Mercurial > hg > plosone_underreview

annotate notebooks/test_music_segments.ipynb @ 6:a35bd818d8e9 branch-tests

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
notebook to test music segments
author Maria Panteli <m.x.panteli@gmail.com>
date Mon, 11 Sep 2017 14:22:17 +0100
parents
children 46b2c713cc73
rev   line source
m@6 1 {
m@6 2 "cells": [
m@6 3 {
m@6 4 "cell_type": "code",
m@6 5 "execution_count": 1,
m@6 6 "metadata": {
m@6 7 "collapsed": true
m@6 8 },
m@6 9 "outputs": [],
m@6 10 "source": [
m@6 11 "import numpy as np\n",
m@6 12 "import pickle"
m@6 13 ]
m@6 14 },
m@6 15 {
m@6 16 "cell_type": "code",
m@6 17 "execution_count": 2,
m@6 18 "metadata": {
m@6 19 "collapsed": true
m@6 20 },
m@6 21 "outputs": [],
m@6 22 "source": [
m@6 23 "filenames = ['/import/c4dm-04/mariap/train_data_melodia_8.pickle', \n",
m@6 24 " '/import/c4dm-04/mariap/val_data_melodia_8.pickle', \n",
m@6 25 " '/import/c4dm-04/mariap/test_data_melodia_8.pickle']"
m@6 26 ]
m@6 27 },
m@6 28 {
m@6 29 "cell_type": "code",
m@6 30 "execution_count": null,
m@6 31 "metadata": {
m@6 32 "collapsed": true
m@6 33 },
m@6 34 "outputs": [],
m@6 35 "source": [
m@6 36 "all_Yaudio = []\n",
m@6 37 "for filename in filenames:\n",
m@6 38 " _, Y, Yaudio = pickle.load(open(filename), 'rb')\n",
m@6 39 " all_Yaudio.append(Yaudio)\n",
m@6 40 "all_Yaudio = np.concatenate(all_Yaudio)"
m@6 41 ]
m@6 42 },
m@6 43 {
m@6 44 "cell_type": "code",
m@6 45 "execution_count": 3,
m@6 46 "metadata": {
m@6 47 "collapsed": false
m@6 48 },
m@6 49 "outputs": [
m@6 50 {
m@6 51 "ename": "NameError",
m@6 52 "evalue": "name 'all_Yaudio' is not defined",
m@6 53 "output_type": "error",
m@6 54 "traceback": [
m@6 55 "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
m@6 56 "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
m@6 57 "\u001b[0;32m<ipython-input-3-4107ada442c0>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0muniq_audio\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0muniq_counts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munique\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mall_Yaudio\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreturn_counts\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
m@6 58 "\u001b[0;31mNameError\u001b[0m: name 'all_Yaudio' is not defined"
m@6 59 ]
m@6 60 }
m@6 61 ],
m@6 62 "source": [
m@6 63 "uniq_audio, uniq_counts = np.unique(all_Yaudio, return_counts=True)"
m@6 64 ]
m@6 65 },
m@6 66 {
m@6 67 "cell_type": "markdown",
m@6 68 "metadata": {},
m@6 69 "source": [
m@6 70 "## Stats on audio files with very few music frames (after the speech/music discrimination)"
m@6 71 ]
m@6 72 },
m@6 73 {
m@6 74 "cell_type": "code",
m@6 75 "execution_count": 4,
m@6 76 "metadata": {
m@6 77 "collapsed": false
m@6 78 },
m@6 79 "outputs": [
m@6 80 {
m@6 81 "ename": "NameError",
m@6 82 "evalue": "name 'uniq_counts' is not defined",
m@6 83 "output_type": "error",
m@6 84 "traceback": [
m@6 85 "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
m@6 86 "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
m@6 87 "\u001b[0;32m<ipython-input-4-700ed156399c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0mmin_n_frames\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m10\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mn_short_files\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwhere\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0muniq_counts\u001b[0m\u001b[0;34m<\u001b[0m\u001b[0mmin_n_frames\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0;32mprint\u001b[0m \u001b[0;34m'%d files out of %d have less than %d frames'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mn_short_files\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0muniq_counts\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmin_n_frames\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
m@6 88 "\u001b[0;31mNameError\u001b[0m: name 'uniq_counts' is not defined"
m@6 89 ]
m@6 90 }
m@6 91 ],
m@6 92 "source": [
m@6 93 "min_n_frames = 10\n",
m@6 94 "n_short_files = np.where(uniq_counts<min_n_frames)[0].shape\n",
m@6 95 "print '%d files out of %d have less than %d frames' % (n_short_files, len(uniq_counts), min_n_frames)"
m@6 96 ]
m@6 97 },
m@6 98 {
m@6 99 "cell_type": "markdown",
m@6 100 "metadata": {},
m@6 101 "source": [
m@6 102 "## Stats on average duration of the music segments for all tracks"
m@6 103 ]
m@6 104 },
m@6 105 {
m@6 106 "cell_type": "code",
