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000003)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000004)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000005)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000006)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000007)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000008)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000009)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000010)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000011)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000012)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000013)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000014)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000015)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000016)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000017)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000018)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000019)\n",
      "\u001b[36m(train_func pid=23609)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17/train_func_91d06_00001_1_h0=1.0000,lr=0.0450_2025-02-24_16-22-17/checkpoint_000020)\n",
      "2025-02-24 16:22:18,562\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/Users/rdecal/ray_results/train_func_2025-02-24_16-22-17' in 0.0061s.\n",
      "2025-02-24 16:22:18,565\tINFO tune.py:1041 -- Total run time: 1.25 seconds (1.23 seconds for the tuning loop).\n"
     ]
    }
   ],
   "source": [
    "if ray.is_initialized():\n",
    "    ray.shutdown()\n",
    "ray.init()\n",
    "\n",
    "tuner = Tuner(\n",
    "    train_func,\n",
    "    param_space={\n",
    "        \"lr\": tune.qloguniform(1e-2, 1e-1, 5e-3),\n",
    "        \"h0\": tune.grid_search([0.0, 1.0]),\n",
    "        \"h1\": tune.sample_from(lambda spec: 1.0 - spec.config[\"h0\"]),\n",
    "    },\n",
    "    tune_config=tune.TuneConfig(\n",
    "        num_samples=1,\n",
    "        metric=\"Q\",\n",
    "        mode=\"max\",\n",
    "    ),\n",
    "    run_config=tune.RunConfig(\n",
    "        stop={\"training_iteration\": 100},\n",
    "        failure_config=tune.FailureConfig(max_failures=3),\n",
    "    ),\n",
    ")\n",
    "\n",
    "grid_results = tuner.fit()\n",
    "if grid_results.errors:\n",
    "    raise RuntimeError\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "81f34f80-0c80-45fa-8266-341cfdad1201",
   "metadata": {},
   "source": [
    "As we can see, neither trial makes it to the optimum, since the search configs are stuck with their original values. This illustrates a key advantage of PBT: while traditional hyperparameter search methods (like grid search) keep fixed search values throughout training, PBT can adapt the search dynamically, allowing it to find better solutions with the same computational budget."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2bff9d33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1300x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 2, figsize=(13, 6), gridspec_kw=dict(width_ratios=[1.5, 1]))\n",
    "\n",
    "colors = [\"red\", \"black\"]\n",
    "labels = [\"h = [1, 0]\", \"h = [0, 1]\"]\n",
    "\n",
    "plot_parameter_history(\n",
    "    grid_results,\n",
    "    colors,\n",
    "    labels,\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    fig=fig,\n",
    "    ax=axs[0],\n",
    ")\n",
    "plot_Q_history(grid_results, colors, labels, ax=axs[1])\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "fca9689c-b4b9-4483-a8f2-43c3d9d50700",
   "metadata": {},
   "source": [
    "Compare the two plots we generated with Figure 2 from the [PBT paper](https://arxiv.org/pdf/1711.09846.pdf) (in particular, we produced the top-left and bottom-right plots).\n",
    "\n",
    "![Figure 2](figure_from_paper.png)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "49efe3ef-5fd3-429e-bf75-73fdfe674019",
   "metadata": {},
   "source": [
    "## Increase PBT population size\n",
    "\n",
    "One last experiment: what does it look like if we increase the PBT population size? Now, low-performing trials will sample one of the multiple high-performing trials to exploit, and it should result in some more interesting behavior.\n",
    "\n",
    "With a larger population:\n",
    "1. There's more diversity in the exploration space\n",
    "2. Multiple \"good\" solutions can be discovered simultaneously\n",
    "3. Different exploitation patterns emerge as trials may choose from multiple well-performing configurations\n",
    "4. The population as a whole can develop more robust strategies for optimization\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ce2daa57-fe86-4f18-9ebb-808b7b449dad",
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"tuneStatus\">\n",
