examples/plots_and_comparisons.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# hposuite Plotting and Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading and printing available Benchmarks and Optimizers in hposuite"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['ackley', 'branin', 'mfh3_bad', 'mfh6_bad', 'mfh3_good', 'mfh6_good', 'mfh3_moderate', 'mfh6_moderate', 'mfh3_terrible', 'mfh6_terrible', 'pd1-cifar100-wide_resnet-2048', 'pd1-imagenet-resnet-512', 'pd1-lm1b-transformer-2048', 'pd1-translate_wmt-xformer_translate-64', 'lcbench_tabular-adult', 'lcbench_tabular-airlines', 'lcbench_tabular-albert', 'lcbench_tabular-Amazon_employee_access', 'lcbench_tabular-APSFailure', 'lcbench_tabular-Australian', 'lcbench_tabular-bank-marketing', 'lcbench_tabular-blood-transfusion-service-center', 'lcbench_tabular-car', 'lcbench_tabular-christine', 'lcbench_tabular-cnae-9', 'lcbench_tabular-connect-4', 'lcbench_tabular-covertype', 'lcbench_tabular-credit-g', 'lcbench_tabular-dionis', 'lcbench_tabular-fabert', 'lcbench_tabular-Fashion-MNIST', 'lcbench_tabular-helena', 'lcbench_tabular-higgs', 'lcbench_tabular-jannis', 'lcbench_tabular-jasmine', 'lcbench_tabular-jungle_chess_2pcs_raw_endgame_complete', 'lcbench_tabular-kc1', 'lcbench_tabular-KDDCup09_appetency', 'lcbench_tabular-kr-vs-kp', 'lcbench_tabular-mfeat-factors', 'lcbench_tabular-MiniBooNE', 'lcbench_tabular-nomao', 'lcbench_tabular-numerai28.6', 'lcbench_tabular-phoneme', 'lcbench_tabular-segment', 'lcbench_tabular-shuttle', 'lcbench_tabular-sylvine', 'lcbench_tabular-vehicle', 'lcbench_tabular-volkert', 'pymoo-ackley', 'pymoo-griewank', 'pymoo-himmelblau', 'pymoo-rastrigin', 'pymoo-rosenbrock', 'pymoo-schwefel', 'pymoo-sphere', 'pymoo-zakharov', 'pymoo-kursawe', 'pymoo-zdt1', 'pymoo-zdt2', 'pymoo-zdt3', 'pymoo-zdt4', 'pymoo-zdt5', 'pymoo-zdt6', 'pymoo-omnitest', 'pymoo-sympart', 'pymoo-sympart_rotated', 'pymoo-dtlz1', 'pymoo-dtlz2', 'pymoo-dtlz3', 'pymoo-dtlz4', 'pymoo-dtlz5', 'pymoo-dtlz6', 'pymoo-dtlz7', 'bbob-f1-2-0', 'bbob-f1-2-1', 'bbob-f1-2-2', 'bbob-f1-3-0', 'bbob-f1-3-1', 'bbob-f1-3-2', 'bbob-f1-5-0', 'bbob-f1-5-1', 'bbob-f1-5-2', 'bbob-f1-10-0', 'bbob-f1-10-1', 'bbob-f1-10-2', 'bbob-f1-20-0', 'bbob-f1-20-1', 'bbob-f1-20-2', 'bbob-f1-40-0', 'bbob-f1-40-1', 'bbob-f1-40-2', 'bbob-f2-2-0', 'bbob-f2-2-1', 'bbob-f2-2-2', 'bbob-f2-3-0', 'bbob-f2-3-1', 'bbob-f2-3-2', 'bbob-f2-5-0', 'bbob-f2-5-1', 'bbob-f2-5-2', 'bbob-f2-10-0', 'bbob-f2-10-1', 'bbob-f2-10-2', 'bbob-f2-20-0', 'bbob-f2-20-1', 'bbob-f2-20-2', 'bbob-f2-40-0', 'bbob-f2-40-1', 'bbob-f2-40-2', 'bbob-f3-2-0', 'bbob-f3-2-1', 'bbob-f3-2-2', 'bbob-f3-3-0', 'bbob-f3-3-1', 'bbob-f3-3-2', 'bbob-f3-5-0', 'bbob-f3-5-1', 'bbob-f3-5-2', 'bbob-f3-10-0', 'bbob-f3-10-1', 'bbob-f3-10-2', 'bbob-f3-20-0', 'bbob-f3-20-1', 'bbob-f3-20-2', 'bbob-f3-40-0', 'bbob-f3-40-1', 'bbob-f3-40-2', 'bbob-f4-2-0', 'bbob-f4-2-1', 'bbob-f4-2-2', 'bbob-f4-3-0', 'bbob-f4-3-1', 'bbob-f4-3-2', 'bbob-f4-5-0', 'bbob-f4-5-1', 'bbob-f4-5-2', 'bbob-f4-10-0', 'bbob-f4-10-1', 'bbob-f4-10-2', 'bbob-f4-20-0', 'bbob-f4-20-1', 'bbob-f4-20-2', 'bbob-f4-40-0', 'bbob-f4-40-1', 'bbob-f4-40-2', 'bbob-f5-2-0', 'bbob-f5-2-1', 'bbob-f5-2-2', 'bbob-f5-3-0', 'bbob-f5-3-1', 'bbob-f5-3-2', 'bbob-f5-5-0', 'bbob-f5-5-1', 'bbob-f5-5-2', 