{"cells":[
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "%matplotlib inline"
 },
 {
  "cell_type": "code",
  "execution_count": null,
  "metadata": {
   "collapsed": false
  },
  "outputs": [],
  "source": "# Sample script naive benchmark that yields 0.609 public LB score WITHOUT any image information\n\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\n\n# Read in data, files are assumed to be in the \"../input/\" directory.\ntrain = pd.read_csv('../input/train.csv')\nsubmit = pd.read_csv('../input/sample_submission.csv')\n\n# convert numeric labels to binary matrix\ndef to_bool(s):\n    return(pd.Series([1 if str(i) in str(s).split(' ') else 0 for i in range(9)]))\nY = train['labels'].apply(to_bool)\n\n# get means proportion of each class\npy = Y.mean()\nplt.bar(Y.columns,py,color='steelblue',edgecolor='white')\n\n# predict classes that are > 0.5, 2,3,5,6,8\n# try using only six labels instead of five\nsubmit['labels'] = '1 2 3 5 6 8'\nsubmit.to_csv('naive.csv',index=False)"
 }
],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"}}, "nbformat": 4, "nbformat_minor": 0}