       "  <div style=\"display: flex;flex-direction: row\">\n",
       "    <div style=\"display: flex;flex-direction: column;\">\n",
       "      <h3>Tune Status</h3>\n",
       "      <table>\n",
       "<tbody>\n",
       "<tr><td>Current time:</td><td>2025-02-24 16:23:40</td></tr>\n",
       "<tr><td>Running for: </td><td>00:01:18.96        </td></tr>\n",
       "<tr><td>Memory:      </td><td>21.3/36.0 GiB      </td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "    </div>\n",
       "    <div class=\"vDivider\"></div>\n",
       "    <div class=\"systemInfo\">\n",
       "      <h3>System Info</h3>\n",
       "      PopulationBasedTraining: 48 checkpoints, 48 perturbs<br>Logical resource usage: 1.0/12 CPUs, 0/0 GPUs\n",
       "    </div>\n",
       "    \n",
       "  </div>\n",
       "  <div class=\"hDivider\"></div>\n",
       "  <div class=\"trialStatus\">\n",
       "    <h3>Trial Status</h3>\n",
       "    <table>\n",
       "<thead>\n",
       "<tr><th>Trial name            </th><th>status    </th><th>loc            </th><th style=\"text-align: right;\">      h0</th><th style=\"text-align: right;\">    lr</th><th style=\"text-align: right;\">  iter</th><th style=\"text-align: right;\">  total time (s)</th><th style=\"text-align: right;\">  Q</th><th style=\"text-align: right;\">     theta0</th><th style=\"text-align: right;\">     theta1</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td>train_func_942f2_00000</td><td>TERMINATED</td><td>127.0.0.1:23974</td><td style=\"text-align: right;\">0.937925</td><td style=\"text-align: right;\">0.1008</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0464976</td><td style=\"text-align: right;\">1.2</td><td style=\"text-align: right;\">2.01666e-06</td><td style=\"text-align: right;\">3.7014e-06 </td></tr>\n",
       "<tr><td>train_func_942f2_00001</td><td>TERMINATED</td><td>127.0.0.1:23979</td><td style=\"text-align: right;\">1.18802 </td><td style=\"text-align: right;\">0.0995</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0468764</td><td style=\"text-align: right;\">1.2</td><td style=\"text-align: right;\">1.74199e-06</td><td style=\"text-align: right;\">2.48858e-06</td></tr>\n",
       "<tr><td>train_func_942f2_00002</td><td>TERMINATED</td><td>127.0.0.1:23981</td><td style=\"text-align: right;\">1.71075 </td><td style=\"text-align: right;\">0.0395</td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0464926</td><td style=\"text-align: right;\">1.2</td><td style=\"text-align: right;\">2.42464e-06</td><td style=\"text-align: right;\">4.55143e-06</td></tr>\n",
       "<tr><td>train_func_942f2_00003</td><td>TERMINATED</td><td>127.0.0.1:23982</td><td style=\"text-align: right;\">1.42562 </td><td style=\"text-align: right;\">0.084 </td><td style=\"text-align: right;\">   100</td><td style=\"text-align: right;\">       0.0461869</td><td style=\"text-align: right;\">1.2</td><td style=\"text-align: right;\">1.68403e-06</td><td style=\"text-align: right;\">3.62265e-06</td></tr>\n",
       "</tbody>\n",
       "</table>\n",
       "  </div>\n",
       "</div>\n",
       "<style>\n",
       ".tuneStatus {\n",
       "  color: var(--jp-ui-font-color1);\n",
       "}\n",
       ".tuneStatus .systemInfo {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus td {\n",
       "  white-space: nowrap;\n",
       "}\n",
       ".tuneStatus .trialStatus {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       ".tuneStatus h3 {\n",
       "  font-weight: bold;\n",
       "}\n",
       ".tuneStatus .hDivider {\n",
       "  border-bottom-width: var(--jp-border-width);\n",
       "  border-bottom-color: var(--jp-border-color0);\n",
       "  border-bottom-style: solid;\n",
       "}\n",
       ".tuneStatus .vDivider {\n",
       "  border-left-width: var(--jp-border-width);\n",
       "  border-left-color: var(--jp-border-color0);\n",
       "  border-left-style: solid;\n",
       "  margin: 0.5em 1em 0.5em 1em;\n",
       "}\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-02-24 16:22:21,301\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "2025-02-24 16:22:21,302\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "2025-02-24 16:22:21,303\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "2025-02-24 16:22:21,304\tWARNING sample.py:469 -- sample_from functions that take a spec dict are deprecated. Please update your function to work with the config dict directly.\n",
      "\u001b[36m(train_func pid=23644)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000000)\n",
      "2025-02-24 16:22:22,342\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 0.090282) into trial 942f2_00001 (score = -0.168306)\n",
      "\n",
      "2025-02-24 16:22:22,343\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.029 --- (resample) --> 0.092\n",