'bbob-f5-10-0', 'bbob-f5-10-1', 'bbob-f5-10-2', 'bbob-f5-20-0', 'bbob-f5-20-1', 'bbob-f5-20-2', 'bbob-f5-40-0', 'bbob-f5-40-1', 'bbob-f5-40-2', 'bbob-f6-2-0', 'bbob-f6-2-1', 'bbob-f6-2-2', 'bbob-f6-3-0', 'bbob-f6-3-1', 'bbob-f6-3-2', 'bbob-f6-5-0', 'bbob-f6-5-1', 'bbob-f6-5-2', 'bbob-f6-10-0', 'bbob-f6-10-1', 'bbob-f6-10-2', 'bbob-f6-20-0', 'bbob-f6-20-1', 'bbob-f6-20-2', 'bbob-f6-40-0', 'bbob-f6-40-1', 'bbob-f6-40-2', 'bbob-f7-2-0', 'bbob-f7-2-1', 'bbob-f7-2-2', 'bbob-f7-3-0', 'bbob-f7-3-1', 'bbob-f7-3-2', 'bbob-f7-5-0', 'bbob-f7-5-1', 'bbob-f7-5-2', 'bbob-f7-10-0', 'bbob-f7-10-1', 'bbob-f7-10-2', 'bbob-f7-20-0', 'bbob-f7-20-1', 'bbob-f7-20-2', 'bbob-f7-40-0', 'bbob-f7-40-1', 'bbob-f7-40-2', 'bbob-f8-2-0', 'bbob-f8-2-1', 'bbob-f8-2-2', 'bbob-f8-3-0', 'bbob-f8-3-1', 'bbob-f8-3-2', 'bbob-f8-5-0', 'bbob-f8-5-1', 'bbob-f8-5-2', 'bbob-f8-10-0', 'bbob-f8-10-1', 'bbob-f8-10-2', 'bbob-f8-20-0', 'bbob-f8-20-1', 'bbob-f8-20-2', 'bbob-f8-40-0', 'bbob-f8-40-1', 'bbob-f8-40-2', 'bbob-f9-2-0', 'bbob-f9-2-1', 'bbob-f9-2-2', 'bbob-f9-3-0', 'bbob-f9-3-1', 'bbob-f9-3-2', 'bbob-f9-5-0', 'bbob-f9-5-1', 'bbob-f9-5-2', 'bbob-f9-10-0', 'bbob-f9-10-1', 'bbob-f9-10-2', 'bbob-f9-20-0', 'bbob-f9-20-1', 'bbob-f9-20-2', 'bbob-f9-40-0', 'bbob-f9-40-1', 'bbob-f9-40-2', 'bbob-f10-2-0', 'bbob-f10-2-1', 'bbob-f10-2-2', 'bbob-f10-3-0', 'bbob-f10-3-1', 'bbob-f10-3-2', 'bbob-f10-5-0', 'bbob-f10-5-1', 'bbob-f10-5-2', 'bbob-f10-10-0', 'bbob-f10-10-1', 'bbob-f10-10-2', 'bbob-f10-20-0', 'bbob-f10-20-1', 'bbob-f10-20-2', 'bbob-f10-40-0', 'bbob-f10-40-1', 'bbob-f10-40-2', 'bbob-f11-2-0', 'bbob-f11-2-1', 'bbob-f11-2-2', 'bbob-f11-3-0', 'bbob-f11-3-1', 'bbob-f11-3-2', 'bbob-f11-5-0', 'bbob-f11-5-1', 'bbob-f11-5-2', 'bbob-f11-10-0', 'bbob-f11-10-1', 'bbob-f11-10-2', 'bbob-f11-20-0', 'bbob-f11-20-1', 'bbob-f11-20-2', 'bbob-f11-40-0', 'bbob-f11-40-1', 'bbob-f11-40-2', 'bbob-f12-2-0', 'bbob-f12-2-1', 'bbob-f12-2-2', 'bbob-f12-3-0', 'bbob-f12-3-1', 'bbob-f12-3-2', 'bbob-f12-5-0', 'bbob-f12-5-1', 'bbob-f12-5-2', 'bbob-f12-10-0', 'bbob-f12-10-1', 'bbob-f12-10-2', 'bbob-f12-20-0', 'bbob-f12-20-1', 'bbob-f12-20-2', 'bbob-f12-40-0', 'bbob-f12-40-1', 'bbob-f12-40-2', 'bbob-f13-2-0', 'bbob-f13-2-1', 'bbob-f13-2-2', 'bbob-f13-3-0', 'bbob-f13-3-1', 'bbob-f13-3-2', 'bbob-f13-5-0', 'bbob-f13-5-1', 'bbob-f13-5-2', 'bbob-f13-10-0', 'bbob-f13-10-1', 'bbob-f13-10-2', 'bbob-f13-20-0', 'bbob-f13-20-1', 'bbob-f13-20-2', 'bbob-f13-40-0', 'bbob-f13-40-1', 'bbob-f13-40-2', 'bbob-f14-2-0', 'bbob-f14-2-1', 'bbob-f14-2-2', 'bbob-f14-3-0', 'bbob-f14-3-1', 'bbob-f14-3-2', 'bbob-f14-5-0', 'bbob-f14-5-1', 'bbob-f14-5-2', 'bbob-f14-10-0', 'bbob-f14-10-1', 'bbob-f14-10-2', 'bbob-f14-20-0', 'bbob-f14-20-1', 'bbob-f14-20-2', 'bbob-f14-40-0', 'bbob-f14-40-1', 'bbob-f14-40-2', 'bbob-f15-2-0', 'bbob-f15-2-1', 'bbob-f15-2-2', 'bbob-f15-3-0', 'bbob-f15-3-1', 'bbob-f15-3-2', 'bbob-f15-5-0', 'bbob-f15-5-1', 'bbob-f15-5-2', 'bbob-f15-10-0', 'bbob-f15-10-1', 'bbob-f15-10-2', 'bbob-f15-20-0', 'bbob-f15-20-1', 'bbob-f15-20-2', 'bbob-f15-40-0', 'bbob-f15-40-1', 'bbob-f15-40-2', 'bbob-f16-2-0', 'bbob-f16-2-1', 'bbob-f16-2-2', 'bbob-f16-3-0', 'bbob-f16-3-1', 'bbob-f16-3-2', 'bbob-f16-5-0', 'bbob-f16-5-1', 'bbob-f16-5-2', 'bbob-f16-10-0', 'bbob-f16-10-1', 'bbob-f16-10-2', 