{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "sixth-surprise",
   "metadata": {},
   "source": [
    "# Benchmark clustering performance"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "appreciated-desktop",
   "metadata": {},
   "source": [
    "Remember to compile `cnnclustering` with `TRACE_CYTHON=0` if timings should be measured. `TRACE_CYTHON=1` is required for line profiling and test coverage and will cause substantial overhead."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fleet-highlight",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Function-definitions\" data-toc-modified-id=\"Function-definitions-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Function definitions</a></span><ul class=\"toc-item\"><li><span><a href=\"#Benchmark-helper\" data-toc-modified-id=\"Benchmark-helper-1.1\"><span class=\"toc-item-num\">1.1&nbsp;&nbsp;</span>Benchmark helper</a></span><ul class=\"toc-item\"><li><span><a href=\"#Collect\" data-toc-modified-id=\"Collect-1.1.1\"><span class=\"toc-item-num\">1.1.1&nbsp;&nbsp;</span>Collect</a></span></li><li><span><a href=\"#Initialisation\" data-toc-modified-id=\"Initialisation-1.1.2\"><span class=\"toc-item-num\">1.1.2&nbsp;&nbsp;</span>Initialisation</a></span></li></ul></li><li><span><a href=\"#Timed-functions\" data-toc-modified-id=\"Timed-functions-1.2\"><span class=\"toc-item-num\">1.2&nbsp;&nbsp;</span>Timed functions</a></span></li><li><span><a href=\"#Data-set-generation-functions\" data-toc-modified-id=\"Data-set-generation-functions-1.3\"><span class=\"toc-item-num\">1.3&nbsp;&nbsp;</span>Data set generation functions</a></span></li><li><span><a href=\"#Data-transformation\" data-toc-modified-id=\"Data-transformation-1.4\"><span class=\"toc-item-num\">1.4&nbsp;&nbsp;</span>Data transformation</a></span></li><li><span><a href=\"#Plotting\" data-toc-modified-id=\"Plotting-1.5\"><span class=\"toc-item-num\">1.5&nbsp;&nbsp;</span>Plotting</a></span></li></ul></li><li><span><a href=\"#Runs\" data-toc-modified-id=\"Runs-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Runs</a></span><ul class=\"toc-item\"><li><span><a href=\"#Example\" data-toc-modified-id=\"Example-2.1\"><span class=\"toc-item-num\">2.1&nbsp;&nbsp;</span>Example</a></span></li><li><span><a href=\"#CommonNN-Clustering\" data-toc-modified-id=\"CommonNN-Clustering-2.2\"><span class=\"toc-item-num\">2.2&nbsp;&nbsp;</span>CommonNN Clustering</a></span><ul class=\"toc-item\"><li><span><a href=\"#Run-a\" data-toc-modified-id=\"Run-a-2.2.1\"><span class=\"toc-item-num\">2.2.1&nbsp;&nbsp;</span>Run a</a></span></li><li><span><a href=\"#Run-b\" data-toc-modified-id=\"Run-b-2.2.2\"><span class=\"toc-item-num\">2.2.2&nbsp;&nbsp;</span>Run b</a></span></li><li><span><a href=\"#Run-c\" data-toc-modified-id=\"Run-c-2.2.3\"><span class=\"toc-item-num\">2.2.3&nbsp;&nbsp;</span>Run c</a></span></li><li><span><a href=\"#Run-c-b\" data-toc-modified-id=\"Run-c-b-2.2.4\"><span class=\"toc-item-num\">2.2.4&nbsp;&nbsp;</span>Run c b</a></span></li><li><span><a href=\"#Run-c-c\" data-toc-modified-id=\"Run-c-c-2.2.5\"><span class=\"toc-item-num\">2.2.5&nbsp;&nbsp;</span>Run c c</a></span></li><li><span><a href=\"#Run-d\" data-toc-modified-id=\"Run-d-2.2.6\"><span class=\"toc-item-num\">2.2.6&nbsp;&nbsp;</span>Run d</a></span></li><li><span><a href=\"#Run-d-b\" data-toc-modified-id=\"Run-d-b-2.2.7\"><span class=\"toc-item-num\">2.2.7&nbsp;&nbsp;</span>Run d b</a></span></li><li><span><a href=\"#Run-d-c\" data-toc-modified-id=\"Run-d-c-2.2.8\"><span class=\"toc-item-num\">2.2.8&nbsp;&nbsp;</span>Run d c</a></span></li><li><span><a href=\"#Plots\" data-toc-modified-id=\"Plots-2.2.9\"><span class=\"toc-item-num\">2.2.9&nbsp;&nbsp;</span>Plots</a></span></li></ul></li></ul></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "second-teacher",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-01T10:12:44.937324Z",
     "start_time": "2021-06-01T10:12:44.926497Z"
    }
   },
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "import json\n",
    "from operator import itemgetter\n",
    "import os\n",
    "import pathlib\n",
    "import time\n",
    "import timeit\n",
    "\n",
    "import cnnclustering\n",
    "from cnnclustering import cluster\n",
    "from cnnclustering import _fit, _primitive_types, _types\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.axes_grid1.inset_locator import zoomed_inset_axes, mark_inset\n",
    "from sklearn import datasets\n",
    "from sklearn.metrics import pairwise_distances\n",
    "from sklearn.neighbors import KDTree\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from tqdm.notebook import tqdm\n",
    "\n",
    "import helper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "signal-democrat",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:14:21.908400Z",
     "start_time": "2021-05-31T11:14:21.905166Z"
    }
   },
   "outputs": [],
   "source": [
    "mpl.rcParams[\"figure.dpi\"] = 300"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "verified-profile",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:52.952275Z",
     "start_time": "2021-05-06T14:49:52.941568Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              Python :   3.8.8 | packaged by conda-forge | (default, Feb 20 2021, 16:22:27)  [GCC 9.3.0]\n"
     ]
    }
   ],
   "source": [
    "print(f\"{'Python':>20} :  \", *sys.version.splitlines())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "aerial-generic",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:52.965778Z",
     "start_time": "2021-05-06T14:49:52.954688Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cnnclustering :   0.3.11\n"
     ]
    }
   ],
   "source": [
    "print(f\"cnnclustering :  \", cnnclustering.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ranking-thickness",
   "metadata": {},
   "source": [
    "## Function definitions"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "precise-recall",
   "metadata": {},
   "source": [
    "### Benchmark helper"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fewer-headline",
   "metadata": {},
   "source": [
    "#### Collect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "graphic-field",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:36.959335Z",
     "start_time": "2021-05-31T11:08:36.942155Z"
    }
   },
   "outputs": [],
   "source": [
    "def collect_timings(\n",
    "        gen_func, setup_func, run_arguments_list,\n",
    "        transform_func=None, timings=None, repeats=10):\n",
    "    \"\"\"Orchestrate timings\n",
    "    \n",
    "    Args:\n",
    "        gen_func: A function, returning data. Called with \n",
    "            run arguments \"gen\".\n",
    "        setup_func: A function, accepting data and returning a\n",
    "            function which should be timed. Called with\n",
    "            run arguments \"setup\".\n",
    "        run_argumens_list: A list of run arguments.\n",
    "        \n",
    "    Keyword args:\n",
    "        transform_func: A function, transforming generated data before setup.\n",
    "        timings: An optional timings mapping which results should\n",
    "            put into.\n",
    "        repeats: Repeats the timing *n* times. Using timeit -n/-r directly would\n",
    "            not ensure running the setup before each timing.\n",
    "    \n",
    "    Returns:\n",
    "        timings mapping\n",
    "    \"\"\"\n",
    "    \n",
    "    # Timed function has to be in global namespace to be discovered by %timeit magic\n",
    "    global timed_args\n",
    "    global timed_kwargs\n",
    "    global timed_func\n",
    "    \n",
    "    if timings is None:\n",
    "        timings = {}\n",
    "    \n",
    "    progress = tqdm(run_arguments_list, desc=\"Run completed\")\n",
    "    \n",
    "    for run_index, arguments in enumerate(progress):\n",
    "        \n",
    "        gen_args, gen_kwargs = arguments.get(\"gen\", ((), {}))\n",
    "        data = gen_func(*gen_args, **gen_kwargs)\n",
    "        \n",
    "        if transform_func is not None:\n",
    "            trans_args, trans_kwargs = arguments.get(\"transform\", ((), {}))\n",
    "            data = transform_func(data, *trans_args, **trans_kwargs)\n",
    "        \n",
    "        timeit_results = []\n",
    "        for _ in range(repeats):\n",
    "            try:\n",
    "                setup_args, setup_kwargs = arguments.get(\"setup\", ((), {}))\n",
    "                timed_func = setup_func(data, *setup_args, **setup_kwargs)\n",
    "\n",
    "                timed_args, timed_kwargs = arguments.get(\"timed\", ((), {}))\n",
    "                o = %timeit -n 1 -r 1 -q -o timed_func(*timed_args, **timed_kwargs)\n",
    "            except MemoryError:\n",
    "                o = \"MemoryError\"\n",
    "                break\n",
    "\n",
    "            finally:\n",
    "                timeit_results.append(o)\n",
    "        \n",
    "        run_id = arguments.get(\"id\", str(run_index))\n",
    "        tqdm.write(f\"Timed run ID: {run_id:>10}    ({datetime.now().strftime('%d.%m.%Y %H:%M:%S')})\")\n",
    "        timings[run_id] = combine_timeit_results(*timeit_results)\n",
    "        \n",
    "    return timings"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "molecular-vacuum",
   "metadata": {},
   "source": [
    "#### Initialisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "fourth-herald",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:44.845259Z",
     "start_time": "2021-05-31T11:08:44.840752Z"
    }
   },
   "outputs": [],
   "source": [
    "def make_maps():\n",
    "    \"\"\"Initialise benchmark setting/result mappings\"\"\"\n",