      "h0 : 0.0 --- (resample) --> 0.21859874791501244\n",
      "h1 : 1.0 --- (resample) --> 0.14995290392498006\n",
      "\n",
      "2025-02-24 16:22:22,343\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 0.090282) into trial 942f2_00002 (score = -0.022182)\n",
      "\n",
      "2025-02-24 16:22:22,344\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.029 --- (* 0.8) --> 0.023200000000000002\n",
      "h0 : 0.0 --- (* 0.8) --> 0.0\n",
      "h1 : 1.0 --- (* 0.8) --> 0.8\n",
      "\n",
      "2025-02-24 16:22:23,155\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/result.json\n",
      "\u001b[36m(train_func pid=23649)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000000)\n",
      "2025-02-24 16:22:23,942\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:24,739\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:25,531\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:25,539\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 0.323032) into trial 942f2_00002 (score = 0.221418)\n",
      "\n",
      "2025-02-24 16:22:25,540\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.092 --- (resample) --> 0.0385\n",
      "h0 : 0.21859874791501244 --- (* 1.2) --> 0.2623184974980149\n",
      "h1 : 0.14995290392498006 --- (* 0.8) --> 0.11996232313998406\n",
      "\n",
      "2025-02-24 16:22:25,540\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 0.323032) into trial 942f2_00003 (score = 0.239975)\n",
      "\n",
      "2025-02-24 16:22:25,541\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.092 --- (* 1.2) --> 0.1104\n",
      "h0 : 0.21859874791501244 --- (resample) --> 0.12144956368659676\n",
      "h1 : 0.14995290392498006 --- (* 1.2) --> 0.17994348470997606\n",
      "\n",
      "2025-02-24 16:22:26,332\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:27,106\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:27,882\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/result.json\n",
      "\u001b[36m(train_func pid=23670)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/checkpoint_000001)\u001b[32m [repeated 10x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:28,670\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:28,678\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 0.506889) into trial 942f2_00000 (score = 0.399434)\n",
      "\n",
      "2025-02-24 16:22:28,678\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.092 --- (* 0.8) --> 0.0736\n",
      "h0 : 0.21859874791501244 --- (resample) --> 0.8250136748029772\n",
      "h1 : 0.14995290392498006 --- (resample) --> 0.5594708426615145\n",
      "\n",
      "2025-02-24 16:22:28,679\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00003 (score = 0.505573) into trial 942f2_00002 (score = 0.406418)\n",
      "\n",
      "2025-02-24 16:22:28,679\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.1104 --- (resample) --> 0.025500000000000002\n",
      "h0 : 0.12144956368659676 --- (* 1.2) --> 0.1457394764239161\n",
      "h1 : 0.17994348470997606 --- (resample) --> 0.8083066244826129\n",
      "\n",
      "\u001b[36m(train_func pid=23671)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000001)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:29,460\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:30,255\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:31,035\tWARNING logger.py:186 -- Remote file not found: /Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/result.json\n",
      "2025-02-24 16:22:31,847\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 0.652138) into trial 942f2_00002 (score = 0.606250)\n",
      "\n",
      "2025-02-24 16:22:31,848\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.092 --- (resample) --> 0.007\n",
      "h0 : 0.21859874791501244 --- (* 0.8) --> 0.17487899833200996\n",
      "h1 : 0.14995290392498006 --- (resample) --> 0.5452206891524898\n",
      "\n",
      "2025-02-24 16:22:31,848\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 0.652138) into trial 942f2_00003 (score = 0.646607)\n",
      "\n",
      "2025-02-24 16:22:31,849\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.092 --- (* 0.8) --> 0.0736\n",
      "h0 : 0.21859874791501244 --- (resample) --> 0.007051230918609708\n",
      "h1 : 0.14995290392498006 --- (* 0.8) --> 0.11996232313998406\n",
      "\n",
      "\u001b[36m(train_func pid=23690)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000004)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23696)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000003)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:35,034\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.038110) into trial 942f2_00002 (score = 0.671646)\n",
      "\n",