m@6 107 "execution_count": 5,
m@6 108 "metadata": {
m@6 109 "collapsed": false
m@6 110 },
m@6 111 "outputs": [
m@6 112 {
m@6 113 "ename": "SyntaxError",
m@6 114 "evalue": "invalid syntax (<ipython-input-5-2c4ab0e943a6>, line 1)",
m@6 115 "output_type": "error",
m@6 116 "traceback": [
m@6 117 "\u001b[0;36m File \u001b[0;32m\"<ipython-input-5-2c4ab0e943a6>\"\u001b[0;36m, line \u001b[0;32m1\u001b[0m\n\u001b[0;31m print 'mean %f' np.mean(uniq_counts)\u001b[0m\n\u001b[0m ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
m@6 118 ]
m@6 119 }
m@6 120 ],
m@6 121 "source": [
m@6 122 "sr = 2.0 # with 8-second window and 0.5-second hop size the sampling rate is 2 about 2 samples per second\n",
m@6 123 "print 'mean %f' % np.mean(uniq_counts)\n",
m@6 124 "print 'median %f' % np.median(uniq_counts)\n",
m@6 125 "print 'std %f' % np.std(uniq_counts)\n",
m@6 126 "print 'mean duration %f' % (np.mean(uniq_counts) / sr)"
m@6 127 ]
m@6 128 },
m@6 129 {
m@6 130 "cell_type": "markdown",
m@6 131 "metadata": {},
m@6 132 "source": [
m@6 133 "## Stats on average duration of the music segments for the British Library tracks"
m@6 134 ]
m@6 135 },
m@6 136 {
m@6 137 "cell_type": "code",
m@6 138 "execution_count": 7,
m@6 139 "metadata": {
m@6 140 "collapsed": false
m@6 141 },
m@6 142 "outputs": [
m@6 143 {
m@6 144 "ename": "NameError",
m@6 145 "evalue": "name 'uniq_audio' is not defined",
m@6 146 "output_type": "error",
m@6 147 "traceback": [
m@6 148 "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
m@6 149 "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
m@6 150 "\u001b[0;32m<ipython-input-7-4ebf50436e4a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m#British library tracks start with 'D:/Audio/...'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0midx_BL_tracks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0muniq_audio\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0muniq_audio\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msplit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'D:/'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m>\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0msr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m2.0\u001b[0m \u001b[0;31m# with 8-second window and 0.5-second hop size the sampling rate is 2 about 2 samples per second\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;32mprint\u001b[0m \u001b[0;34m'mean %f'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmean\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0muniq_counts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx_BL_tracks\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0;32mprint\u001b[0m \u001b[0;34m'median %f'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmedian\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0muniq_counts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx_BL_tracks\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
m@6 151 "\u001b[0;31mNameError\u001b[0m: name 'uniq_audio' is not defined"
m@6 152 ]
m@6 153 }
m@6 154 ],
m@6 155 "source": [
m@6 156 "#British library tracks start with 'D:/Audio/...'\n",
m@6 157 "idx_BL_tracks = np.array([i for i in range(len(uniq_audio)) if len(uniq_audio[i].split('D:/'))>1])\n",
m@6 158 "sr = 2.0 # with 8-second window and 0.5-second hop size the sampling rate is 2 about 2 samples per second\n",
m@6 159 "print 'mean %f' % np.mean(uniq_counts[idx_BL_tracks])\n",
m@6 160 "print 'median %f' % np.median(uniq_counts[idx_BL_tracks])\n",
m@6 161 "print 'std %f' % np.std(uniq_counts[idx_BL_tracks])\n",
m@6 162 "print 'mean duration %f' % (np.mean(uniq_counts[idx_BL_tracks]) / sr)"
m@6 163 ]
m@6 164 },
m@6 165 {
m@6 166 "cell_type": "code",
m@6 167 "execution_count": null,
m@6 168 "metadata": {
m@6 169 "collapsed": true
m@6 170 },
m@6 171 "outputs": [],
m@6 172 "source": []
m@6 173 }
m@6 174 ],
m@6 175 "metadata": {
m@6 176 "kernelspec": {
m@6 177 "display_name": "Python 2",
m@6 178 "language": "python",
m@6 179 "name": "python2"
m@6 180 },
m@6 181 "language_info": {
m@6 182 "codemirror_mode": {
m@6 183 "name": "ipython",
m@6 184 "version": 2
m@6 185 },
m@6 186 "file_extension": ".py",
m@6 187 "mimetype": "text/x-python",
m@6 188 "name": "python",
m@6 189 "nbconvert_exporter": "python",
m@6 190 "pygments_lexer": "ipython2",
m@6 191 "version": "2.7.12"
m@6 192 }
m@6 193 },
m@6 194 "nbformat": 4,
m@6 195 "nbformat_minor": 0
m@6 196 }