numpy/aggregations.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SUM, MIN, MAX"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "129 µs ± 603 ns per loop (mean ± std. dev. of 7 runs, 10000 loops each)\n",
      "2.44 µs ± 230 ns per loop (mean ± std. dev. of 7 runs, 100000 loops each)\n"
     ]
    }
   ],
   "source": [
    "big_array = np.random.rand(1000)\n",
    "%timeit sum(big_array)\n",
    "%timeit np.sum(big_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.00098561856427836769, 0.99947556031324558)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.min(big_array), np.max(big_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "48.8 µs ± 65 ns per loop (mean ± std. dev. of 7 runs, 10000 loops each)\n",
      "2.29 µs ± 6.81 ns per loop (mean ± std. dev. of 7 runs, 100000 loops each)\n"
     ]
    }
   ],
   "source": [
    "%timeit min(big_array)\n",
    "%timeit np.min(big_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.000985618564278 0.999475560313 497.102229897\n"
     ]
    }
   ],
   "source": [
    "print(big_array.min(), big_array.max(), big_array.sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Многомерные сводные показатели"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.78234112  0.21550833  0.62292744  0.53012043]\n",
      " [ 0.2525845   0.12572385  0.05469664  0.38775828]\n",
      " [ 0.83382787  0.22656239  0.98431005  0.21766844]]\n"
     ]
    }
   ],
   "source": [
    "M = np.random.random((3, 4))\n",
    "print(M)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.2525845 ,  0.12572385,  0.05469664,  0.21766844])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# минимальное значение каждого из столбцов\n",
    "M.min(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.78234112,  0.38775828,  0.98431005])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# максимальное значение в каждом из строк\n",
    "M.max(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "np.mean - среднее  \n",
    "np.std - стандартное отклонение  \n",
    "np.var - дисперсия  \n",
    "np.median - медиана  \n",
    "np.percentile - квантили  \n",
    "np.any - существуют ли элементы со значением true  \n",
    "np.all - все ли элементы имеют значение true  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}