bar_plot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f7888d75",
   "metadata": {},
   "outputs": [],
   "source": [
    "import argparse\n",
    "import json\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import csv\n",
    "\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "colors = {\"tblis\" : \"olive\", \"gsl\" : \"green\", \"blas\" : \"red\", \"taco\" : \"blue\", \"gsl_tensor\" : \"cyan\", \\\n",
    "          \"dot_blas\" : \"gold\", \"dot_gsl\" : \"grey\", \"gemv_blas\" : \"purple\", \"gemv_gsl\" : \"pink\", \"mkl\" :\"black\", \"cuda\":\"green\",\\\n",
    "         \"dot_mkl\" : \"silver\", \"gemv_mkl\" : \"yellow\", \"stardust\":\"orange\",\n",
    "         \"taco_csr\" : \"green\", \"taco_coo\" : \"red\", \"row\" : \"olive\", \"col\" : \"green\", \"block\" : \"red\", \n",
    "         \"block_diagonal\" : \"blue\", \"random\" : \"cyan\", \"bug_mkl\":\"purple\"}\n",
    "\n",
    "markers = {\"tblis\" : \"o\", \"gsl\" : \"p\", \"blas\" : \"*\", \"taco\" : \".\", \"gsl_tensor\" : \".\", \\\n",
    "          \"dot_blas\" : \".\", \"dot_gsl\" : \".\", \"gemv_blas\" : \".\", \"gemv_gsl\" : \"p\", \"mkl\" :\"v\", \"cuda\":\"s\",\\\n",
    "         \"dot_mkl\" : \".\", \"gemv_mkl\" : \".\", \"stardust\":\"1\", \"taco_coo\" : \"p\", \"taco_csr\" : \".\", \\\n",
    "          \"row\" : \"o\", \"col\" : \"p\", \"random\" : \"*\", \"block\" : \".\", \"block_diagonal\" : \".\", \"bug_mkl\": \".\"}\n",
    "                      \n",
    "linestyles = {\"tblis\" : \"-\", \"gsl\" : \"-\", \"blas\" : \"-\", \"taco\" : \"-\", \"gsl_tensor\" : \"-\", \\\n",
    "          \"dot_blas\" : \"-\", \"dot_gsl\" : \"-\", \"gemv_blas\" : \"-\", \"gemv_gsl\" : \"-\", \"mkl\" :\"--\", \"cuda\":\"-\",\\\n",
    "         \"dot_mkl\" : \"-\", \"gemv_mkl\" : \"-\", \"stardust\":\"-\", \"taco_coo\" : \"-\", \"taco_csr\" : \"--\",\\\n",
    "           \"row\" : \"-\", \"col\" : \"-\", \"block\" : \"-\", \"block_diagonal\" : \"-\", \"random\" : \"-\", \"bug_mkl\":\"-\" }\n",
    "\n",
    "hatches = {\"mkl\" : \"/\", \"taco\" : \"o\", \"blas\" : \"*\"}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8ea0aabf",
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_bar_plots(name, directory, systems, matrices, expr, unit=\"ms\"):\n",
    "    all_dfs = None\n",
    "    for matrix in matrices:\n",
    "        result = None\n",
    "        for system in systems:\n",
    "            data = json.load(open(f'{directory}/{matrix}/{system}'))\n",
    "            df = pd.DataFrame(data[\"benchmarks\"])\n",
    "            df = df[df['aggregate_name'] == \"median\"]['real_time']\n",
    "            df = df.rename_axis('Label').reset_index()\n",
    "            df.rename(columns = {'real_time': f'{system}'}, inplace = True)\n",
    "           \n",
    "\n",
    "            if result is None:\n",
    "                result = df\n",
    "            else:\n",
    "                result = pd.merge(result, df, on='Label', how='outer')\n",
    "        result['SuiteSparseMatrix'] = matrix\n",
    "        \n",
    "        if all_dfs is None:\n",
    "            all_dfs = result\n",
    "        else:\n",
    "            all_dfs = pd.concat([all_dfs, result])\n",
    "\n",
    "    all_dfs = all_dfs.drop(['Label'], axis=1)\n",
    "    \n",
    "    full_plt = all_dfs.plot.bar(x='SuiteSparseMatrix', logy=True, label=systems)\n",
    "    \n",
    "    plt.ylabel(f'Runtime ({unit})')\n",
    "    plt.xlabel('SuiteSparse Matrix Name')\n",
    "    \n",
    "    for system in systems:\n",
    "        if system == \"taco\":\n",
    "            continue\n",
    "        all_dfs[system] = all_dfs[\"taco\"]/all_dfs[system]\n",
    "    \n",
    "    print(all_dfs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "38c47091",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          taco       mkl      blas SuiteSparseMatrix\n",
      "0     0.006613  0.353353  0.540896             Trec5\n",
      "0     0.007034  0.327080  0.437292          Ragusa18\n",
      "0     0.006763  0.311285  0.412813        lpi_bgprtr\n",
      "0     0.012566  0.276119  0.312097          lp_sc50b\n",
      "0     1.177303  1.331247  1.678987          cavity02\n",
      "0     1.197664  1.332807  1.528687          cavity03\n",
      "0    12.089265  0.539295  0.944674          lp_nug08\n",
      "0  1281.862909  0.063495  0.124263          m3plates\n",
      "0   816.734070  3.569406  6.478009            IG5-12\n",
      "0  2480.098870  0.805035  1.496346        g7jac020sc\n",
      "0  2513.326384  0.638913  1.186281            gemat1\n",
      "0  9987.979106  0.292063  0.578269  mimo28x28_system\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_bar_plots(\"Suitesparse\", \"./data-benchmarks/suitesparse/\", [\"taco\", \"mkl\", \"blas\"],\\\n",
    "                   [\"Trec5\",\"Ragusa18\", \"lpi_bgprtr\", \"lp_sc50b\",\\\n",
    "                    \"cavity02\", \"cavity03\", \"lp_nug08\", \"m3plates\",\\\n",
    "                    \"IG5-12\", \"g7jac020sc\", \"gemat1\", \"mimo28x28_system\"], \"SDDMM\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c9c9fc4a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cf0e966f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}