'bbob-f16-20-0', 'bbob-f16-20-1', 'bbob-f16-20-2', 'bbob-f16-40-0', 'bbob-f16-40-1', 'bbob-f16-40-2', 'bbob-f17-2-0', 'bbob-f17-2-1', 'bbob-f17-2-2', 'bbob-f17-3-0', 'bbob-f17-3-1', 'bbob-f17-3-2', 'bbob-f17-5-0', 'bbob-f17-5-1', 'bbob-f17-5-2', 'bbob-f17-10-0', 'bbob-f17-10-1', 'bbob-f17-10-2', 'bbob-f17-20-0', 'bbob-f17-20-1', 'bbob-f17-20-2', 'bbob-f17-40-0', 'bbob-f17-40-1', 'bbob-f17-40-2', 'bbob-f18-2-0', 'bbob-f18-2-1', 'bbob-f18-2-2', 'bbob-f18-3-0', 'bbob-f18-3-1', 'bbob-f18-3-2', 'bbob-f18-5-0', 'bbob-f18-5-1', 'bbob-f18-5-2', 'bbob-f18-10-0', 'bbob-f18-10-1', 'bbob-f18-10-2', 'bbob-f18-20-0', 'bbob-f18-20-1', 'bbob-f18-20-2', 'bbob-f18-40-0', 'bbob-f18-40-1', 'bbob-f18-40-2', 'bbob-f19-2-0', 'bbob-f19-2-1', 'bbob-f19-2-2', 'bbob-f19-3-0', 'bbob-f19-3-1', 'bbob-f19-3-2', 'bbob-f19-5-0', 'bbob-f19-5-1', 'bbob-f19-5-2', 'bbob-f19-10-0', 'bbob-f19-10-1', 'bbob-f19-10-2', 'bbob-f19-20-0', 'bbob-f19-20-1', 'bbob-f19-20-2', 'bbob-f19-40-0', 'bbob-f19-40-1', 'bbob-f19-40-2', 'bbob-f20-2-0', 'bbob-f20-2-1', 'bbob-f20-2-2', 'bbob-f20-3-0', 'bbob-f20-3-1', 'bbob-f20-3-2', 'bbob-f20-5-0', 'bbob-f20-5-1', 'bbob-f20-5-2', 'bbob-f20-10-0', 'bbob-f20-10-1', 'bbob-f20-10-2', 'bbob-f20-20-0', 'bbob-f20-20-1', 'bbob-f20-20-2', 'bbob-f20-40-0', 'bbob-f20-40-1', 'bbob-f20-40-2', 'bbob-f21-2-0', 'bbob-f21-2-1', 'bbob-f21-2-2', 'bbob-f21-3-0', 'bbob-f21-3-1', 'bbob-f21-3-2', 'bbob-f21-5-0', 'bbob-f21-5-1', 'bbob-f21-5-2', 'bbob-f21-10-0', 'bbob-f21-10-1', 'bbob-f21-10-2', 'bbob-f21-20-0', 'bbob-f21-20-1', 'bbob-f21-20-2', 'bbob-f21-40-0', 'bbob-f21-40-1', 'bbob-f21-40-2', 'bbob-f22-2-0', 'bbob-f22-2-1', 'bbob-f22-2-2', 'bbob-f22-3-0', 'bbob-f22-3-1', 'bbob-f22-3-2', 'bbob-f22-5-0', 'bbob-f22-5-1', 'bbob-f22-5-2', 'bbob-f22-10-0', 'bbob-f22-10-1', 'bbob-f22-10-2', 'bbob-f22-20-0', 'bbob-f22-20-1', 'bbob-f22-20-2', 'bbob-f22-40-0', 'bbob-f22-40-1', 'bbob-f22-40-2', 'bbob-f23-2-0', 'bbob-f23-2-1', 'bbob-f23-2-2', 'bbob-f23-3-0', 'bbob-f23-3-1', 'bbob-f23-3-2', 'bbob-f23-5-0', 'bbob-f23-5-1', 'bbob-f23-5-2', 'bbob-f23-10-0', 'bbob-f23-10-1', 'bbob-f23-10-2', 'bbob-f23-20-0', 'bbob-f23-20-1', 'bbob-f23-20-2', 'bbob-f23-40-0', 'bbob-f23-40-1', 'bbob-f23-40-2', 'bbob-f24-2-0', 'bbob-f24-2-1', 'bbob-f24-2-2', 'bbob-f24-3-0', 'bbob-f24-3-1', 'bbob-f24-3-2', 'bbob-f24-5-0', 'bbob-f24-5-1', 'bbob-f24-5-2', 'bbob-f24-10-0', 'bbob-f24-10-1', 'bbob-f24-10-2', 'bbob-f24-20-0', 'bbob-f24-20-1', 'bbob-f24-20-2', 'bbob-f24-40-0', 'bbob-f24-40-1', 'bbob-f24-40-2'])\n",
      "dict_keys(['DEHB', 'HEBO', 'Nevergrad', 'Optuna', 'RandomSearch', 'RandomSearchWithPriors', 'Scikit_Optimize', 'SMAC_BO', 'SMAC_Hyperband'])\n"
     ]
    }
   ],
   "source": [
    "from hposuite.benchmarks import BENCHMARKS\n",
    "from hposuite.optimizers import OPTIMIZERS\n",
    "\n",
    "print(BENCHMARKS.keys())\n",
    "print(OPTIMIZERS.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparing RandomSearch and RandomSearchWithPriors on Ackley Functional Benchmark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:hposuite.study:Generated 10 runs\n",
      "INFO:hposuite.study:Created study at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp\n",
      "INFO:hposuite.study:Dumped experiments to /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/dump_all.txt\n",