    "    \n",
    "    global RUN_ARGUMENTS_MAP\n",
    "    global RUN_TIMINGS_MAP\n",
    "    \n",
    "    if \"RUN_ARGUMENTS_MAP\" not in dir():\n",
    "        RUN_ARGUMENTS_MAP = {}\n",
    "\n",
    "    if \"RUN_TIMINGS_MAP\" not in dir():\n",
    "        RUN_TIMINGS_MAP = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "impressive-excuse",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:46.259512Z",
     "start_time": "2021-05-31T11:08:46.254828Z"
    }
   },
   "outputs": [],
   "source": [
    "def del_maps():\n",
    "    %xdel RUN_ARGUMENTS_MAP\n",
    "    %xdel RUN_TIMINGS_MAP"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "republican-elephant",
   "metadata": {},
   "source": [
    "### Timed functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "consecutive-repeat",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:49.996463Z",
     "start_time": "2021-05-31T11:08:49.993019Z"
    }
   },
   "outputs": [],
   "source": [
    "def setup_sleep(data):\n",
    "    \"\"\"Dummy example\"\"\"\n",
    "\n",
    "    return time.sleep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "guilty-dress",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:50.397668Z",
     "start_time": "2021-05-31T11:08:50.392225Z"
    }
   },
   "outputs": [],
   "source": [
    "def setup_commonnn_clustering__fit(data, preparation_hook=cluster.prepare_pass, recipe=None):\n",
    "    \"\"\"Prepare benchmark of :meth:`cnnclustering.cluster.Clustering._fit`\"\"\"\n",
    "    \n",
    "    if recipe is None:\n",
    "        recipe = {}\n",
    "\n",
    "    clustering = cluster.prepare_clustering(data, preparation_hook, **recipe)\n",
    "    clustering._labels = _types.Labels(\n",
    "        np.zeros(clustering._input_data.n_points, order=\"C\", dtype=_primitive_types.P_AINDEX)\n",
    "        )\n",
    "\n",
    "    return clustering._fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "atmospheric-persian",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:50.795442Z",
     "start_time": "2021-05-31T11:08:50.791268Z"
    }
   },
   "outputs": [],
   "source": [
    "def setup_commonnn_clustering_fit(data, preparation_hook=cluster.prepare_pass, recipe=None):\n",
    "    \"\"\"Prepare benchmark of :meth:`cnnclustering.cluster.Clustering.fit`\"\"\"\n",
    "    \n",
    "    if recipe is None:\n",
    "        recipe = {}\n",
    "\n",
    "    clustering = cluster.prepare_clustering(data, preparation_hook, **recipe)\n",
    "\n",
    "    return clustering.fit"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "military-harvard",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-04-29T19:49:44.479666Z",
     "start_time": "2021-04-29T19:49:44.476020Z"
    }
   },
   "source": [
    "### Data set generation functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "threaded-liechtenstein",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:52.407504Z",
     "start_time": "2021-05-31T11:08:52.404252Z"
    }
   },
   "outputs": [],
   "source": [
    "def gen_dummy():\n",
    "    return []"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "brave-marks",
   "metadata": {},
   "source": [
    "### Data transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "operating-diagram",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:53.986759Z",
     "start_time": "2021-05-31T11:08:53.981888Z"
    }
   },
   "outputs": [],
   "source": [
    "def compute_neighbours(data, radius, sort=False):\n",
    "    \n",
    "    tree = KDTree(data)\n",
    "    neighbourhoods = tree.query_radius(\n",
    "        data, r=radius, return_distance=False\n",
    "        )\n",
    "\n",
    "    if sort:\n",
    "        for n in neighbourhoods:\n",
    "            n.sort()\n",
    "            \n",
    "    return neighbourhoods"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "accepted-handling",
   "metadata": {},
   "source": [
    "### Plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "spoken-military",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:08:58.501118Z",
     "start_time": "2021-05-31T11:08:58.497471Z"
    },
    "run_control": {
     "marked": true
    }
   },
   "outputs": [],
   "source": [
    "def annotate_memory_error(\n",
    "        ax, line, memory_error_id):\n",
    "    \n",
    "    last_sample_pos = line.get_xydata()[-1]\n",
    "    memory_error_pos = np.array([memory_error_id, last_sample_pos[-1]])\n",
    "    ax.plot(\n",
    "        *np.vstack([last_sample_pos, memory_error_pos]).T,\n",
    "        color=\"k\",\n",
    "        linestyle=\"--\",\n",
    "        zorder=0\n",
    "    )\n",
    "    ax.plot(\n",
    "        *memory_error_pos,\n",
    "        color=line.get_color(),\n",
    "        marker=\"X\",\n",
    "        markeredgecolor=\"k\",\n",
    "        zorder=0\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "incident-dealing",
   "metadata": {},
   "source": [
    "## Runs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "recovered-instrument",
   "metadata": {},
   "source": [
    "### Example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "ancient-truth",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:56.344959Z",
     "start_time": "2021-05-06T14:49:53.246931Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "210180cd645842b19c0962839774b41d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/2 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:          1    (06.05.2021 16:49:54)\n",
      "Timed run ID:          2    (06.05.2021 16:49:56)\n",
      "\n",
      "    Run ID: Factor\n",
      "=======================\n",
      "         1: 1.00\n",
      "         2: 2.00\n"
     ]
    }
   ],
   "source": [
    "example_run = [\n",
    "    {\n",
    "        \"id\": \"1\",\n",
    "        \"gen\": (\n",
    "            (), {}\n",
    "        ),\n",
    "        \"setup\": (\n",
    "            (), {}\n",
    "        ),\n",
    "        \"timed\": (\n",
    "            (0.1,), {\n",
    "            }\n",
    "        ),\n",
    "    },\n",
    "    {\n",
    "        \"id\": \"2\",\n",
    "        \"gen\": (\n",
    "            (), {}\n",
    "        ),\n",
    "        \"setup\": (\n",
    "            (), {}\n",
    "        ),\n",
    "        \"timed\": (\n",
    "            (0.2,), {\n",
    "            }\n",
    "        ),\n",
    "    }\n",
    "]\n",
    "\n",
    "sleep_timings = collect_timings(\n",
    "    gen_dummy,\n",
    "    setup_sleep,\n",
    "    example_run\n",
    ")\n",
    "\n",
    "print()\n",
    "print_ratios(get_ratios(sleep_timings))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "animated-constitution",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:56.353310Z",
     "start_time": "2021-05-06T14:49:56.347698Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'1': <TimeitResult : 100 ms ± 2.42 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2': <TimeitResult : 200 ms ± 3.87 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sleep_timings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "white-material",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:56.483054Z",
     "start_time": "2021-05-06T14:49:56.355152Z"
    },
    "run_control": {
     "marked": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAboAAAFHCAYAAADX+4lCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAC4jAAAuIwF4pT92AAAtXklEQVR4nO3dd5wV5dn/8c9FR3ovht6L2BtWEIyICviImhiNLfbkieb5pZqoSZ6UJz0REzVGg4kmIQpIVCyoqNiCGBWW3kGadKRsu35/zFmZXbacXfacOTP7fb9e+9qZOfe956Lsfndm7rlvc3dERESSql7UBYiIiGSSgk5ERBJNQSciIommoBMRkURT0ImISKIp6EREJNEUdCIikmgKOhERSTQFnYiIJJqCTkREEk1BJyIiiaagExGRRFPQiYhIoinoREQk0RR0IiKSaAo6ERFJNAWdiIgkmoJOREQSTUEnIiKJpqATEZFEU9CJiEiiKehERCTRFHQiIpJoCjoREUk0BZ2IiCSagk5ERBJNQSciIommoBMRkURT0ImISKIp6CJkZk3N7Etm9oqZrTOzvWa2yMzuN7Ojoq5PRCQJzN2jrqFOMrNjgBnAZypp9iBwm7vn1/J7Nwe+BowF+gDNgI3AImAa8Fd3312b7ykiEhUFXQTMbDgwE2iROrQReBHYCxwDnAhY6rWpwCXuXlxL730+8AegWyXNNgJfdfe/18Z7iohESUGXZWbWBPgQ6Js6dD9wh7vvDbUZDfwF6Jg6dLu7/7oW3vsC4CmCEN0PvAe8D3QmCNcjy3T5urv/7HDfV0QkSgq6LDOz7wPfTe0+SXC2dsg/gpmNAp4nCKU9QHd3334Y79sN+A/QFlgA/Je7Ly7T5rPAZA4GrAMXuPszNX1fEZGoKeiyyMwaAB8BHYAiYKC7L6uk/VRgfGr32+7+48N478eBywkuSw5w910VtOsMvAX0SB1aC/Rx94KavndNmVkr4KzQobVArd6vFJHD0ojSt0Fmu/vOqIqpiIIui1KXDmekdme4+0VVtL8ImJ7afdvdT6nh+7YH1hP8p7zW3R+uov1E4B+hQ1e6+19q8t6Ho8yfX0Ry3zh3fyrqIsrS4wXZdW5o++002r8W2j7RzNrW8H0/TxBy64FHqmrs7lMI7iOWuKCG7ysiEjkFXXadENr+sMJWKal7cltSu/WAYTV832nAVuCp8u4HVmBuaPukGr6viEjkGkRdQB0Tfgj8gzT7LCa4pwfBSM1Xqvum7r7GzK4GCqvRbUFou3N137OWrA3vTJs2jb59+1bUVkRqSUFBAaNGjWLr1q2fHmvcsjP9zv1/1G/Q6NNj+3duYPGzPwp3LfU9mysUdFmSGojSPHRoTZpdlwCnp7bLDv9Pm7v/q5pdwoNPttX0fQ9TqYEnffv2ZciQIRGVIlJ3TJkypVTIAfQbfQctOlb5i2ZODhaL1aVLM2trZkebWcNK2piZ3Wxmc81sl5ktNbNHzaxnFkstT6vQdkE1HgDfEdpuXHvlVKlnaHt+Ft9XRCI2c+bMUvstug6hRacBEVVz+OJ2RvdV4DtAvpmNcfdXwi+mAvBp4JySQwTTW/UGLjazq1MDLaIQDrrq/NazN7TdpJZqScf5oe13D/eLmVlHDl6CTVefw31fEam+uXPnltpv16tGA75zRtyC7lqC8NpJ6RGJJX4GjCJ40LnkQetGqY+mwMNm9rq7b8hOuaW0DG1XJ+jCZ35ZCTozOxUo+fVtL/C7WviytwB31cLXEZEMW7JkSan9Zh3i/TtnbC5dmtkAoCtBiE1296Iyr3cn+GEKQThcTTALSGvgr6njTYGbslBueVqEtqszb6WFtjN+6dLMDPh16NCv3H1jpt9XRHLHgQMHSu3Xb9wsokpqR2yCDjg+tD2nnNcnEJyhOvCAu09290J33w/cCGxKtRud2TIrFA5mq7DVoRqFtqszarKmbufg4wRzgB9k4T1FJIc0blz6d+qiA59EVEntiNOly66h7fJGLIaninok/IK77zWzlwmmwIrqHDz8K1J1zszCQZfREU1mdiHB5V+A5cB4dz9QSZfquA+o7v3RPmhmFJGsylu1nWZtjmT/huWfHvtky3Jado3viOc4BV3T0HZ5QVfyMPZeyh88sT71uVU5r2XD/tD2EWZWL82Rl+Gg219hq8OUui/3OMFZ/jJgtLt/XFtf3903A5urWVNtvb2IVGHdlk+4f8ZCZr37EfVa9YJQ0G1d+RZdjq50xsKcFqegC48+LHUJz8z6Eixg6sB/KgiQTqnPOzJSXdU2hbaNYCLU1Wn06xjazsi9MjM7mWB9vGYEM7acq/tyInXDjj0HeGTmUp58dSWFRcHESW26H8eWRS9+2mb3RwvYvWlxbB8xiNM9uvBMHUeXeS08FP6tCvqXXLKMYsQlqbOj8GXAdKf4CF8vqPXazewk4DmCUaFvAGcp5ESS70B+EZOfX8rEu2bxj5dXfBpyAG16nUTDpq1LtV/+0u8oKqytOxnZFaegm8PBAR33mFk9ADNrDXw51O7pCvr3Izjjq3BZnCwIj9ntV1Xj1HOB4XaLarMYMzuRYM27VgRhN/pw1rwTkdxXVOw8/eYaLrtnFn+YvpBP9h86xq1e/YZ0OWZcqWP7tq9lycyfxjLsYhN07r4beJHgst+ZwAYzm04QHr0JQmxe2YfIAczsNKB9aveFrBRcvvBo0XSuAfQDSmaB2Uv682NWycyO52DIPQFcGF7lXESSxd15K28z1/xkNv/7l/+weUflt/y7DLvokOfndqx5lw+nfI3dmxZX0Cs3xSboUr5G8LA4BMF1AdCOIPx2AFdV0O/G1OdiKj7jy4aXQ9tj0mg/IrQ9x91r5fECMzuOIPBbEzxjeFkUC6uKSHYsXruTr977JndMeotl68tdc/kQ/bq14cE/PkzLli1LHd+3fS3zn/g686d+iy2LX66gd26J02AU3D3PzEYBkyi9dMybwFXuvrxsHzM7Frgitfu0u68v2yaLniII5NbAADM7yd3fqaT9daHtR2qjgNTfxwtAG+BPwJeqMe+miMTIhq17eWDGIp7797q0+3Ro3YQbLhjIeSd3o349o8ezzzJmzBh27SodkLs35LF7Q15tl5wRsQo6AHd/FzglNXdiD2Cju1e2NMSHwDGpj7mVtMs4d99vZn8G/jt16LdmNry8oDGzLwHHpna3AE8e7vub2dEEIdcW+CNwQzXWpxORmNi1N5/JM5cyZfZKCgrT+z22WZMGXPnZflx2dm8aN6r/6fHhw4fz5ptvcuWVVzJv3rxMlZxRpp9z2ZUaPLOIg487/B243t33hNpcDvwFKPnfdunhTkZtZsOAlwgu9T4I3JjrIWdmQwitnDB//nwt0yNSifyCIp54dRWPzFzC7r3p3Y1oUN+4+MxeXH1eP1o3r3gui4KCAn71q1/xy1/+kk2bNlXUbKi7L6joxago6CJgZmMIztBKJmneTDBQZR0wnNLTnf3J3a+jAmb2MMG8ngCvuPuIctp0A+YR3NfcCdxP9ebbxN2/VZ32tUFBJ5K+dxZu5qePf8CGremPKTvnuK7ceNEgPtMh/bks8/PzmT59OjNnzmTOnDksXlxqYEpOBl3sLl0mgbs/m7rX+ATBmV1Hgrk6w4qBH7p7bcz435uDo05bAV+vwdfIetCJSPqaNm6Qdsgd07cdt44fzJBebar9Po0aNWLixIlMnDiRBQsWMHTo0Gp/jWxT0EXE3eeYWS+CpYcuBfoTrEC+AngV+I27R/nMn4jEyFG923L2MV145T8VzyvRs3Nzbhk/mNOGdqpTU+wp6CLk7vsIRpBOOoyvcQ1wTRVtZlO9FRNEJIZuumgQr32wkaLi0rek2rVszPUXDGTsKd1oUD9uT5UdPgWdiEhCdO/UnHGn9+DJV1cBcETj+lwxui+Xj+xD08Z198d93f2Ti4jEQEFhMfNXbufYfu3San/t+QN4Ye56Rp9wJNeO6U/blk2q7pRwCjoRkRzk7rz83gZ+Pz2Pjdv28fh3R/CZjs2r7Ne2RWOe/P4omjVtWGXbukJBJyKSY/6zbCuTpuaxYNXBOdbvn7GIH1x3QiW9DlLIlaagExHJEas27ub30xfy2geHrpQ1a95HXD5ye40eCajrFHQiIhHbunM/Dz2zmBlvrDlkxGTYvVMXcN/tp9WpRwNqg4JORCQie/cX8tisZTz+4nL25RdV2f795dt4/cNNnDGscxaqSw4FnYhIlhUWFTPjjTU89PRitu1ObyHTegYXDO/BwO6tM1tcAinoRESyxN157YON3Dd9IWs27am6Q8ppQztx87hB9O7asurGcggFnYhIFsxfuY1JU/N4f/m2tPsM7N6a2yYM5rj+7atuLBVS0ImIZNDazXv4w1MLefm9iuegLKtruyO4adwgRh7blXr1NPDkcEUSdGbWFbgI+Je7p7/0rYhITGzffYCHn13C1NdWVTqSMqxls4Zcc15/JpzRk0YN61fdQdIS1RndaQQTGd9rZrMJFhl9wt13Vd5NRCT3PffOOn729w/Yu78wrfaNGtTj0hG9ufLcfrQ4Qg9717ZIgs7dp5jZ+8BVwOeBh4BJZjaDIPSedff0/oeIiOSYLu2aphVyZjDmpG5cf8EAOrc9IguV1U2Rrdfg7kvc/U537w2cCUwGRgHTgI1mNsnMhkdVn4hITQ3r046zjq78WbeTB3XgkW+exZ1XHauQy7CcGIzi7q8Dr5vZl4GxwBcIFiS9ycxWE5zlPebuiyIsU0QkbTeNG8zrH2465P5cvyNbcuuEwZw0qGNEldU9ObUCn7sXuPs0d78E6AzcCKwBvgMsMLN/m9lXzKxTpIWKiFShR6fmXHRaj0/3O7VpynevOpaHv3mWQi7Lcirowtx9p7v/0d3PBnoShF1j4NfAOjN71syuMLNm0VUpInXJrk/yefX99B8TuO78/nRs3YRbxw/mb3eNZMzJ3fS4QARy4tJlVdx9LfAT4CdmNgy4Evgc8Ciw18ymElzefMHdi6OrVESS6EBBEf98ZSWTn1vKvvxCHv/eSI5sX/Xv2G1bNuGf3x9Fg/o5e05RJ8Tub9/dP3D3/wd0A0YDUwieyXsG+MjMfm1m6S3aJCJSieJiZ+Y7a7n8npeYNC2P3fsKKCxy/vDUwrS/hkIuerH9F/DALHe/BuhE8JjCv4FbgLfNbLGZ3WlmvSMtVERi6d+LtnDNT2fz/T+/x6bt+0q9Nuvdj8gLLYoquS22QRfm7vvd/e/ufiHQBfgKsBX4PrDUzOaY2c1m1i7SQkUk5y1dt5Pb732T//7dmyxdV/EcFpOm5eGe3ownEq1EBF2Yu29190nuPhzoRxB2HQhmYvnIzKab2UQzaxJpoSKSUzZt38cPJ7/H1T+ZzdsLt1TZ/r2lW3ljweYsVCaHKxaDUWrK3ZcD9wD3mNnJBM/nXQpcCOw2s38CfwVedv1qJlIn7d5bwKPPL+Ufr6wgvyC9sWxHNGnAlaP7cnx/XSSKg0QHXZi7v01w7+6rwHnAFcBlwNUEM7H8Ffiru78fWZEikjX5BUVMfW0VD89cwq5PCtLqU7+eMeGMnlw9pj9tWzTOcIVSW+pM0JVw9yLgaeBpM2sOXExwpncH8DUzW0gwHdlkd98YXaUikgnFxc5L733EH6Yv5KOte9PuN+LYLtx00SC6dWyeweokE+pc0IW5+x5SoWZmnQlGbl5B8MzeLQQPqotIQsxb8jGTpuWxcPWOtPsM692W2y4ezNBebTNXmGRUnQ66sNTZ2y+BX5rZQED/q0USYsVHu/j99IXMmb8p7T7dOzXnlnGDOGNYZ8w0m0mcKejKocmjRZLjsReXcd+0PNJc+5S2LRpz/dgBXDC8ux72TggFnYgk2uCebdIKuSaN6vP5UX343Dl9adZEPxqTRP+aIpJox/RtxxnDOvPaB+WPLatfz7hweHeuPX8A7Vvp8dokUtCJSOLdPG4Qb8w/dG2404/qzC3jB9Gzc4uIKpNsUNCJSOL17NyCC4d3Z9rrqwEY3KM1t04YwrH99MB3XaCgE5FYWrNpD/NXbuP8U7qn1f668wewYOV2vnhef0Yc20UjKesQBZ2IxMq2Xfv507NLmP76aszg6L7t0lobrl2rJjzyrbMUcHWQxs6KSCzsO1DIw88u5tK7Z/Hkq6soKnYKi5z7n0r/aSCFXN2kMzoRyWmFRcU8/dZaHnp6ER/vPHDI6y++u57Lz+nN4B5tIqhO4kBBJyI5yd2ZM38T903LY9XGPZW2nTQ1j3v/e7jO2KRcCjoRyTl5q7czaWoe7y3dmlb795Zu5d0lH3PCgA4ZrkziSEEnIjlj3ZZPuH/GQma9+1HafTq3bcqNFw7iuH7tM1iZxJmCTkQit3NPPg/PXMKTr66ksCi9SSlbNG3IF8/rx3+d1YvGDetnuEKJMwWdiETmQH4R/3hlBY8+v5Q9+wrT6tOwQT0mntWLqz7bj5bNGmW4QkkCBZ2IZF1RsfPcO2t5YMYiNu/Yn3a/z574GW64cCBd2h2RweokaRR0IpJVb+Vt5r5peSxbvyvtPicMaM+t4wczoHvrzBUmiRXroDOz5sBwoAvQCnjf3WeXadMV6A40dfeXs1+liJT47RPz+dtLK9Ju36drC26dMISTB3XQowNSY7EMutQK4N8DLgHCd6F/Aswu0/xG4M5Uv/Pc/YWsFCkihzhjWJe0gq5D6ybccMFAzju5G/XrKeDk8MQu6MxsFDAVOAIwIB9oBFQ0VOsHwHjgKOC3wKDMVyki5Tm2XztOP6ozr39Y/tpwzZo04MrP9uOys3vTuJFGUkrtiNVcl2Z2AvAvoBnwCfAVoGNlfdy9EPhxare/mX02o0WKSKVuGT/okLO0BvWNS0f0Zso953DVuf0UclKrYhV0wM8Jzt7ygYvd/V53T+eO9j+B7antiZkqTkSqVrI2XIlzjuvKY98dyVcvGUrr5o0jrEySKjaXLs2sL3AmwSXKSe7+Yrp93b3QzN4CxhAMXhGRWjR38RZWbdjDJWf3Sqv9decPYOO2fVx3/gCG9NJkzJJZsQk64PTQ9mM16L8s9fnIWqhFRIDl63dx37Q83szbTMMG9Rg+tCNd01wb7pe3npKFCkXiFXSdQ9vLKmxVsZInTOP0ZxbJSZu37+PBfy3imbfX4qlhYAWFxTwwYxF3X3N8tMWJlBGnH/rbQ9udgZ3V7H9S6nP6s8WKSCl79hXwlxeW8feXVnCgoOiQ15+fu57LRvZhUI/W2S9OpAJxGoyyMLRdrWseZtYTGEJwf+/NWqxJpE4oKCxmyisruPTuWUx+bmm5IVdi0tQFuKc3MbNINsTpjO51YBPB4wTfNbPH3L2gqk5m1gB4nCDUHfhbRqsUSRB35+X3NvD76Xms/3hvWn3mLd3KojU7dVYnOSM2QefuxWb2Q+B3QC/g72Z2WWVhZ2btgHuBk0mdzbn7M1kpWCTm/rNsK5Om5rFg1faqG6cM7dWG2yYMVshJTolN0KX8HpgAjATGAe+Y2T3hBmbWlGD2k7OBbwMlY5d3ANdlq1CRuFq1cTf3TVtY4ewl5enWsRk3jxvEWUd30ZyUknNiFXSps7rxwDMEjxsMA57g4PRfXwG+QTA1WAkDtgHj3X1x9qoViZetO/fzx6cXM+ON1RSneYutdfNGXDd2AONO60GD+nG65S91SayCDsDd95jZCOC7wNcIpgODIOxK5r8Mex64yd1XZa1IkRjZu7+Qx2Yt4/EXl7Mvv+JBJmFNGtXnc+f04fPn9KFZ04YZrlDk8MQu6ADcvQi428x+B1wFnAcMBjoAe4D1wEvA3939rcgKFclhhUXFzHhjDQ89vZhtuw+k1aeewdhTu3P92IF0aN0kwxWK1I5YBl0Jd98K/Cr1ISLV8IPJ7/HC3PVptz9taCduHjeI3l1bZrAqkdoX66ATkZobf3qPtIJuYPfW3DZhMMf1b5+FqkRqn4JOpI46tl97Tj+qE69/uKnc17u2O4Kbxg1i5LFdqafFTyXGFHQiddjN4wbzxvxNpUZZtmzWkGvO68+EM3rSqKHWhZP4i23QmdkgoA/QiWCNurS5++8zUpRIzPTq0oILhvfgqTmradSgHpeO6M2V5/ajxREaSSnJEbugM7NLCJ6VO+4wvoyCThKpqNh55q01bN+Tz1Xn9kurz/VjB+DuXDOmP53bHlF1B5GYiVXQmdl1wAMluzX8MpptVhLH3XljwWbum5bHyg27adigHqOPP5Iu7aoOrvatmvCtK47JfJEiEYlN0KWm9voFQcCVhNUOgrXpPkEBJnXUwtU7mDQtj3lLPv70WEFhMffPWMjdV2ttOJHYBB1wDtCSINAc+DLwR3fPj7QqkYh89PEn3D9jUYWPCDz/7/VcPrIPA7u3zm5hIjkmTkHXK7T9c3e/L7JKRCK0c08+jzy3hCdmr6SwqPILGfdOzeN3XzlVEy1LnRanoAvPGDs5sipEInIgv4gps1cw+bml7NlXmFafD5ZvZfWmPfTs3CLD1YnkrjgFXV5oe21kVYhkWXGx89y/1/HAjEVs2r4v7X6jTziSGy8cSNf2zapuLJJgcQq694FCoD7QHtgdbTkimffOws1MmpbH0nW70u5zXL923DphiBY/FUmJTdC5+2Yz+xPwJeAa4HsRlySSMUvX7WTStDzeWbgl7T69urTg1vGDOXVIR92TEwmJTdCl3AWMBb5jZrvc/edRFyRSmzZt38cDMxYy8511eJoPzLRv1YQbLhjImFO6UV9zUoocIlZB5+6bzGwM8BzwUzO7lmBgyr8J1qD7JM2vo3t8klN27y3g0eeX8o+XV5BfWJxWnyOaNOALo/ty2YjeNG0cq29lkayK43fHWmAGcAMwAPjfavZ34vnnlgT77p/mpn2Zsn49Y8IZPbl6TH/atmic4cpE4i9WP/DNrA/wBsFglLIXdqq6ZuNptBGJxFXn9ksr6EYc24WbLhpEt47Ns1CVSDLEKuiAKUCHGvZVyEnOOq5/e04b2ok588tfG25Y77bcdvFghvZqm+XKROIvNkFnZscBx3DwTO4Z4FfAfDTXpSTAzeMG8eaC0mvDde/UnFvGDeKMYZ01klKkhmITdEB4dtoXgXHunt5de5EY6N21JWNP7c6MN9bQtkVjrh87gAuGd6dB/XpVdxaRCsUp6NqEtn+okJNc98m+Av764nLqGVx/wcC0+lw/diAdWzfl8nP60KxJnL49RXJXnL6TVoa28ypsJRKxwqJips9ZzUNPL2bHnnwaNqjH2FO7p7U2XIfWTbhu7IAsVClSd8Qp6BaEtnUtR3KOuzP7/Q38fvpC1m4++EhnQWExD8xYxF1XHxdhdSJ1V2wCw93zgDdTu2dHWIrIIT5YvpUbf/E6335wbqmQK/Hcv9exeO3OCCoTkdgEXco3gCLgt2Z2dNTFiKzetIdvPfAON/1yDvNXbq+07aSpC/B05/USkVoTp0uXuPvrZnYLcD8w28x+Azzq7ssiLk3qmG279vPQM0t4as5qiorTC69l63exZcd+OrZpmuHqRCQsNkFnZucBY1K7S4H+wJ3AnWa2i/TnunR3PyUzVUrS7TtQyOOzlvPYi8vYe6AorT6NGtbj8pF9+MLovjRv2jDDFYpIWbEJOuAk4MscfDC85LMBrYCWaXwNQw+WSw0UFhXz9Ftr+eO/FrF114G0+pjB2FO6c/3YATqLE4lQnIKuREXTQ2jaCKl17s6c+Zu4b1oeqzbuSbvfqUM6csu4wfQ5Mp3fv0Qkk+IUdNOAVRHXIHVI3qrt3Ds1j/8s25p2nwHdWnHrhMGcMKCmU7KKSG2LTdC5+wfAB1HXIcm3bssn3P/UQmbN+yjtPp3bNuWmiwYx6vgjqafFT0VySmyCTiRbfjB5Hh+uqPxRgRItjmjI1ef157/O7EmjhvUzXJmI1ETcnqMTybgbLhxUZZtGDepxxag+TLnnHD53Th+FnEgO0xmdSBnH92/P8KGdeKOCteE+e+JnuOHCgWnNXSki0VPQiZTj5nGDeKvM2nAnDGjPrROGMKBbq+gKE5Fqy5mgM7Mrw/vu/mg5bbrVxnu5+9ra+DoSP+6e1gKmfUJrw/Xp2oJbJwzh5EEdtPipSAzlTNABf6b0w9yHBB2wmsN/4NvJkT+3mTUFvgBcAfQF2gJrgNnAve7+YYTlJcqGrXt5YMYiOrRuwi3jB6fV5/qxAxjWuy3nndyN+hpJKRJbOfEDv4x0Zi+pyU8dr2G/jDCzY4AZwGfKvDQg9XGDmT0I3Obu+RmupR5wG/ALd0/UHFW79uYzeeZSpsxeSUFhMY0a1OPiM3vSuW06a8M1Zeyp3bNQpYhkUq6NukwniGoaVrkUcsOBVzkYchuBvwAPAO9wMOi/BPwtFUSZqKOFmX0OeBv4Dbn5i8+nLr74Yq677jqmTJlCfn7l2Z9fUMTjs5Yz8a5ZPDZrOQWFwYL0+YXFPPivRdkoV0RyhOXKsiFmdmZ4391fjaqWTDKzJsCHBJcqIViJ4Q533xtqM5og+DqmDt3u7r+uhfceDYwDugJHAsOAJuE27p5LvxAMAeaX91qnTp244447uP3222nY8OBJaHGx8+K767l/xiI2bN1bXlfM4OFvnEV/DSoROSwLFixg6NCh4UND3X1BRe2jkjNBV1eY2feB76Z2nwQu8XL+EcxsFPA8wZnoHqC7u6f3FHPF7/0wcHVlbeISdCWOO+44Hn30UQYPHszcxVuYNDUvrQVOTxzYgd98+dTaKlWkTopL0OX0paqkMbMGwE2p3SLgG+WFHIC7v2hm04HxQPNUvx8fZgnFwA6CS6XzgVeArcDjh/l1IzNv3jxOPuVULvjSz1i9v3Pa/bbs2MeuT/Jp2axRBqsTkVyQa/foku48oGS232fSWDD24dD2uMN9c3e/zt3buPsgd5/o7pOADYf7dbOl67EX06LrkEOO79m9iyn33s7ebWuq/BrtWzXmm58/msnfPlshJ1JH1IkzOjNrDYwE9rv7MxGWcm5o++002r8W2j7RzNq6+7Zarik2OgwYQY9Tv8juTYtZ/tLv2Lf94OOQRfl7WTbr1wy9+P+oV//Q/9ZHNK7PFaP7cvnIPjRtXCf+24tISqzO6Mxseerjf6rZ9efAFOABM4tyUsITQttVPiOXuie3JbVbj2DwSJ3XotMAjpr4C1p3P77U8U+2LGfDB0+VOla/nnHxmT35x93ncM2YAQo5kTooVkEH9AJ6Am2q2e9bQAHQBbiglmuqjqNC2+kuObQ4tN23wlZ1TP0Gjel/3jdo2qb0ZDkb3n+K4qICAM4+pgt/uXME/3PZMNq2bFLelxGROiBuQVcj7r4FmJPaHRtFDamBKM1Dh6q+oRRYEto+svYqir/6DRrTZ8SXSx0r2LudFnsXcP/XTudHXzqRHp2aV9BbROqKOhF0KSsIhuqfEtH7hx/aKnD34jT77QhtN669cpKhRecBtOhSekqvbg1WcVTvthFVJCK5pi7dsOif+twlovcPB111pvQKP/Uc2+tvZtaRgyNO09UnnUbtep/K7g15n+7Pmzevmm8jIkmWk0GXxioFrdJcyaAhweW+q4EzUseiGlPeMrRdnaALn/nFNuiAW4C7MvGFj2jfu9T+kiVLKmgpInVRTgYdla9SYMDNqY/qcuCNmhZ1mFqEttO9bAml5+jUpctyNGhc+j7c/v37I6pERHJRrgYdVD4Jc02nqdoN/KiGfQ9XUWi7OvWHz0ALa6mWRCk8sKfUfpMmcT7xFZHalqtBN5vyz+jOTh3fBaQ77+N+gmfRXgMeSWM2kkw5ENquzplZOOgyulxPht1H8CxjdfQBplfVaO/HK0rt9+/fv4KWIlIX5WTQufuI8o6bWcklv/vc/TtZLKk2hK+nHWFm9dIceRkOuthek3P3zcDm6vRJdzXvrSveLLV//PHHV9BSROqinAy6hNoU2jagG8G9yKp0DG1vrNWKcl+pgUP7dx46LecnH68sNeISYMiQISxYkHMTqIskzrJlh1wgy8kJZGO1TE/ojO7HMTyjw8z2c/Cy5Sh3n5VGnwVAyYNiV7j7Y7Vc01kEqxgAObdMz0WkcelSRHLGOHd/qupm2RW3M7qzU5/TnVUk1yzh4DRg/YBKg87