      "2025-02-24 16:22:35,034\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0736 --- (resample) --> 0.018000000000000002\n",
      "h0 : 0.8250136748029772 --- (resample) --> 0.002064710166551409\n",
      "h1 : 0.5594708426615145 --- (resample) --> 0.5725196002079377\n",
      "\n",
      "2025-02-24 16:22:35,035\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 0.766900) into trial 942f2_00003 (score = 0.688034)\n",
      "\n",
      "2025-02-24 16:22:35,035\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.092 --- (* 1.2) --> 0.1104\n",
      "h0 : 0.21859874791501244 --- (resample) --> 0.6821981346240038\n",
      "h1 : 0.14995290392498006 --- (* 0.8) --> 0.11996232313998406\n",
      "\n",
      "2025-02-24 16:22:38,261\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.121589) into trial 942f2_00001 (score = 0.857585)\n",
      "\n",
      "2025-02-24 16:22:38,262\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.0736 --- (* 0.8) --> 0.05888\n",
      "h0 : 0.8250136748029772 --- (resample) --> 0.4514076493559237\n",
      "h1 : 0.5594708426615145 --- (* 0.8) --> 0.4475766741292116\n",
      "\n",
      "2025-02-24 16:22:38,262\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.050600) into trial 942f2_00003 (score = 0.947136)\n",
      "\n",
      "2025-02-24 16:22:38,263\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.018000000000000002 --- (resample) --> 0.039\n",
      "h0 : 0.002064710166551409 --- (* 0.8) --> 0.0016517681332411272\n",
      "h1 : 0.5725196002079377 --- (* 1.2) --> 0.6870235202495252\n",
      "\n",
      "\u001b[36m(train_func pid=23715)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000006)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23719)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000005)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:41,544\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.161966) into trial 942f2_00002 (score = 1.061179)\n",
      "\n",
      "2025-02-24 16:22:41,544\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0736 --- (* 0.8) --> 0.05888\n",
      "h0 : 0.8250136748029772 --- (* 0.8) --> 0.6600109398423818\n",
      "h1 : 0.5594708426615145 --- (resample) --> 0.7597397486004039\n",
      "\n",
      "2025-02-24 16:22:41,545\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.146381) into trial 942f2_00003 (score = 1.075142)\n",
      "\n",
      "2025-02-24 16:22:41,545\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.05888 --- (resample) --> 0.022\n",
      "h0 : 0.4514076493559237 --- (* 1.2) --> 0.5416891792271085\n",
      "h1 : 0.4475766741292116 --- (* 0.8) --> 0.3580613393033693\n",
      "\n",
      "2025-02-24 16:22:44,761\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.179472) into trial 942f2_00003 (score = 1.153187)\n",
      "\n",
      "2025-02-24 16:22:44,762\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.05888 --- (resample) --> 0.077\n",
      "h0 : 0.6600109398423818 --- (* 1.2) --> 0.7920131278108581\n",
      "h1 : 0.7597397486004039 --- (* 0.8) --> 0.6077917988803232\n",
      "\n",
      "2025-02-24 16:22:44,762\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.179472) into trial 942f2_00001 (score = 1.163228)\n",
      "\n",
      "2025-02-24 16:22:44,763\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.05888 --- (* 0.8) --> 0.04710400000000001\n",
      "h0 : 0.6600109398423818 --- (resample) --> 0.9912816837768351\n",
      "h1 : 0.7597397486004039 --- (resample) --> 0.14906117271353014\n",
      "\n",
      "\u001b[36m(train_func pid=23743)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/checkpoint_000002)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23748)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000007)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:47,992\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.191012) into trial 942f2_00001 (score = 1.185283)\n",
      "\n",
      "2025-02-24 16:22:47,993\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.0736 --- (resample) --> 0.017\n",
      "h0 : 0.8250136748029772 --- (* 1.2) --> 0.9900164097635725\n",
      "h1 : 0.5594708426615145 --- (resample) --> 0.8982838603244675\n",
      "\n",
      "2025-02-24 16:22:47,994\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00003 (score = 1.190555) into trial 942f2_00002 (score = 1.188719)\n",
      "\n",
      "2025-02-24 16:22:47,994\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.077 --- (resample) --> 0.008\n",
      "h0 : 0.7920131278108581 --- (resample) --> 0.6807322169820972\n",
      "h1 : 0.6077917988803232 --- (* 0.8) --> 0.4862334391042586\n",
      "\n",
