Mercurial > hg > dhoxss15

--- a/Vamp.ipynb	Thu Jul 16 11:06:35 2015 +0100
+++ b/Vamp.ipynb	Thu Jul 16 16:25:00 2015 +0100
@@ -14,7 +14,7 @@
    "source": [
     "### Setup\n",
     "\n",
-    "First import some necessary modules.\n",
+    "First we import some necessary modules.\n",
     "\n",
     "The `vamp` module loads and runs Vamp plugins. `librosa` is an audio analysis module from LabROSA at Columbia University. We are using it here only to load audio files, though it can also carry out some analysis functions. `matplotlib` is the usual plotting library."
    ]
@@ -23,7 +23,7 @@
    "cell_type": "code",
    "execution_count": 1,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -38,24 +38,7 @@
    "source": [
     "import vamp\n",
     "import librosa\n",
-    "import matplotlib.pyplot as plt"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "This bit of magic ensures that plots show up in the notebook rather than in a separate window:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
+    "import matplotlib.pyplot as plt\n",
     "%matplotlib inline"
    ]
   },
@@ -72,9 +55,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 2,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -178,7 +161,7 @@
        " 'vamp-test-plugin:vamp-test-plugin-freq']"
       ]
      },
-     "execution_count": 3,
+     "execution_count": 2,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -196,7 +179,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 3,
    "metadata": {
     "collapsed": true
    },
@@ -214,7 +197,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 4,
    "metadata": {
     "collapsed": true
    },
@@ -234,7 +217,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "metadata": {
     "collapsed": false
    },
@@ -252,9 +235,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 6,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -263,7 +246,7 @@
        "22050"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -281,7 +264,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 7,
    "metadata": {
     "collapsed": true
    },
@@ -304,9 +287,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 8,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -639,7 +622,7 @@
        "            0.00000000e+00,   3.16866152e-02,   5.19289337e-02]], dtype=float32))}"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -659,7 +642,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 9,
    "metadata": {
     "collapsed": false
    },
@@ -670,9 +653,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 10,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -681,7 +664,7 @@
        " 0.092879819"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -699,9 +682,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 11,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -710,7 +693,7 @@
        "2048.00000895"
       ]
      },
-     "execution_count": 12,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -721,9 +704,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 12,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
@@ -732,7 +715,7 @@
        "2048"
       ]
      },
-     "execution_count": 13,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -750,18 +733,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 13,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.image.AxesImage at 0x7fb0c6a80710>"
+       "<matplotlib.image.AxesImage at 0x7fc227986710>"
       ]
      },
-     "execution_count": 14,
+     "execution_count": 13,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -769,7 +752,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAABOCAYAAADvu5czAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7fb0c6d57cd0>"
+       "<matplotlib.figure.Figure at 0x7fc227c5acd0>"
       ]
      },
      "metadata": {},
@@ -784,13 +767,189 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "### How did we know which parameters were available?\n",
+    "A bit further back, we set up some plugin parameters in a Python dictionary:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "plugin_params = { \"tuningmode\": 1, \"s\": 0.9, \"chromanormalize\": 3 }"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We also included `output = chroma` in the `vamp.collect` call to make sure we got the right output from the plugin. How did we know which parameters and outputs were available?\n",
+    "\n",
+    "Outputs are pretty simple, there's a `vamp.get_outputs_of` call that returns them directly."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['logfreqspec',\n",
+       " 'tunedlogfreqspec',\n",
+       " 'semitonespectrum',\n",
+       " 'chroma',\n",
+       " 'basschroma',\n",
+       " 'bothchroma']"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "vamp.get_outputs_of(\"nnls-chroma:nnls-chroma\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Querying details of a plugin's parameters is a bit more tricky with the current version of the module. You have to use a lower-level interface for this. It looks like the following."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[{'defaultValue': 1.0,\n",
+       "  'description': 'Toggles approximate transcription (NNLS).',\n",
+       "  'identifier': 'useNNLS',\n",
+       "  'isQuantized': True,\n",
+       "  'maxValue': 1.0,\n",
+       "  'minValue': 0.0,\n",
+       "  'name': 'use approximate transcription (NNLS)',\n",
+       "  'quantizeStep': 1.0,\n",
+       "  'unit': ''},\n",
+       " {'defaultValue': 0.0,\n",
+       "  'description': 'Consider the cumulative energy spectrum (from low to high frequencies). All bins below the first bin whose cumulative energy exceeds the quantile [bass noise threshold] x [total energy] will be set to 0. A threshold value of 0 means that no bins will be changed.',\n",
+       "  'identifier': 'rollon',\n",
+       "  'isQuantized': True,\n",
+       "  'maxValue': 5.0,\n",
+       "  'minValue': 0.0,\n",
+       "  'name': 'bass noise threshold',\n",
+       "  'quantizeStep': 0.5,\n",
+       "  'unit': '%'},\n",
+       " {'defaultValue': 0.0,\n",
+       "  'description': 'Tuning can be performed locally or on the whole extraction segment. Local tuning is only advisable when the tuning is likely to change over the audio, for example in podcasts, or in a cappella singing.',\n",