MGqbalahrS2NrZVSReMnJ79lYBV0CVh2fw8GgG5BG+34EzwJC8OB4uvNjJkXLqpuISA7Jye/ZWAVdArzMwYVXxwC3V9E+PChnjrvXtccL5pbZv4S6d1YrkssGAv8M7Zf9ns0JCrrseopg7srWwAAzO8nd36mk/XWh7UcyV1bO2lNmf5G7a5SJSI4oZ2R02e/ZnFCXJnWOnLvvB/4cOvRbMyv338DMvgQcm9rdAjyZ4fJERBJJQZd9d3PwUYOTgcfMrNQcVmZ2OfD70KFbUyEpIiLVpEuXWebuO8zsGoIztCbAZcAIM5sDrAOGA+Ennv/k7hXOKGJmDxNMWg3wSkUP26faHgdMLHO4e5k2Py6n5m9V+AcSEclxCroIuPuzZjYKeALoRPBQ+IQyzYqBH7p7bc74fxTwzSralPe6gk5EYktBFxF3n2NmvYBrgUsJ1strTrBA7KvAbyKcl1NEJDFiNTOK1C1mNgSYHzo0VKMuRXJHXL5HNRhFREQSTUEnIiKJpqATEZFEU9CJiEiiKehERCTR9HiB5LItwD1l9kUkd8Tie1SPF4iISKLp0qWIiCSagk5ERBJNQSciIommoBMRkURT0ImISKIp6EREJNEUdCIikmgKOhERSTQFnYiIJJqCTkREEk1BJyIiiaagExGRRFPQiYhIoinoREQk0RR0EgtmNtLMNpnZWVHXIlKXWeBaM3vezDaY2YHU5zfN7E4z6xF1jWVpPTrJWWbWADgbuAGYmDp8trvPjqwokTrMzAYCDwKnV9LsAPAz4G53L8pKYVXQCuOSM8ysL3AL0BXoAgwB2kValIgAYGbdgTlAW8CB+cB7gAEnAgNS242BO4GhZjbR3QujqfggBZ3kktOA26MuQkRKS11d+RtByG0HrnL3f5Vp0zfV5vjUofHAD4FvZq/S8ukeneSaPcBqYCbwPeCoaMsREeA64FSCM7kxZUMOwN2XAcOBGaHDXzezQdkpsWK6Ryc5z8zC/0l1j04ky8zsXeA4YLK7f7GKtj2AhUDT1KGH3P36DJdYKZ3RiYhIhcxsGEHIAfxvVe3dfTXwx9ChCzJRV3Uo6EREpDILgTeAhe6+JM0+c0PbnVIDWSKjoBMRkQq5ewFwOTC5Gt0WlNnvXHsVVZ9GXYqISKXcfS3wk2p0KSizv60Wy6k2ndGJiEht6xna3gesiKgOQEEnIiK17/zQ9nvuXhxZJSjoRESkFplZC+CS0KEfR1VLCQWdiIjUpu9xcOq+OeU9XJ5tCjoREakVZnYy8N+p3S3AVRGW8ykFnYiIHDYz6wk8BTQE9gPj3D3SQSglFHQiInJYzKwj8CzQEdgLjHf3N6Ot6iA9RyciIjVmZh2Al4CBwA5grLu/EWlRZSjoRESkRsysPTCLYO3IjcB57v5+tFUdSkEnIiLVZmbtCELuKGAlMNrdl0dbVfkUdCIiUi1m1pYg5IYBi4Bz3P2jaKuqmAajiIhI2sysDfAicDTwIXBWLoccKOhERCRNoZA7FphHsBDy5mirqpqCTkREqmRmrYEXCBZhnQuMcvdIVyVIl+7RiYhIpcysFfA8cDzwLsHAkx2RFlUNCjrJGanfGL9RRbMbzOy8MscecPeVmalKRAhC7kTACYLuG2ZWnf6Rfo+au0f13iKlmFkPYFUNup7t7rNruRwRSTGzww2KSL9HdY9OREQSTWd0IiKSaDqjExGRRFPQiYhIoinoREQk0RR0IiKSaAo6ERFJNAWdiIgkmoJOREQSTUEnIiKJpqATEZFEU9CJiEiiKehERCTRFHQiIpJoCjoREUk0BZ2IiCSagk5ERBJNQSciIommoBMRkURT0ImISKIp6EREJNEUdCIikmgKOhERSTQFnYiIJJqCTkREEk1BJyIiiaagE5HYMrOVZuapj5eraPtyqG1VH9vM7D9m9g8zm2hmTbL1Z5Lap6ATETlUG+BoYCLwD2CDmX3ZzPQzM4YaRF2AiEgE1gCPVfBaS+AzwIlAl9Sx1sBvgSvMbKK7r814hVJrFHQiUhetcPdvVdYgdfZ2BnAXMCJ1+GRgtpmNdPdVmS1RaotOw0VEyuHuxe4+291HAtcCBamXehGEXdvoqpPqUNCJiFTB3R8GxgMHUoe6A/dHVpBUi4JORCQN7v4MEL7ceYmZXRpVPZI+BZ2ISPp+DcwJ7d8ZUR1SDQo6EZE0ubsDPwodOsrMxkRVj6RHQSciUj3PAstC+1dGVYikR48XiMSAmX0R6Jnavc/dt4Reaw8MA/oAz7n7mgzXcldqM9/dfxw63gYYRTBQo5DgWbUX3H1PJuvJNnd3M5sJ3JY6NDLKeqRqOqMTiYergbtTHx0BzKy/mT0MrANmAQ8QDH3/VJkpssZW9gZm9nCo7c8qaVpSx7dT/VqZ2QPABoJZRH5OcC/rSWCzmd1RnT9oTLwa2u5kZoMjq0SqpDM6kRgys88BDwLNIq6jN/AywVlceZoCvzCzA+4+KXuVZVxemf0+5RyTHKEzOpH4uRz4K0HIOfBn4AvAUOD1LNbRiOBMsjvwK2AgwVRZAwlGIxaF2n7dzCyLtWXax2X2O0ZShaRFZ3Qi8VMypH0RcL27z6mscQY1Irhv+CN3/07o+E7gf81sFfCX1LHuwOnAa9ksMIO2ldnvEEkVkhad0YnE01bg7AhDrsRqKn6W7G/AltD+wMyXkzVlLxnvjaQKSYuCTiSebnb3TVEXATyWerbsEO5eBKwIHUrSWU/7Mvu58G8hFVDQicTPe+4+JeoiUt6u4vX9oe2GmSwky7qV2d8QSRWSFgWdSPw8H3UBIVX9gC/3bC8BzgxtHwDmRlWIVE1BJxI/GX0gvJo+ibqAiIwKbc9xd92jy2EKOpH4qavhkhPM7CiCEaQlpkVUiqRJQSciUj3hpXo+Bh6KqhBJj4JOREo0j7qAXGdm44HPhQ79Spctc5+CTkRKHB11AbnMzI4G/hQ69C7BvJ6S4xR0Isl2ILTdtKJGZnYM0Dfj1cSUmZ1PMKtLm9Sh3cDn3D0/uqokXQo6kWT7KLR9QnkNzKwrMAVI0lyUtcLMTjSzqcDTQIvU4W3AaHdfGl1lUh2a61Ik2eYAI1LbX03Nq/wqwQ/rzwCnAjcRnO19BHSNoMYo9DazH1fwWgugC3ASwd9R2FLgYnefn8nipHYp6ESS7XcEQdYeaAx8I/URth6YAIwG/ier1UWnO/DNarTfC/wU+Km7H6iqseQWBZ1Igrn75tQgiv8DjidYN62IYDLmDwguWc5w9/1mNjq6SnPOTmAtsIxgAdlp7r472pKkpqyC+VhFREQSQYNRREQk0RR0IiKSaAo6ERFJNAWdiIgkmoJOREQSTUEnIiKJpqATEZFEU9CJiEiiKehERCTRFHQiIpJoCjoREUk0BZ2IiCSagk5ERBJNQSciIommoBMRkURT0ImISKIp6EREJNEUdCIikmgKOhERSTQFnYiIJJqCTkREEk1BJyIiiaagExGRRFPQiYhIoinoREQk0RR0IiKSaAo6ERFJNAWdiIgkmoJOREQSTUEnIiKJpqATEZFEU9CJiEiiKehERCTRFHQiIpJoCjoREUk0BZ2IiCTa/wePf7PqP0uHPgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 300x185.414 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(1, 1/1.618))\n",
    "plot_timings(sleep_timings, ax=ax, id_to_x=lambda x: int(x))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "falling-council",
   "metadata": {},
   "source": [
    "### CommonNN Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "ultimate-fraction",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:07.256232Z",
     "start_time": "2021-05-31T11:09:07.252231Z"
    }
   },
   "outputs": [],
   "source": [
    "report_dir = pathlib.Path(\"reports/qcm07/cnnclustering_fit\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "weighted-williams",
   "metadata": {},
   "source": [
    "Clustering of a 2D toy data set (no structure) with increasing number of points while dimensionality, cluster parameters, and cluster recipe are fixed:\n",
    "\n",
    "  - Default recipe:\n",
    "    - $d = 2$, $r = 0.1$, $c = 0$ :  [run a](#run_a)\n",
    "    \n",
    "  - Distance recipe:\n",
    "    - $d = 2$, $r = 0.1$, $c = 0$ :  [run b](#run_a)\n",
    "    \n",
    "  - Neighbours recipe:\n",
    "    - $d = 2$, $r = 0.1$, $c = 0$ :  [run c](#run_a)\n",
    "    \n",
    "  - Neighbours recipe:\n",
    "    - $d = 2$, $r = 0.1$, $c = 0$ :  [run d](#run_a)\n",
    "    - $d = 2$, $r = 0.1$, $c = 100$ :  [run d b](#run_a_b)\n",
    "    - $d = 2$, $r = 0.3$, $c = 0$ :  [run d c](#run_a_c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "impaired-investor",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:56.616017Z",
     "start_time": "2021-05-06T14:49:56.488488Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f26f1d37a00>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x300 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# The test data\n",
    "fig, ax = plt.subplots(figsize=(1, 1))\n",
    "ax.scatter(\n",
    "    *helper.gen_no_structure_points((5000, 2)).T,\n",
    "    s=0.5,\n",
    "    \n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "brutal-arnold",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:16.742565Z",
     "start_time": "2021-05-31T11:09:16.738831Z"
    }
   },
   "outputs": [],
   "source": [
    "n_points_list = [500 * 2**x for x in range(10)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "promising-ribbon",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:17.102674Z",
     "start_time": "2021-05-31T11:09:17.097580Z"
    }
   },
   "outputs": [],
   "source": [
    "default_recipe = {\n",
    "    \"input_data\": _types.InputDataExtPointsMemoryview,\n",
    "    \"neighbours_getter\": _types.NeighboursGetterExtBruteForce,\n",
    "    \"neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"neighbour_neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"metric\": _types.MetricExtEuclideanReduced,\n",
    "    \"similarity_checker\": _types.SimilarityCheckerExtContains,\n",
    "    \"queue\": _types.QueueExtFIFOQueue,\n",
    "    \"fitter\": _fit.FitterExtBFS,\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "variable-threshold",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:17.443862Z",
     "start_time": "2021-05-31T11:09:17.438698Z"
    }
   },
   "outputs": [],
   "source": [
    "distance_recipe = {\n",
    "    \"input_data\": _types.InputDataExtPointsMemoryview,\n",
    "    \"neighbours_getter\": _types.NeighboursGetterExtBruteForce,\n",
    "    \"neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"neighbour_neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"metric\": _types.MetricExtPrecomputed,\n",
    "    \"similarity_checker\": _types.SimilarityCheckerExtContains,\n",
    "    \"queue\": _types.QueueExtFIFOQueue,\n",
    "    \"fitter\": _fit.FitterExtBFS,\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "banned-eugene",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:17.915650Z",
     "start_time": "2021-05-31T11:09:17.910607Z"
    }
   },
   "outputs": [],
   "source": [
    "neighbours_recipe = {\n",
    "    \"input_data\": _types.InputDataExtNeighboursMemoryview,\n",
    "    \"neighbours_getter\": _types.NeighboursGetterExtLookup,\n",
    "    \"neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"neighbour_neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"metric\": _types.MetricExtDummy,\n",
    "    \"similarity_checker\": _types.SimilarityCheckerExtContains,\n",
    "    \"queue\": _types.QueueExtFIFOQueue,\n",
    "    \"fitter\": _fit.FitterExtBFS,\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "covered-contractor",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:19.138868Z",
     "start_time": "2021-05-31T11:09:19.133679Z"
    }
   },
   "outputs": [],
   "source": [
    "neighbours_sorted_recipe = {\n",
    "    \"input_data\": _types.InputDataExtNeighboursMemoryview,\n",
    "    \"neighbours_getter\": _types.NeighboursGetterExtLookup,\n",
    "    \"neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"neighbour_neighbours\": (_types.NeighboursExtVector, (5000,), {}),\n",
    "    \"metric\": _types.MetricExtDummy,\n",
    "    \"similarity_checker\": _types.SimilarityCheckerExtScreensorted,\n",
    "    \"queue\": _types.QueueExtFIFOQueue,\n",
    "    \"fitter\": _fit.FitterExtBFS,\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "elect-connection",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:09:19.858500Z",
     "start_time": "2021-05-31T11:09:19.855098Z"
    }
   },
   "outputs": [],
   "source": [
    "make_maps()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "greek-symbol",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:10:05.907803Z",
     "start_time": "2021-05-31T11:10:05.888633Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name_list = [\n",
    "    'no_structure_run_a',\n",
    "    'no_structure_run_b',\n",
    "    'no_structure_run_c',\n",
    "    'no_structure_run_c_b',\n",
    "    'no_structure_run_c_c',\n",
    "    'no_structure_run_d',\n",
    "    'no_structure_run_d_b',\n",
    "    'no_structure_run_d_c',\n",
    "]\n",
    "\n",