      "\u001b[36m(train_func pid=23768)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/checkpoint_000008)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:51,175\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.195622) into trial 942f2_00002 (score = 1.191142)\n",
      "\n",
      "2025-02-24 16:22:51,175\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0736 --- (resample) --> 0.0205\n",
      "h0 : 0.8250136748029772 --- (* 1.2) --> 0.9900164097635725\n",
      "h1 : 0.5594708426615145 --- (resample) --> 0.6233012271154452\n",
      "\n",
      "2025-02-24 16:22:51,176\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.195622) into trial 942f2_00001 (score = 1.192855)\n",
      "\n",
      "2025-02-24 16:22:51,177\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.0736 --- (* 0.8) --> 0.05888\n",
      "h0 : 0.8250136748029772 --- (resample) --> 0.6776393680340219\n",
      "h1 : 0.5594708426615145 --- (resample) --> 0.5972686909595455\n",
      "\n",
      "\u001b[36m(train_func pid=23773)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/checkpoint_000002)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:54,409\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.197864) into trial 942f2_00002 (score = 1.196497)\n",
      "\n",
      "2025-02-24 16:22:54,410\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0736 --- (resample) --> 0.094\n",
      "h0 : 0.8250136748029772 --- (* 1.2) --> 0.9900164097635725\n",
      "h1 : 0.5594708426615145 --- (resample) --> 0.916496614878753\n",
      "\n",
      "2025-02-24 16:22:54,411\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00003 (score = 1.198000) into trial 942f2_00001 (score = 1.197464)\n",
      "\n",
      "2025-02-24 16:22:54,411\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.077 --- (resample) --> 0.009000000000000001\n",
      "h0 : 0.7920131278108581 --- (resample) --> 0.09724457530695019\n",
      "h1 : 0.6077917988803232 --- (* 0.8) --> 0.4862334391042586\n",
      "\n",
      "\u001b[36m(train_func pid=23796)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000011)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23801)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000010)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:22:57,678\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199463) into trial 942f2_00001 (score = 1.198073)\n",
      "\n",
      "2025-02-24 16:22:57,678\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.094 --- (resample) --> 0.011\n",
      "h0 : 0.9900164097635725 --- (* 1.2) --> 1.188019691716287\n",
      "h1 : 0.916496614878753 --- (resample) --> 0.854735155913485\n",
      "\n",
      "2025-02-24 16:22:57,679\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00003 (score = 1.199079) into trial 942f2_00000 (score = 1.198957)\n",
      "\n",
      "2025-02-24 16:22:57,679\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.077 --- (* 1.2) --> 0.0924\n",
      "h0 : 0.7920131278108581 --- (resample) --> 0.8783500284482123\n",
      "h1 : 0.6077917988803232 --- (* 1.2) --> 0.7293501586563879\n",
      "\n",
      "2025-02-24 16:23:00,836\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199862) into trial 942f2_00001 (score = 1.199540)\n",
      "\n",
      "2025-02-24 16:23:00,836\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.094 --- (* 0.8) --> 0.0752\n",
      "h0 : 0.9900164097635725 --- (resample) --> 0.06185563216172696\n",
      "h1 : 0.916496614878753 --- (resample) --> 0.06868522206070948\n",
      "\n",
      "2025-02-24 16:23:00,837\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199862) into trial 942f2_00003 (score = 1.199576)\n",
      "\n",
      "2025-02-24 16:23:00,837\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.094 --- (* 1.2) --> 0.1128\n",
      "h0 : 0.9900164097635725 --- (resample) --> 0.3672068732350573\n",
      "h1 : 0.916496614878753 --- (resample) --> 0.3263725487154706\n",
      "\n",
      "\u001b[36m(train_func pid=23821)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000013)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23822)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/checkpoint_000011)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:04,072\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199964) into trial 942f2_00001 (score = 1.199871)\n",
      "\n",
      "2025-02-24 16:23:04,073\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.094 --- (* 0.8) --> 0.0752\n",
      "h0 : 0.9900164097635725 --- (resample) --> 0.8143417145384867\n",
      "h1 : 0.916496614878753 --- (* 1.2) --> 1.0997959378545035\n",
      "\n",
      "2025-02-24 16:23:04,073\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199964) into trial 942f2_00000 (score = 1.199896)\n",
      "\n",
      "2025-02-24 16:23:04,074\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.094 --- (* 0.8) --> 0.0752\n",