      "INFO:hposuite.study:Overwrite flag is set to True. Existing results will be overwritten!\n",
      "INFO:hposuite.study:Running 10 experiments sequentially\n",
      "INFO:hposuite.study:Running experiment 1/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=383329928 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=383329928.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearch'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=383329928\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=383329928/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=383329928.parquet\n",
      "INFO:hposuite.study:Running experiment 2/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=3324115917 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=3324115917.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearch'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=3324115917\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=3324115917/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=3324115917.parquet\n",
      "INFO:hposuite.study:Running experiment 3/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=2811363265 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=2811363265.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearch'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=2811363265\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=2811363265/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=2811363265.parquet\n",
      "INFO:hposuite.study:Running experiment 4/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1884968545 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1884968545.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearch'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1884968545\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1884968545/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1884968545.parquet\n",
      "INFO:hposuite.study:Running experiment 5/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1859786276 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1859786276.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearch'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1859786276\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1859786276/optimizer=RandomSearch.benchmark=ackley.objectives=y.TrialBudget=1000.seed=1859786276.parquet\n",
      "INFO:hposuite.study:Running experiment 6/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=383329928 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=383329928.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearchWithPriors'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=383329928\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=383329928/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=383329928.parquet\n",
      "INFO:hposuite.study:Running experiment 7/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=3324115917 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=3324115917.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearchWithPriors'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=3324115917\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=3324115917/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=3324115917.parquet\n",