    "for run_name in run_name_list:\n",
    "    report_file = report_dir / f\"{run_name}.json\"\n",
    "    RUN_TIMINGS_MAP[run_name] = load_report(report_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "absent-uzbekistan",
   "metadata": {},
   "source": [
    "<a id=\"run_a\"></a>\n",
    "#### Run a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "academic-intake",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:56.693660Z",
     "start_time": "2021-05-06T14:49:56.683855Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_a\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = None\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "extended-interface",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T14:49:56.707315Z",
     "start_time": "2021-05-06T14:49:56.695989Z"
    }
   },
   "outputs": [],
   "source": [
    "radius_cutoff = 0.1\n",
    "cnn_cutoff = 0\n",
    "\n",
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\"recipe\": default_recipe}\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "flying-mapping",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:18:45.663263Z",
     "start_time": "2021-05-06T14:49:56.709340Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_a\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c8022f4edb8c4f8a854a88f5007d8d72",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 16:49:56)\n",
      "Timed run ID:       1000    (06.05.2021 16:49:56)\n",
      "Timed run ID:       2000    (06.05.2021 16:49:57)\n",
      "Timed run ID:       4000    (06.05.2021 16:49:58)\n",
      "Timed run ID:       8000    (06.05.2021 16:50:01)\n",
      "Timed run ID:      16000    (06.05.2021 16:50:16)\n",
      "Timed run ID:      32000    (06.05.2021 16:51:13)\n",
      "Timed run ID:      64000    (06.05.2021 16:55:05)\n",
      "Timed run ID:     128000    (06.05.2021 17:11:19)\n",
      "Timed run ID:     256000    (06.05.2021 18:18:45)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 2.01 ms ± 259 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 7.54 ms ± 1.68 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 27.5 ms ± 9.73 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 89.4 ms ± 24.7 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 359 ms ± 98.9 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 1.43 s ± 394 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 5.72 s ± 1.57 s per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 23.2 s ± 6.31 s per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 1min 37s ± 29.4 s per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 6min 44s ± 2min 26s per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "appointed-brass",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:18:45.673173Z",
     "start_time": "2021-05-06T16:18:45.665330Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "wireless-guidance",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:18:45.761253Z",
     "start_time": "2021-05-06T16:18:45.674469Z"
    },
    "run_control": {
     "marked": true
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 3.84\n",
      "      2000: 11.38\n",
      "      4000: 45.04\n",
      "      8000: 180.10\n",
      "     16000: 720.78\n",
      "     32000: 2884.09\n",
      "     64000: 11527.52\n",
      "    128000: 46168.29\n",
      "    256000: 184473.71\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "effective-springfield",
   "metadata": {},
   "source": [
    "<a id=\"run_b\"></a>\n",
    "#### Run b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "lonely-yorkshire",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:18:45.770175Z",
     "start_time": "2021-05-06T16:18:45.763867Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_b\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = pairwise_distances\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "constitutional-istanbul",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:18:45.782433Z",
     "start_time": "2021-05-06T16:18:45.771999Z"
    }
   },
   "outputs": [],
   "source": [
    "radius_cutoff = 0.1\n",
    "cnn_cutoff = 0\n",
    "\n",
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "for n_points in [500 * 2**x for x in range(8)]:  # Memory error on 128000\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\"recipe\": distance_recipe}\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "driving-taste",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:24:16.758636Z",
     "start_time": "2021-05-06T16:18:45.784411Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_b\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df05dbf1af2148a79b1b6af19ab42b7b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/8 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 18:18:45)\n",
      "Timed run ID:       1000    (06.05.2021 18:18:45)\n",
      "Timed run ID:       2000    (06.05.2021 18:18:46)\n",
      "Timed run ID:       4000    (06.05.2021 18:18:46)\n",
      "Timed run ID:       8000    (06.05.2021 18:18:52)\n",
      "Timed run ID:      16000    (06.05.2021 18:19:06)\n",
      "Timed run ID:      32000    (06.05.2021 18:20:06)\n",
      "Timed run ID:      64000    (06.05.2021 18:24:16)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 724 µs ± 108 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 2.8 ms ± 629 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 12.7 ms ± 4.81 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 64.1 ms ± 23.3 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 454 ms ± 196 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 1.17 s ± 239 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 4.25 s ± 33.7 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 17.1 s ± 64.7 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "technological-conditions",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:24:16.762529Z",
     "start_time": "2021-05-06T16:24:16.760209Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_TIMINGS_MAP[run_name][\"128000\"] = \"MemoryError\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "otherwise-evanescence",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:24:16.780758Z",
     "start_time": "2021-05-06T16:24:16.764289Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "technical-sister",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:24:16.786885Z",
     "start_time": "2021-05-06T16:24:16.782566Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 3.60\n",
      "      2000: 11.49\n",
      "      4000: 47.19\n",
      "      8000: 367.65\n",
      "     16000: 1498.82\n",
      "     32000: 6146.07\n",
      "     64000: 24953.98\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "equivalent-ecuador",
   "metadata": {},
   "source": [
    "<a id=\"run_c\"></a>\n",
    "#### Run c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "amended-melbourne",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:24:16.798482Z",
     "start_time": "2021-05-06T16:24:16.789484Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_c\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = compute_neighbours\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "hawaiian-relative",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:24:16.812577Z",
     "start_time": "2021-05-06T16:24:16.800651Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "\n",
    "radius_cutoff = 0.1\n",
    "cnn_cutoff = 0\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"transform\": (\n",
    "                (radius_cutoff,), {}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\n",
    "                    \"preparation_hook\": cluster.prepare_neighbourhoods,\n",
    "                    \"recipe\": neighbours_recipe\n",
    "                    }\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "broadband-questionnaire",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:27:06.607873Z",
     "start_time": "2021-05-06T16:24:16.814485Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_c\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9ee0ee50049c4a4dbbb5053003d910f3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 18:24:17)\n",
      "Timed run ID:       1000    (06.05.2021 18:24:17)\n",
      "Timed run ID:       2000    (06.05.2021 18:24:17)\n",
      "Timed run ID:       4000    (06.05.2021 18:24:18)\n",
      "Timed run ID:       8000    (06.05.2021 18:24:20)\n",
      "Timed run ID:      16000    (06.05.2021 18:24:24)\n",
      "Timed run ID:      32000    (06.05.2021 18:24:32)\n",
      "Timed run ID:      64000    (06.05.2021 18:24:49)\n",
      "Timed run ID:     128000    (06.05.2021 18:25:28)\n",
      "Timed run ID:     256000    (06.05.2021 18:27:06)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 83.8 µs ± 32 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 93.1 µs ± 19.6 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 185 µs ± 92.6 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 965 µs ± 577 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 7.92 ms ± 5.55 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 13.9 ms ± 9.04 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 40.7 ms ± 8.03 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 138 ms ± 21.4 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 530 ms ± 70.4 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 2.26 s ± 259 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "naval-residence",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:27:06.616692Z",
     "start_time": "2021-05-06T16:27:06.609716Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "dependent-ticket",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:27:06.626040Z",