      "h0 : 0.9900164097635725 --- (resample) --> 0.28845453300169044\n",
      "h1 : 0.916496614878753 --- (resample) --> 0.02235127072371279\n",
      "\n",
      "2025-02-24 16:23:07,516\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.199986) into trial 942f2_00003 (score = 1.199955)\n",
      "\n",
      "2025-02-24 16:23:07,516\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.0752 --- (* 0.8) --> 0.060160000000000005\n",
      "h0 : 0.8143417145384867 --- (* 1.2) --> 0.9772100574461839\n",
      "h1 : 1.0997959378545035 --- (* 0.8) --> 0.8798367502836029\n",
      "\n",
      "2025-02-24 16:23:07,517\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.199986) into trial 942f2_00000 (score = 1.199969)\n",
      "\n",
      "2025-02-24 16:23:07,517\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.0752 --- (resample) --> 0.0155\n",
      "h0 : 0.8143417145384867 --- (* 1.2) --> 0.9772100574461839\n",
      "h1 : 1.0997959378545035 --- (* 0.8) --> 0.8798367502836029\n",
      "\n",
      "\u001b[36m(train_func pid=23846)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/checkpoint_000007)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23846)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/checkpoint_000006)\u001b[32m [repeated 6x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:10,721\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.199994) into trial 942f2_00000 (score = 1.199989)\n",
      "\n",
      "2025-02-24 16:23:10,722\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.0752 --- (resample) --> 0.005\n",
      "h0 : 0.8143417145384867 --- (resample) --> 0.14093804696635504\n",
      "h1 : 1.0997959378545035 --- (resample) --> 0.04714342092680601\n",
      "\n",
      "2025-02-24 16:23:10,723\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199997) into trial 942f2_00003 (score = 1.199994)\n",
      "\n",
      "2025-02-24 16:23:10,723\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.094 --- (* 0.8) --> 0.0752\n",
      "h0 : 0.9900164097635725 --- (resample) --> 0.4368194817950344\n",
      "h1 : 0.916496614878753 --- (resample) --> 0.7095403843032826\n",
      "\n",
      "\u001b[36m(train_func pid=23867)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/checkpoint_000015)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23867)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/checkpoint_000014)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:13,989\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199999) into trial 942f2_00000 (score = 1.199994)\n",
      "\n",
      "2025-02-24 16:23:13,989\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.094 --- (resample) --> 0.0925\n",
      "h0 : 0.9900164097635725 --- (resample) --> 0.998683166515384\n",
      "h1 : 0.916496614878753 --- (* 1.2) --> 1.0997959378545035\n",
      "\n",
      "2025-02-24 16:23:13,990\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.199999) into trial 942f2_00001 (score = 1.199998)\n",
      "\n",
      "2025-02-24 16:23:13,990\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00001:\n",
      "lr : 0.094 --- (resample) --> 0.0995\n",
      "h0 : 0.9900164097635725 --- (* 1.2) --> 1.188019691716287\n",
      "h1 : 0.916496614878753 --- (* 0.8) --> 0.7331972919030024\n",
      "\n",
      "2025-02-24 16:23:17,224\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.200000) into trial 942f2_00003 (score = 1.199999)\n",
      "\n",
      "2025-02-24 16:23:17,224\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.0925 --- (resample) --> 0.006500000000000001\n",
      "h0 : 0.998683166515384 --- (* 0.8) --> 0.7989465332123072\n",
      "h1 : 1.0997959378545035 --- (* 0.8) --> 0.8798367502836029\n",
      "\n",
      "2025-02-24 16:23:17,225\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.200000) into trial 942f2_00002 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:17,225\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0995 --- (* 0.8) --> 0.0796\n",
      "h0 : 1.188019691716287 --- (* 0.8) --> 0.9504157533730297\n",
      "h1 : 0.7331972919030024 --- (* 0.8) --> 0.586557833522402\n",
      "\n",
      "\u001b[36m(train_func pid=23892)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000018)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23892)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000017)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:20,513\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.200000) into trial 942f2_00003 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:20,514\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.0995 --- (resample) --> 0.0325\n",
      "h0 : 1.188019691716287 --- (* 0.8) --> 0.9504157533730297\n",
      "h1 : 0.7331972919030024 --- (resample) --> 0.19444236619090172\n",