      "INFO:hposuite.study:Running experiment 8/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=2811363265 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=2811363265.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearchWithPriors'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=2811363265\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=2811363265/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=2811363265.parquet\n",
      "INFO:hposuite.study:Running experiment 9/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1884968545 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1884968545.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearchWithPriors'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1884968545\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1884968545/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1884968545.parquet\n",
      "INFO:hposuite.study:Running experiment 10/10\n",
      "INFO:hposuite.run:Overwriting optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1859786276 in `state=<State.COMPLETE: 'COMPLETE'>` at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1859786276.\n",
      "WARNING:hposuite.run:Continuations are not supported for the optimizer: <class 'hposuite.optimizers.random_search.RandomSearchWithPriors'>. Ignoring continuations=True.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1859786276\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1859786276/optimizer=RandomSearchWithPriors.benchmark=ackley.objectives=y.TrialBudget=1000.priors=y_optimum.seed=1859786276.parquet\n",
      "INFO:hposuite.study:Completed study with 10 runs\n"
     ]
    }
   ],
   "source": [
    "from hposuite import create_study\n",
    "\n",
    "ackley = BENCHMARKS[\"ackley\"]\n",
    "\n",
    "study = create_study(\n",
    "    name=\"ex-rs-comp\",\n",
    "    output_dir=\"example-outputs\",\n",
    "    optimizers=[\n",
    "        \"RandomSearch\",\n",
    "        \"RandomSearchWithPriors\",\n",
    "    ],\n",
    "    benchmarks=(\n",
    "        \"ackley\",\n",
    "        {\n",
    "            \"priors\": (\n",
    "                \"y_optimum\",\n",
    "                {\n",
    "                    \"y\": ackley.desc.predefined_points[\"min\"]\n",
    "                }\n",
    "            )\n",
    "        }\n",
    "    ),\n",
    "    num_seeds=5,\n",
    "    budget=1000,\n",
    ")\n",
    "\n",
    "study.optimize(overwrite=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plotting the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Some stuff might fail: issue in joblib\n",
      "INFO:__main__:Found benchmarks: ['ackley']\n",
      "INFO:__main__:Processing benchmark: ackley\n",
      "INFO:__main__:Plotting RandomSearchWithPriors_default\n",
      "INFO:__main__:Plotting RandomSearch_default\n",
      "INFO:__main__:Saved plot to /home/soham/repos/hposuite/examples/example-outputs/ex-rs-comp/plots/RandomSearchWithPriors_default,RandomSearch_default.ackley, objective='y', fidelity=None, cost=None, TrialBudget=1000.0, to_minimize=True, error_bars='std'.png\n"
     ]
    }
   ],
   "source": [