     "start_time": "2021-05-06T16:27:06.618310Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 1.13\n",
      "      2000: 1.45\n",
      "      4000: 2.68\n",
      "      8000: 26.48\n",
      "     16000: 81.55\n",
      "     32000: 409.02\n",
      "     64000: 1472.91\n",
      "    128000: 5848.36\n",
      "    256000: 25515.87\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "similar-humidity",
   "metadata": {},
   "source": [
    "<a id=\"run_c_b\"></a>\n",
    "#### Run c b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "impressive-reply",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:27:06.635423Z",
     "start_time": "2021-05-06T16:27:06.627622Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_c_b\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = compute_neighbours\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "representative-soviet",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:27:06.648240Z",
     "start_time": "2021-05-06T16:27:06.637298Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "\n",
    "radius_cutoff = 0.1\n",
    "cnn_cutoff = 100\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"transform\": (\n",
    "                (radius_cutoff,), {}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\n",
    "                    \"preparation_hook\": cluster.prepare_neighbourhoods,\n",
    "                    \"recipe\": neighbours_recipe\n",
    "                    }\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "familiar-bible",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:33:43.545136Z",
     "start_time": "2021-05-06T16:27:06.650320Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_c_b\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a8a25349883459d876e0174ae5c64aa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 18:27:06)\n",
      "Timed run ID:       1000    (06.05.2021 18:27:07)\n",
      "Timed run ID:       2000    (06.05.2021 18:27:07)\n",
      "Timed run ID:       4000    (06.05.2021 18:27:08)\n",
      "Timed run ID:       8000    (06.05.2021 18:27:10)\n",
      "Timed run ID:      16000    (06.05.2021 18:27:13)\n",
      "Timed run ID:      32000    (06.05.2021 18:27:21)\n",
      "Timed run ID:      64000    (06.05.2021 18:30:14)\n",
      "Timed run ID:     128000    (06.05.2021 18:31:14)\n",
      "Timed run ID:     256000    (06.05.2021 18:33:43)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 73.3 µs ± 7.55 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 82.6 µs ± 2.25 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 106 µs ± 952 ns per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 182 µs ± 3.08 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 481 µs ± 103 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 1.54 ms ± 134 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 24.3 ms ± 26.5 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 15.7 s ± 2.49 s per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 2.54 s ± 409 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 7.16 s ± 329 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "arctic-borough",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:33:43.553924Z",
     "start_time": "2021-05-06T16:33:43.547117Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "foster-quick",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:33:43.563850Z",
     "start_time": "2021-05-06T16:33:43.555443Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 1.14\n",
      "      2000: 1.48\n",
      "      4000: 2.56\n",
      "      8000: 6.30\n",
      "     16000: 21.03\n",
      "     32000: 89.47\n",
      "    128000: 29191.25\n",
      "    256000: 94533.45\n",
      "     64000: 168392.29\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cognitive-currency",
   "metadata": {},
   "source": [
    "<a id=\"run_c_c\"></a>\n",
    "#### Run c c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "offensive-shepherd",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:33:43.575793Z",
     "start_time": "2021-05-06T16:33:43.565361Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_c_c\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = compute_neighbours\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "asian-hydrogen",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:33:43.590090Z",
     "start_time": "2021-05-06T16:33:43.577929Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "\n",
    "radius_cutoff = 0.3\n",
    "cnn_cutoff = 0\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"transform\": (\n",
    "                (radius_cutoff,), {}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\n",
    "                    \"preparation_hook\": cluster.prepare_neighbourhoods,\n",
    "                    \"recipe\": neighbours_recipe\n",
    "                    }\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "conservative-panic",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:54:40.040542Z",
     "start_time": "2021-05-06T16:33:43.592223Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_c_c\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "12764808cf3d42429c45cc88c51d93c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 18:33:43)\n",
      "Timed run ID:       1000    (06.05.2021 18:33:44)\n",
      "Timed run ID:       2000    (06.05.2021 18:33:44)\n",
      "Timed run ID:       4000    (06.05.2021 18:33:45)\n",
      "Timed run ID:       8000    (06.05.2021 18:33:47)\n",
      "Timed run ID:      16000    (06.05.2021 18:33:52)\n",
      "Timed run ID:      32000    (06.05.2021 18:34:05)\n",
      "Timed run ID:      64000    (06.05.2021 18:34:46)\n",
      "Timed run ID:     128000    (06.05.2021 18:37:36)\n",
      "Timed run ID:     256000    (06.05.2021 18:54:39)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 187 µs ± 76.8 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 1.09 ms ± 855 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 1.59 ms ± 722 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 4.27 ms ± 919 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 16.7 ms ± 2.68 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 72 ms ± 9.88 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 355 ms ± 42.4 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 1.84 s ± 189 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 10.8 s ± 892 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 1min 17s ± 4.69 s per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "played-entertainment",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:54:40.049320Z",
     "start_time": "2021-05-06T16:54:40.042499Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "particular-shoulder",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:54:40.134241Z",
     "start_time": "2021-05-06T16:54:40.050870Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 3.46\n",
      "      2000: 10.44\n",
      "      4000: 36.08\n",
      "      8000: 155.47\n",
      "     16000: 704.80\n",
      "     32000: 3539.87\n",
      "     64000: 18977.51\n",
      "    128000: 115620.64\n",
      "    256000: 863166.43\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "given-martin",
   "metadata": {},
   "source": [
    "<a id=\"run_d\"></a>\n",
    "#### Run d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "boolean-burst",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:54:40.147331Z",
     "start_time": "2021-05-06T16:54:40.136771Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_d\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = compute_neighbours\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "initial-armor",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:54:40.161765Z",
     "start_time": "2021-05-06T16:54:40.149531Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "\n",
    "radius_cutoff = 0.1\n",
    "cnn_cutoff = 0\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"transform\": (\n",
    "                (radius_cutoff,), {\"sort\": True}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\n",
    "                    \"preparation_hook\": cluster.prepare_neighbourhoods,\n",
    "                    \"recipe\": neighbours_sorted_recipe\n",
    "                    }\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "charged-excuse",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:57:25.310269Z",
     "start_time": "2021-05-06T16:54:40.163363Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_d\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6a72abb851974632a1d96a4ea12b3726",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 18:54:40)\n",
      "Timed run ID:       1000    (06.05.2021 18:54:40)\n",
      "Timed run ID:       2000    (06.05.2021 18:54:41)\n",
      "Timed run ID:       4000    (06.05.2021 18:54:42)\n",
      "Timed run ID:       8000    (06.05.2021 18:54:44)\n",
      "Timed run ID:      16000    (06.05.2021 18:54:48)\n",
      "Timed run ID:      32000    (06.05.2021 18:54:55)\n",
      "Timed run ID:      64000    (06.05.2021 18:55:13)\n",
      "Timed run ID:     128000    (06.05.2021 18:55:51)\n",
