{"cells":[
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "import pandas as pd\nimport numpy as np\nimport time as t\nimport math\ncpu_start = t.clock()\nreal_start = t.time()\n\ndf = pd.read_csv('../input/train.csv', nrows=3500000)\ndfp = pd.read_csv('../input/sample_submission.csv', nrows=3500000)\nn = dfp.shape[0]\n\na=[] #actual demand\nfor line in df.itertuples():\n    a.append(line.Demanda_uni_equil)\n    \np=[] #predicted demans\nfor line in dfp.itertuples():\n    p.append(line.Demanda_uni_equil)\n    \nx=0\nfor i in range(n):\n    x += (math.pow((math.log1p(p[i] + 1) - math.log1p(a[i]+1)), 2))\n\ne = math.sqrt((1/n)*(x))\n\nprint (e)\n                    \ncpu_time = t.clock()-cpu_start\nreal_time = t.time()-real_start\nprint('CPU: %s, Valós:%s' %(cpu_time, real_time))"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": ""
 }
],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"}}, "nbformat": 4, "nbformat_minor": 0}