    "!python -m hposuite.plotting.incumbent_trace --study_dir ex-rs-comp --output_dir example-outputs --figsize 10 7 --logscale"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Displaying the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "\n",
    "Image(\n",
    "    filename=\"example-outputs/ex-rs-comp/plots/\"\n",
    "        \"RandomSearchWithPriors_default,RandomSearch_default.\"\n",
    "        \"ackley, objective='y', fidelity=None, cost=None, TrialBudget=1000.0, \"\n",
    "        \"to_minimize=True, error_bars='std'.png\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparing the effect of Continuations using a Multi-fidelity Optimizer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this example we use the DEHB on the PD1 Cifar100 Wide Resnet 2048 Benchmark \\\n",
    "In hposuite, continuations is set to True by default for each run, unless the optimizer doesn't support it. In that case, it automatically gets set to False."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Creating and Running the Study"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Some stuff might fail: issue in joblib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:hposuite.study:Generated 5 runs\n",
      "INFO:hposuite.study:Created study at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp\n",
      "INFO:hposuite.study:Dumped experiments to /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/dump_all.txt\n",
      "INFO:hposuite.study:Overwrite flag is set to True. Existing results will be overwritten!\n",
      "INFO:hposuite.study:Running 5 experiments sequentially\n",
      "INFO:hposuite.study:Running experiment 1/5\n",
      "INFO:hposuite.run:Overwriting optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=383329928 in `state=<State.PENDING: 'PENDING'>` at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=383329928.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m2025-03-17 14:45:16.100\u001b[0m | \u001b[33m\u001b[1mWARNING \u001b[0m | \u001b[36mdehb.optimizers.dehb\u001b[0m:\u001b[36m__init__\u001b[0m:\u001b[36m264\u001b[0m - \u001b[33m\u001b[1mA checkpoint already exists, results could potentially be overwritten.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:hpoglue._run:COMPLETED running optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=383329928\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=383329928/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=383329928.parquet\n",
      "INFO:hposuite.study:Running experiment 2/5\n",
      "INFO:hposuite.run:Overwriting optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=3324115917 in `state=<State.PENDING: 'PENDING'>` at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=3324115917.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=3324115917\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=3324115917/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=3324115917.parquet\n",
      "INFO:hposuite.study:Running experiment 3/5\n",
      "INFO:hposuite.run:Overwriting optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=2811363265 in `state=<State.PENDING: 'PENDING'>` at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=2811363265.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=2811363265\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=2811363265/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=2811363265.parquet\n",