      "Timed run ID:     256000    (06.05.2021 18:57:25)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 81.9 µs ± 25.6 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 94.6 µs ± 17.3 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 208 µs ± 126 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 774 µs ± 427 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 5.18 ms ± 3 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 9.47 ms ± 3.72 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 30.8 ms ± 3.11 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 102 ms ± 6.33 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 365 ms ± 12.7 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 1.27 s ± 24.4 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "ready-irish",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:57:25.316421Z",
     "start_time": "2021-05-06T16:57:25.312027Z"
    }
   },
   "outputs": [],
   "source": [
    "previous_timings = load_report(report_file)\n",
    "for run_id, timing in previous_timings.items():\n",
    "    if run_id in RUN_TIMINGS_MAP[run_name]:\n",
    "        combined = combine_timeit_results(\n",
    "            RUN_TIMINGS_MAP[run_name][run_id], timing\n",
    "        )\n",
    "    else:\n",
    "        combined = timing\n",
    "        \n",
    "    RUN_TIMINGS_MAP[run_name][run_id] = combined"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "french-optimization",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:57:25.330321Z",
     "start_time": "2021-05-06T16:57:25.317765Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "normal-marina",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:57:25.338809Z",
     "start_time": "2021-05-06T16:57:25.331633Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 1.14\n",
      "      2000: 1.50\n",
      "      4000: 2.67\n",
      "      8000: 22.84\n",
      "     16000: 80.77\n",
      "     32000: 359.05\n",
      "     64000: 1280.41\n",
      "    128000: 4773.54\n",
      "    256000: 17251.74\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "received-secret",
   "metadata": {},
   "source": [
    "<a id=\"run_d_b\"></a>\n",
    "#### Run d b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "exterior-writer",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:57:25.349689Z",
     "start_time": "2021-05-06T16:57:25.343830Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_d_b\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = compute_neighbours\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "atlantic-surrey",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T16:57:25.362300Z",
     "start_time": "2021-05-06T16:57:25.351461Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "\n",
    "radius_cutoff = 0.1\n",
    "cnn_cutoff = 100\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"transform\": (\n",
    "                (radius_cutoff,), {\"sort\": True}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\n",
    "                    \"preparation_hook\": cluster.prepare_neighbourhoods,\n",
    "                    \"recipe\": neighbours_sorted_recipe\n",
    "                    }\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "adequate-detective",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:00:43.329634Z",
     "start_time": "2021-05-06T16:57:25.364015Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_d_b\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c99111f4c046439191202ccc6de50c09",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 18:57:25)\n",
      "Timed run ID:       1000    (06.05.2021 18:57:25)\n",
      "Timed run ID:       2000    (06.05.2021 18:57:26)\n",
      "Timed run ID:       4000    (06.05.2021 18:57:27)\n",
      "Timed run ID:       8000    (06.05.2021 18:57:28)\n",
      "Timed run ID:      16000    (06.05.2021 18:57:32)\n",
      "Timed run ID:      32000    (06.05.2021 18:57:40)\n",
      "Timed run ID:      64000    (06.05.2021 18:58:22)\n",
      "Timed run ID:     128000    (06.05.2021 18:59:04)\n",
      "Timed run ID:     256000    (06.05.2021 19:00:43)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 78.8 µs ± 9.26 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 88 µs ± 11 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 109 µs ± 4.72 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 185 µs ± 3.57 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 452 µs ± 9.96 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 1.5 ms ± 14.5 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 11.1 ms ± 6.44 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 2.65 s ± 443 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 625 ms ± 37.4 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 1.71 s ± 26.1 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "rotary-bolivia",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:00:43.338418Z",
     "start_time": "2021-05-06T17:00:43.331546Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "sought-vacation",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:00:43.348244Z",
     "start_time": "2021-05-06T17:00:43.339672Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 1.10\n",
      "      2000: 1.43\n",
      "      4000: 2.43\n",
      "      8000: 5.96\n",
      "     16000: 19.97\n",
      "     32000: 87.27\n",
      "    128000: 7739.12\n",
      "    256000: 22501.85\n",
      "     64000: 26586.38\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "uniform-element",
   "metadata": {},
   "source": [
    "<a id=\"run_d_c\"></a>\n",
    "#### Run d c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "valuable-seafood",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:00:43.359492Z",
     "start_time": "2021-05-06T17:00:43.349701Z"
    }
   },
   "outputs": [],
   "source": [
    "run_name = \"no_structure_run_d_c\"\n",
    "report_file = report_dir / f\"{run_name}.json\"\n",
    "\n",
    "gen_func = helper.gen_no_structure_points\n",
    "transform_func = compute_neighbours\n",
    "setup_func = setup_commonnn_clustering__fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "precise-eating",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:00:43.373816Z",
     "start_time": "2021-05-06T17:00:43.361703Z"
    }
   },
   "outputs": [],
   "source": [
    "RUN_ARGUMENTS_MAP[run_name] = []\n",
    "\n",
    "radius_cutoff = 0.3\n",
    "cnn_cutoff = 0\n",
    "for n_points in n_points_list:\n",
    "    RUN_ARGUMENTS_MAP[run_name].append(\n",
    "        {\n",
    "            \"id\": str(n_points),\n",
    "            \"gen\": (\n",
    "                ((n_points, 2),), {}\n",
    "            ),\n",
    "            \"transform\": (\n",
    "                (radius_cutoff,), {\"sort\": True}\n",
    "            ),\n",
    "            \"setup\": (\n",
    "                (), {\n",
    "                    \"preparation_hook\": cluster.prepare_neighbourhoods,\n",
    "                    \"recipe\": neighbours_sorted_recipe\n",
    "                    }\n",
    "            ),\n",
    "            \"timed\": (\n",
    "                (_types.ClusterParameters(radius_cutoff, cnn_cutoff),), {}\n",
    "            ),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "fifth-census",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:08:48.274343Z",
     "start_time": "2021-05-06T17:00:43.375951Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collection for run: no_structure_run_d_c\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "30ebf17add2046809b3e6d2e78b5a881",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Run completed:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Timed run ID:        500    (06.05.2021 19:00:43)\n",
      "Timed run ID:       1000    (06.05.2021 19:00:43)\n",
      "Timed run ID:       2000    (06.05.2021 19:00:44)\n",
      "Timed run ID:       4000    (06.05.2021 19:00:45)\n",
      "Timed run ID:       8000    (06.05.2021 19:00:47)\n",
      "Timed run ID:      16000    (06.05.2021 19:00:52)\n",
      "Timed run ID:      32000    (06.05.2021 19:01:03)\n",
      "Timed run ID:      64000    (06.05.2021 19:01:34)\n",
      "Timed run ID:     128000    (06.05.2021 19:03:09)\n",
      "Timed run ID:     256000    (06.05.2021 19:08:48)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'500': <TimeitResult : 163 µs ± 58.9 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '1000': <TimeitResult : 747 µs ± 484 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '2000': <TimeitResult : 1.08 ms ± 309 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '4000': <TimeitResult : 2.97 ms ± 398 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '8000': <TimeitResult : 10.1 ms ± 666 µs per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '16000': <TimeitResult : 39.4 ms ± 1.34 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '32000': <TimeitResult : 150 ms ± 3.93 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '64000': <TimeitResult : 566 ms ± 6.94 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '128000': <TimeitResult : 2.2 s ± 10.9 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>,\n",
       " '256000': <TimeitResult : 8.99 s ± 26.7 ms per loop (mean ± std. dev. of 10 runs, 1 loop each)>}"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Collection for run: {run_name}\")\n",
    "RUN_TIMINGS_MAP[run_name] = {}\n",
    "collect_timings(\n",
    "    gen_func,\n",
    "    setup_func,\n",
    "    RUN_ARGUMENTS_MAP[run_name],\n",
    "    transform_func=transform_func,\n",
    "    timings=RUN_TIMINGS_MAP[run_name]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "smoking-oxygen",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:08:48.284424Z",
     "start_time": "2021-05-06T17:08:48.276160Z"
    }
   },
   "outputs": [],
   "source": [
    "save_report(RUN_TIMINGS_MAP[run_name], report_file, overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "level-extension",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-06T17:08:48.291530Z",
