pumpp - h5py vs npz.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pumpp\n",
    "import h5py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "C = pumpp.feature.CQT(name='cqt', sr=44100, hop_length=4096, log=True, conv='tf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "B = pumpp.task.BeatTransformer(sr=44100, hop_length=4096)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "P = pumpp.Pump(C, B)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "datum = P('/home/bmcfee/data/MedleyDB_sample/Audio/LizNelson_Rainfall/LizNelson_Rainfall_MIX.wav')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'beat/_valid': array([[0, 0]]), 'beat/beat': array([[[False],\n",
       "         [False],\n",
       "         [False],\n",
       "         ..., \n",
       "         [False],\n",
       "         [False],\n",
       "         [False]]], dtype=bool), 'beat/downbeat': array([[[False],\n",
       "         [False],\n",
       "         [False],\n",
       "         ..., \n",
       "         [False],\n",
       "         [False],\n",
       "         [False]]], dtype=bool), 'beat/mask_downbeat': array([False], dtype=bool), 'cqt/mag': array([[[[-60.45588684],\n",
       "          [-56.91991043],\n",
       "          [-64.60596466],\n",
       "          ..., \n",
       "          [-80.        ],\n",
       "          [-80.        ],\n",
       "          [-80.        ]],\n",
       " \n",
       "         [[-60.5687294 ],\n",
       "          [-57.38772202],\n",
       "          [-61.41232681],\n",
       "          ..., \n",
       "          [-80.        ],\n",
       "          [-80.        ],\n",
       "          [-80.        ]],\n",
       " \n",
       "         [[-61.04840088],\n",
       "          [-58.86682129],\n",
       "          [-57.50798798],\n",
       "          ..., \n",
       "          [-75.88668823],\n",
       "          [-74.88679504],\n",
       "          [-70.69407654]],\n",
       " \n",
       "         ..., \n",
       "         [[-65.34263611],\n",
       "          [-60.15421295],\n",
       "          [-59.2531662 ],\n",
       "          ..., \n",
       "          [-80.        ],\n",
       "          [-80.        ],\n",
       "          [-80.        ]],\n",
       " \n",
       "         [[-65.20644379],\n",
       "          [-61.90522385],\n",
       "          [-57.836483  ],\n",
       "          ..., \n",
       "          [-80.        ],\n",
       "          [-80.        ],\n",
       "          [-80.        ]],\n",
       " \n",
       "         [[-65.43597412],\n",
       "          [-64.44393921],\n",
       "          [-57.01498795],\n",
       "          ..., \n",
       "          [-80.        ],\n",
       "          [-80.        ],\n",
       "          [-80.        ]]]], dtype=float32), 'cqt/phase': array([[[[  1.40390837e+00],\n",
       "          [  1.50876081e+00],\n",
       "          [  1.53741777e+00],\n",
       "          ..., \n",
       "          [  1.91018534e+00],\n",
       "          [  1.60638404e+00],\n",
       "          [ -5.47881901e-01]],\n",
       " \n",
       "         [[  1.21011412e+00],\n",
       "          [  7.63346970e-01],\n",
       "          [  1.21743667e+00],\n",
       "          ..., \n",
       "          [  9.93008493e-04],\n",
       "          [ -7.67094553e-01],\n",
       "          [  5.39158344e-01]],\n",
       " \n",
       "         [[  9.93348539e-01],\n",
       "          [ -1.74790490e-02],\n",
       "          [  3.73399168e-01],\n",
       "          ..., \n",
       "          [ -5.14139295e-01],\n",
       "          [  1.18385446e+00],\n",
       "          [  2.87618685e+00]],\n",
       " \n",
       "         ..., \n",
       "         [[  1.70295131e+00],\n",
       "          [  1.21126425e+00],\n",
       "          [  1.33812416e+00],\n",
       "          ..., \n",
       "          [  2.14315009e+00],\n",
       "          [  1.38004375e+00],\n",
       "          [ -2.85812318e-01]],\n",
       " \n",
       "         [[  1.37173688e+00],\n",
       "          [  4.30214912e-01],\n",
       "          [  3.83782387e-01],\n",
       "          ..., \n",
       "          [  2.80652738e+00],\n",
       "          [  2.77391362e+00],\n",
       "          [ -3.04661632e+00]],\n",
       " \n",
       "         [[  1.11487710e+00],\n",
       "          [ -4.43137944e-01],\n",
       "          [ -6.32018566e-01],\n",
       "          ..., \n",
       "          [ -2.12892151e+00],\n",
       "          [ -1.42968285e+00],\n",
       "          [ -1.63801491e-01]]]], dtype=float32)}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "datum"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "with h5py.File('/tmp/arr.h5', mode='w') as hf:\n",
    "    hf.update(datum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.savez('/tmp/arr.npz', **datum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.13 ms ± 42.6 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit \n",
    "with h5py.File('/tmp/arr.h5', mode='r', ) as hf:\n",
    "    cqt = hf['cqt/phase'].value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.17 ms ± 40.9 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit \n",
    "with np.load('/tmp/arr.npz') as nf:\n",
    "    cqt = nf['cqt/phase']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-rw-rw-r-- 1 bmcfee bmcfee 7078759 Apr 28 16:35 /tmp/arr.h5\r\n",
      "-rw-rw-r-- 1 bmcfee bmcfee 7073691 Apr 28 16:35 /tmp/arr.npz\r\n"
     ]
    }
   ],
   "source": [
    "!ls -l /tmp/arr.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import crema.utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = crema.utils.load_h5('/tmp/arr.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.26 ms ± 59.5 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "data = crema.utils.load_h5('/tmp/arr.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13.2 ms ± 31.6 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "data = dict(np.load('/tmp/arr.npz'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data1 = dict(np.load('/tmp/arr.npz'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data2 = crema.utils.load_h5('/tmp/arr.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1.keys() == data2.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all([np.allclose(data1[k], data2[k]) for k in data1])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.5",
   "language": "python",
   "name": "python3.5"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}