      "\n",
      "2025-02-24 16:23:20,515\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.200000) into trial 942f2_00002 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:20,515\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0925 --- (* 0.8) --> 0.074\n",
      "h0 : 0.998683166515384 --- (* 1.2) --> 1.1984197998184607\n",
      "h1 : 1.0997959378545035 --- (resample) --> 0.6632564869583678\n",
      "\n",
      "2025-02-24 16:23:23,779\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00000 (score = 1.200000) into trial 942f2_00003 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:23,779\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.0925 --- (resample) --> 0.0205\n",
      "h0 : 0.998683166515384 --- (* 0.8) --> 0.7989465332123072\n",
      "h1 : 1.0997959378545035 --- (* 1.2) --> 1.319755125425404\n",
      "\n",
      "2025-02-24 16:23:23,780\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.200000) into trial 942f2_00002 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:23,780\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0995 --- (resample) --> 0.059500000000000004\n",
      "h0 : 1.188019691716287 --- (* 1.2) --> 1.4256236300595444\n",
      "h1 : 0.7331972919030024 --- (resample) --> 0.19309431415014977\n",
      "\n",
      "\u001b[36m(train_func pid=23917)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000020)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23917)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00000_0_h0=0.0000,lr=0.0290_2025-02-24_16-22-21/checkpoint_000019)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:27,089\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.200000) into trial 942f2_00003 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:27,090\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.059500000000000004 --- (* 0.8) --> 0.0476\n",
      "h0 : 1.4256236300595444 --- (* 0.8) --> 1.1404989040476357\n",
      "h1 : 0.19309431415014977 --- (* 0.8) --> 0.15447545132011983\n",
      "\n",
      "2025-02-24 16:23:27,090\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.200000) into trial 942f2_00000 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:27,091\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.059500000000000004 --- (resample) --> 0.051000000000000004\n",
      "h0 : 1.4256236300595444 --- (resample) --> 0.5322491694545954\n",
      "h1 : 0.19309431415014977 --- (resample) --> 0.4907896898235511\n",
      "\n",
      "2025-02-24 16:23:30,403\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.200000) into trial 942f2_00003 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:30,403\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00003:\n",
      "lr : 0.0995 --- (resample) --> 0.084\n",
      "h0 : 1.188019691716287 --- (* 1.2) --> 1.4256236300595444\n",
      "h1 : 0.7331972919030024 --- (resample) --> 0.7068936194953941\n",
      "\n",
      "2025-02-24 16:23:30,404\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00002 (score = 1.200000) into trial 942f2_00000 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:30,404\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.059500000000000004 --- (resample) --> 0.041\n",
      "h0 : 1.4256236300595444 --- (* 1.2) --> 1.7107483560714531\n",
      "h1 : 0.19309431415014977 --- (resample) --> 0.6301738678453057\n",
      "\n",
      "\u001b[36m(train_func pid=23942)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00003_3_h0=0.9900,lr=0.0530_2025-02-24_16-22-21/checkpoint_000008)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23942)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00002_2_h0=0.0100,lr=0.0170_2025-02-24_16-22-21/checkpoint_000019)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:33,643\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.200000) into trial 942f2_00002 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:33,643\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.0995 --- (resample) --> 0.08\n",
      "h0 : 1.188019691716287 --- (* 1.2) --> 1.4256236300595444\n",
      "h1 : 0.7331972919030024 --- (resample) --> 0.12615387675586676\n",
      "\n",
      "2025-02-24 16:23:33,644\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00001 (score = 1.200000) into trial 942f2_00000 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:33,644\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.0995 --- (resample) --> 0.0185\n",
      "h0 : 1.188019691716287 --- (* 1.2) --> 1.4256236300595444\n",
      "h1 : 0.7331972919030024 --- (* 0.8) --> 0.586557833522402\n",
      "\n",
      "\u001b[36m(train_func pid=23962)\u001b[0m Checkpoint successfully created at: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000023)\u001b[32m [repeated 6x across cluster]\u001b[0m\n",