     "start_time": "2021-05-06T17:08:48.285914Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Run ID: Factor\n",
      "=======================\n",
      "       500: 1.00\n",
      "      1000: 3.58\n",
      "      2000: 8.80\n",
      "      4000: 30.28\n",
      "      8000: 113.11\n",
      "     16000: 464.12\n",
      "     32000: 1816.65\n",
      "     64000: 6954.94\n",
      "    128000: 27366.46\n",
      "    256000: 112138.82\n"
     ]
    }
   ],
   "source": [
    "print_ratios(get_ratios(RUN_TIMINGS_MAP[run_name], which=\"best\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dirty-confirmation",
   "metadata": {},
   "source": [
    "#### Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "dried-holly",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:11:10.464147Z",
     "start_time": "2021-05-31T11:11:10.459659Z"
    }
   },
   "outputs": [],
   "source": [
    "figsrc_dir = pathlib.Path(os.path.expandvars(\"$WD/CommonNN/Manuscript/figsrc/Benchmark\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "played-terminology",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:55:23.295587Z",
     "start_time": "2021-05-31T11:55:22.376125Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 999x617.429 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mpl.rcParams[\"font.size\"] = 8\n",
    "mpl.rcParams[\"axes.labelpad\"] = 2\n",
    "mpl.rcParams[\"xtick.major.pad\"] = 2\n",
    "mpl.rcParams[\"xtick.minor.pad\"] = 1.9\n",
    "mpl.rcParams[\"ytick.major.pad\"] = 2\n",
    "mpl.rcParams[\"ytick.minor.pad\"] = 1.9\n",
    "\n",
    "fig, (normal_ax, log_ax) = plt.subplots(2, 1, figsize=(3.33, 3.33/1.618))\n",
    "\n",
    "run_name_list = [\n",
    "    ('no_structure_run_a', 'points'),\n",
    "    ('no_structure_run_b', 'distances'),\n",
    "    ('no_structure_run_c', 'neighbours'),\n",
    "    ('no_structure_run_d', 'sorted n.'),\n",
    "]\n",
    "markers = iter([\"o\", \"v\", \"^\", \"s\", \"p\", \"P\", \"*\", \"h\", \"d\"])\n",
    "\n",
    "normal_ax.set_position([0.13, 0.51, 0.84, 0.40])\n",
    "log_ax.set_position([0.13, 0.09, 0.84, 0.40])\n",
    "\n",
    "# Inset\n",
    "inset_ax = fig.add_axes(\n",
    "    [0.125, 0.66, 0.39, 0.26],\n",
    "    zorder=3,\n",
    "    )\n",
    "\n",
    "for name, label in run_name_list:\n",
    "    marker = next(markers)\n",
    "    \n",
    "    plot_timings(\n",
    "        RUN_TIMINGS_MAP[name],\n",
    "        ax=normal_ax,\n",
    "        id_to_x=lambda x: int(x),\n",
    "        sort_ids=True,\n",
    "        set_ax_props=False,\n",
    "        plot_props={\n",
    "            \"label\": label,\n",
    "            \"marker\": marker,\n",
    "            \"markersize\": 4\n",
    "        }\n",
    "    )\n",
    "    \n",
    "    line = normal_ax.lines[-1]\n",
    "    log_line, = log_ax.plot(*line.get_xydata().T)\n",
    "    saved_transform = log_line.get_transform()\n",
    "    saved_clipbox = log_line.clipbox\n",
    "    log_line.update_from(line)\n",
    "    log_line.set_transform(saved_transform)\n",
    "    log_line.clipbox = saved_clipbox\n",
    "    \n",
    "    line = normal_ax.lines[-1]\n",
    "    inset_line, = inset_ax.plot(*line.get_xydata().T)\n",
    "    saved_transform = inset_line.get_transform()\n",
    "    saved_clipbox = inset_line.clipbox\n",
    "    inset_line.update_from(line)\n",
    "    inset_line.set_transform(saved_transform)\n",
    "    inset_line.clipbox = saved_clipbox\n",
    "    \n",
    "    inset_line.set_markersize(3)\n",
    "    \n",
    "normal_ax.xaxis.tick_top()\n",
    "\n",
    "normal_ax.set(**{\n",
    "    \"xlabel\": None,\n",
    "    \"xlabel\": None\n",
    "})\n",
    "\n",
    "log_ax.set(**{\n",
    "    \"xlabel\": None,\n",
    "    \"xlabel\": None,\n",
    "    \"xscale\": \"log\",\n",
    "    \"yscale\": \"log\",\n",
    "})\n",
    "\n",
    "inset_ax.set(**{\n",
    "    \"xlim\": (400, 9000),\n",
    "    \"ylim\": (-0.0005, 0.01),\n",
    "    \"xticks\": (),\n",
    "    \"yticks\": (),\n",
    "})\n",
    "\n",
    "mark_inset(normal_ax, inset_ax, loc1=3, loc2=4, edgecolor=\"k\")\n",
    "\n",
    "# Annotate memory error\n",
    "annotate_memory_error(\n",
    "    ax=normal_ax, line=normal_ax.get_lines()[1],\n",
    "    memory_error_id=128000\n",
    ")\n",
    "annotate_memory_error(\n",
    "    ax=log_ax, line=log_ax.get_lines()[1],\n",
    "    memory_error_id=128000\n",
    ")\n",
    "    \n",
    "commonax = fig.add_axes(\n",
    "    [0.04, 0., 0.97, 1],\n",
    "    zorder=-1,\n",
    "    frame_on=False\n",
    "    )\n",
    "\n",
    "commonax.set(**{\n",
    "    \"xticks\": (),\n",
    "    \"yticks\": ()\n",
    "})\n",
    "commonax.set_ylabel(\"time / s\", labelpad=0)\n",
    "\n",
    "# Legend\n",
    "legend = normal_ax.legend(\n",
    "    fancybox=False,\n",
    "    framealpha=1,\n",
    "    edgecolor=\"k\",\n",
    "    fontsize=\"xx-small\",\n",
    "    loc=(0.80, 0.34)\n",
    "    )\n",
    "legend.get_frame().set_linewidth(0.5)\n",
    "\n",
    "log_ax.get_xaxis().set_major_formatter(mpl.ticker.ScalarFormatter())\n",
    "\n",
    "# fig.subplots_adjust(left=0.11, bottom=0.1, right=0.86, top=0.89, hspace=0.1)\n",
    "fig.savefig(figsrc_dir / \"bm_cnnclustering_fit_no_structure_a_b_c_d.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "defensive-state",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-31T11:55:19.436944Z",
     "start_time": "2021-05-31T11:55:18.817996Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 999x617.429 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mpl.rcParams[\"font.size\"] = 8\n",
    "mpl.rcParams[\"axes.labelpad\"] = 2\n",
    "mpl.rcParams[\"xtick.major.pad\"] = 2\n",
    "mpl.rcParams[\"xtick.minor.pad\"] = 1.9\n",
    "mpl.rcParams[\"ytick.major.pad\"] = 2\n",
    "mpl.rcParams[\"ytick.minor.pad\"] = 1.9\n",
    "\n",
    "fig, (normal_ax, log_ax) = plt.subplots(2, 1, figsize=(3.33, 3.33/1.618))\n",
    "\n",
    "run_name_list = [\n",
    "    ('no_structure_run_c', '0.1, 0'),\n",
    "    ('no_structure_run_c_b', '0.1, 100'),\n",
    "    ('no_structure_run_c_c', '0.3, 0'),\n",
    "    ('no_structure_run_d', '0.1, 0, sorted'),\n",
    "    ('no_structure_run_d_b', '0.1, 100, sorted'),\n",
    "    ('no_structure_run_d_c', '0.3, 0, sorted'),\n",
    "]\n",
    "markers = iter([\"o\", \"v\", \"^\", \"s\", \"p\", \"P\", \"*\", \"h\", \"d\"])\n",
    "\n",
    "normal_ax.set_position([0.13, 0.51, 0.84, 0.40])\n",
    "log_ax.set_position([0.13, 0.09, 0.84, 0.40])\n",
    "\n",
    "# Inset\n",
    "inset_ax = fig.add_axes(\n",
    "    [0.125, 0.66, 0.39, 0.26],\n",
    "    zorder=3,\n",
    "    )\n",
    "\n",
    "for name, label in run_name_list:\n",
    "    marker = next(markers)\n",
    "    \n",
    "    plot_timings(\n",
    "        RUN_TIMINGS_MAP[name],\n",
    "        ax=normal_ax,\n",
    "        id_to_x=lambda x: int(x),\n",
    "        sort_ids=True,\n",
    "        set_ax_props=False,\n",
    "        plot_props={\n",
    "            \"label\": label,\n",
    "            \"marker\": marker,\n",
    "            \"markersize\": 4\n",
    "        }\n",
    "    )\n",
    "    \n",
    "    line = normal_ax.lines[-1]\n",
    "    log_line, = log_ax.plot(*line.get_xydata().T)\n",
    "    saved_transform = log_line.get_transform()\n",
    "    saved_clipbox = log_line.clipbox\n",
    "    log_line.update_from(line)\n",
    "    log_line.set_transform(saved_transform)\n",
    "    log_line.clipbox = saved_clipbox\n",
    "    \n",
    "    line = normal_ax.lines[-1]\n",
    "    inset_line, = inset_ax.plot(*line.get_xydata().T)\n",
    "    saved_transform = inset_line.get_transform()\n",
    "    saved_clipbox = inset_line.clipbox\n",
    "    inset_line.update_from(line)\n",
    "    inset_line.set_transform(saved_transform)\n",
    "    inset_line.clipbox = saved_clipbox\n",
    "    \n",
    "    inset_line.set_markersize(3)\n",
    "    \n",
    "normal_ax.xaxis.tick_top()\n",
    "\n",
    "normal_ax.set(**{\n",
    "    \"xlabel\": None,\n",
    "    \"xlabel\": None\n",
    "})\n",
    "\n",
    "log_ax.set(**{\n",
    "    \"xlabel\": None,\n",
    "    \"xlabel\": None,\n",
    "    \"xscale\": \"log\",\n",
    "    \"yscale\": \"log\",\n",
    "})\n",
    "\n",
    "inset_ax.set(**{\n",
    "    \"xlim\": (400, 9000),\n",
    "    \"ylim\": (-0.0005, 0.005),\n",
    "    \"xticks\": (),\n",
    "    \"yticks\": (),\n",
    "})\n",
    "\n",
    "mark_inset(normal_ax, inset_ax, loc1=3, loc2=4, edgecolor=\"k\")\n",
    "    \n",
    "commonax = fig.add_axes(\n",
    "    [0.04, 0., 0.97, 1],\n",
    "    zorder=-1,\n",
    "    frame_on=False\n",
    "    )\n",
    "\n",
    "commonax.set(**{\n",
    "    \"xticks\": (),\n",
    "    \"yticks\": ()\n",
    "})\n",
    "commonax.set_ylabel(\"time / s\", labelpad=0)\n",
    "\n",
    "# Legend\n",
    "legend = normal_ax.legend(\n",
    "    fancybox=False,\n",
    "    framealpha=1,\n",
    "    edgecolor=\"k\",\n",
    "    fontsize=\"xx-small\",\n",
    "    loc=(0.8, 0.33)\n",
    "    )\n",
    "legend.get_frame().set_linewidth(0.5)\n",
    "\n",
    "log_ax.get_xaxis().set_major_formatter(mpl.ticker.ScalarFormatter())\n",
    "\n",
    "# fig.subplots_adjust(left=0.11, bottom=0.1, right=0.86, top=0.89, hspace=0.1)\n",
    "fig.savefig(figsrc_dir / \"bm_cnnclustering_fit_no_structure_c_cb_cc_d_db_dc.png\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "md38",
   "language": "python",
   "name": "md38"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "263.3px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