      "INFO:hposuite.study:Running experiment 4/5\n",
      "INFO:hposuite.run:Overwriting optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1884968545 in `state=<State.PENDING: 'PENDING'>` at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1884968545.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1884968545\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1884968545/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1884968545.parquet\n",
      "INFO:hposuite.study:Running experiment 5/5\n",
      "INFO:hposuite.run:Overwriting optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1859786276 in `state=<State.PENDING: 'PENDING'>` at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1859786276.\n",
      "INFO:hpoglue._run:COMPLETED running optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1859786276\n",
      "INFO:hposuite.run:Results dumped at /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1859786276/optimizer=DEHB.benchmark=pd1-cifar100-wide_resnet-2048.objectives=valid_error_rate.fidelities=epoch.TrialBudget=100.seed=1859786276.parquet\n",
      "INFO:hposuite.study:Completed study with 5 runs\n"
     ]
    }
   ],
   "source": [
    "from hposuite import create_study\n",
    "\n",
    "study = create_study(\n",
    "    name=\"continuations-comp\",\n",
    "    output_dir=\"example-outputs\",\n",
    "    optimizers=\"DEHB\",\n",
    "    benchmarks=(\n",
    "        \"pd1-cifar100-wide_resnet-2048\",\n",
    "        {\n",
    "            \"objectives\": \"valid_error_rate\"\n",
    "        }\n",
    "    ),\n",
    "    num_seeds=5,\n",
    "    budget=100,\n",
    ")\n",
    "\n",
    "study.optimize(overwrite=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plotting the Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Some stuff might fail: issue in joblib\n",
      "INFO:__main__:Found benchmarks: ['pd1-cifar100-wide_resnet-2048']\n",
      "INFO:__main__:Processing benchmark: pd1-cifar100-wide_resnet-2048\n",
      "INFO:__main__:Plotting DEHB_default\n",
      "INFO:__main__:Saved plot to /home/soham/repos/hposuite/examples/example-outputs/continuations-comp/plots/DEHB_default.pd1-cifar100-wide_resnet-2048, objective='valid_error_rate', fidelity='epoch', cost=None, FidelityBudget=19110.0, to_minimize=True, error_bars='std'.png\n"
     ]
    }
   ],
   "source": [
    "!python -m hposuite.plotting.incumbent_trace --study_dir continuations-comp --output_dir example-outputs --figsize 12 7"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Displaying the Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "\n",
    "Image(\n",
    "    filename=\"example-outputs/continuations-comp/plots/DEHB_default.\"\n",
    "        \"pd1-cifar100-wide_resnet-2048, objective='valid_error_rate', \"\n",
    "        \"fidelity='epoch', cost=None, FidelityBudget=19110.0, \"\n",
    "        \"to_minimize=True, error_bars='std'.png\"\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "hposuite_env",
   "language": "python",
   "name": "hposuite_env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}