      "\u001b[36m(train_func pid=23967)\u001b[0m Restored on 127.0.0.1 from checkpoint: Checkpoint(filesystem=local, path=/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21/train_func_942f2_00001_1_h0=1.0000,lr=0.0070_2025-02-24_16-22-21/checkpoint_000022)\u001b[32m [repeated 7x across cluster]\u001b[0m\n",
      "2025-02-24 16:23:36,961\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00003 (score = 1.200000) into trial 942f2_00000 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:36,961\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00000:\n",
      "lr : 0.084 --- (* 1.2) --> 0.1008\n",
      "h0 : 1.4256236300595444 --- (resample) --> 0.9379248877817841\n",
      "h1 : 0.7068936194953941 --- (* 0.8) --> 0.5655148955963153\n",
      "\n",
      "2025-02-24 16:23:36,962\tINFO pbt.py:878 -- \n",
      "\n",
      "[PopulationBasedTraining] [Exploit] Cloning trial 942f2_00003 (score = 1.200000) into trial 942f2_00002 (score = 1.200000)\n",
      "\n",
      "2025-02-24 16:23:36,962\tINFO pbt.py:905 -- \n",
      "\n",
      "[PopulationBasedTraining] [Explore] Perturbed the hyperparameter config of trial942f2_00002:\n",
      "lr : 0.084 --- (resample) --> 0.0395\n",
      "h0 : 1.4256236300595444 --- (* 1.2) --> 1.7107483560714531\n",
      "h1 : 0.7068936194953941 --- (* 1.2) --> 0.8482723433944729\n",
      "\n",
      "2025-02-24 16:23:40,264\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/Users/rdecal/ray_results/train_func_2025-02-24_16-22-21' in 0.0086s.\n",
      "2025-02-24 16:23:40,265\tINFO tune.py:1041 -- Total run time: 78.97 seconds (78.95 seconds for the tuning loop).\n"
     ]
    }
   ],
   "source": [
    "if ray.is_initialized():\n",
    "    ray.shutdown()\n",
    "ray.init()\n",
    "perturbation_interval = 4\n",
    "pbt_scheduler = PopulationBasedTraining(\n",
    "    time_attr=\"training_iteration\",\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    quantile_fraction=0.5,\n",
    "    resample_probability=0.5,\n",
    "    hyperparam_mutations={\n",
    "        \"lr\": tune.qloguniform(5e-3, 1e-1, 5e-4),\n",
    "        \"h0\": tune.uniform(0.0, 1.0),\n",
    "        \"h1\": tune.uniform(0.0, 1.0),\n",
    "    },\n",
    "    synch=True,\n",
    ")\n",
    "tuner = Tuner(\n",
    "    train_func,\n",
    "    param_space={\n",
    "        \"lr\": tune.qloguniform(5e-3, 1e-1, 5e-4),\n",
    "        \"h0\": tune.grid_search([0.0, 1.0, 0.01, 0.99]),  # 4 trials\n",
    "        \"h1\": tune.sample_from(lambda spec: 1.0 - spec.config[\"h0\"]),\n",
    "        \"num_training_iterations\": 100,\n",
    "        \"checkpoint_interval\": perturbation_interval,\n",
    "    },\n",
    "    tune_config=TuneConfig(\n",
    "        num_samples=1,\n",
    "        metric=\"Q\",\n",
    "        mode=\"max\",\n",
    "        # Set the PBT scheduler in this config\n",
    "        scheduler=pbt_scheduler,\n",
    "    ),\n",
    "    run_config=tune.RunConfig(\n",
    "        stop={\"training_iteration\": 100},\n",
    "        failure_config=tune.FailureConfig(max_failures=3),\n",
    "    ),\n",
    ")\n",
    "pbt_4_results = tuner.fit()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a1188de3-57c5-45a2-a1a3-0f5fbb5607b0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1300x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 2, figsize=(13, 6), gridspec_kw=dict(width_ratios=[1.5, 1]))\n",
    "\n",
    "colors = [\"red\", \"black\", \"blue\", \"green\"]\n",
    "labels = [\"h = [1, 0]\", \"h = [0, 1]\", \"h = [0.01, 0.99]\", \"h = [0.99, 0.01]\"]\n",
    "\n",
    "plot_parameter_history(\n",
    "    pbt_4_results,\n",
    "    colors,\n",
    "    labels,\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    fig=fig,\n",
    "    ax=axs[0],\n",
    ")\n",
    "plot_Q_history(pbt_4_results, colors, labels, ax=axs[1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "577c47a7-b68a-412b-8902-1c46e73340a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "make_animation(\n",
    "    pbt_4_results,\n",
    "    colors,\n",
    "    labels,\n",
    "    perturbation_interval=perturbation_interval,\n",
    "    filename=\"pbt4.gif\",\n",
    ")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "33aebed2-5d9c-45be-a2c1-b0529fae2762",
   "metadata": {},
   "source": [
    "![PBT 4 Trial Visualization](pbt4.gif)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "24f15e34-485d-4f27-96e2-8fba7402a376",
   "metadata": {},
   "source": [
    "## Summary\n",
    "\n",
    "Hopefully, this guide has given you a better understanding of the PBT algorithm. Please file any issues you run into when running this notebook and ask any questions you might have in the Ray Slack"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}