+       "  'identifier': 'tuningmode',\n",
+       "  'isQuantized': True,\n",
+       "  'maxValue': 1.0,\n",
+       "  'minValue': 0.0,\n",
+       "  'name': 'tuning mode',\n",
+       "  'quantizeStep': 1.0,\n",
+       "  'unit': '',\n",
+       "  'valueNames': ['global tuning', 'local tuning']},\n",
+       " {'defaultValue': 1.0,\n",
+       "  'description': 'Spectral whitening: no whitening - 0; whitening - 1.',\n",
+       "  'identifier': 'whitening',\n",
+       "  'isQuantized': False,\n",
+       "  'maxValue': 1.0,\n",
+       "  'minValue': 0.0,\n",
+       "  'name': 'spectral whitening',\n",
+       "  'unit': ''},\n",
+       " {'defaultValue': 0.699999988079071,\n",
+       "  'description': 'Determines how individual notes in the note dictionary look: higher values mean more dominant higher harmonics.',\n",
+       "  'identifier': 's',\n",
+       "  'isQuantized': False,\n",
+       "  'maxValue': 0.8999999761581421,\n",
+       "  'minValue': 0.5,\n",
+       "  'name': 'spectral shape',\n",
+       "  'unit': ''},\n",
+       " {'defaultValue': 0.0,\n",
+       "  'description': 'How shall the chroma vector be normalized?',\n",
+       "  'identifier': 'chromanormalize',\n",
+       "  'isQuantized': True,\n",
+       "  'maxValue': 3.0,\n",
+       "  'minValue': 0.0,\n",
+       "  'name': 'chroma normalization',\n",
+       "  'quantizeStep': 1.0,\n",
+       "  'unit': '',\n",
+       "  'valueNames': ['none', 'maximum norm', 'L1 norm', 'L2 norm']}]"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "plug = vamp.vampyhost.load_plugin(\"nnls-chroma:nnls-chroma\", 44100, vamp.vampyhost.ADAPT_NONE)\n",
+    "plug.parameters"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "From this we can work out that (for example) to set the chroma vector normalisation to \"L2 norm\" we should supply something like `parameters = { \"chromanormalize\" : 3 }` in the call to `vamp.collect`.\n",
+    "\n",
+    "(Strictly speaking we're supposed to call"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "True"
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "plug.unload()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "after doing the above, as well; otherwise a bit of memory is wasted.)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "### Exporting the result to a CSV file\n",
     "This is a pretty standard use of Python's `csv` module."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 18,
    "metadata": {
     "collapsed": true
    },
@@ -822,7 +981,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 19,
    "metadata": {
     "collapsed": true
    },
@@ -846,9 +1005,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 20,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [
     {
--- a/Vamp.v3.ipynb	Thu Jul 16 11:06:35 2015 +0100
+++ b/Vamp.v3.ipynb	Thu Jul 16 16:25:00 2015 +0100
@@ -38,18 +38,19 @@
      "source": [
       "### Setup\n",
       "\n",
-      "First import some necessary modules.\n",
+      "First we import some necessary modules.\n",
       "\n",
       "The `vamp` module loads and runs Vamp plugins. `librosa` is an audio analysis module from LabROSA at Columbia University. We are using it here only to load audio files, though it can also carry out some analysis functions. `matplotlib` is the usual plotting library."
      ]
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "import vamp\n",
       "import librosa\n",
-      "import matplotlib.pyplot as plt"
+      "import matplotlib.pyplot as plt\n",
+      "%matplotlib inline"
      ],
      "language": "python",
      "metadata": {},
@@ -69,24 +70,6 @@
      "cell_type": "markdown",
      "metadata": {},
      "source": [
-      "This bit of magic ensures that plots show up in the notebook rather than in a separate window:"
-     ]
-    },
-    {
-     "cell_type": "code",
-     "collapsed": true,
-     "input": [
-      "%matplotlib inline"
-     ],
-     "language": "python",
-     "metadata": {},
-     "outputs": [],
-     "prompt_number": 2
-    },
-    {
-     "cell_type": "markdown",
-     "metadata": {},
-     "source": [
       "### Getting started with the Vamp module\n",
       "\n",
       "List the plugins that are installed. The strings that are returned here are referred to in the module documentation as _plugin keys_ -- each one consists of the name of the library file that contains the plugin, then a colon, then the identifier for the plugin itself. So a set of plugins with the same text before the colon (such as `qm-vamp-plugins:`) were all distributed in the same plugin library file.\n",
@@ -96,7 +79,7 @@
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "vamp.list_plugins()"
      ],
@@ -106,7 +89,7 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 3,
+       "prompt_number": 2,
        "text": [
         "['bbc-vamp-plugins:bbc-energy',\n",
         " 'bbc-vamp-plugins:bbc-intensity',\n",
@@ -207,7 +190,7 @@
        ]
       }
      ],
-     "prompt_number": 3
+     "prompt_number": 2
     },
     {
      "cell_type": "markdown",
@@ -225,7 +208,7 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 4
+     "prompt_number": 3
     },
     {
      "cell_type": "markdown",
@@ -243,7 +226,7 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 5
+     "prompt_number": 4
     },
     {
      "cell_type": "markdown",
@@ -263,7 +246,7 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 6
+     "prompt_number": 5
     },
     {
      "cell_type": "markdown",
@@ -274,7 +257,7 @@
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "rate"
      ],
@@ -284,13 +267,13 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 7,
+       "prompt_number": 6,
        "text": [
         "22050"
        ]
       }
      ],
-     "prompt_number": 7
+     "prompt_number": 6
     },
     {
      "cell_type": "markdown",
@@ -313,7 +296,7 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 8
+     "prompt_number": 7
     },
     {
      "cell_type": "markdown",
@@ -324,7 +307,7 @@
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "out"
      ],
@@ -334,7 +317,7 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 9,
+       "prompt_number": 8,
        "text": [
         "{'matrix': ( 0.092879819,\n",
         "  array([[  0.00000000e+00,   3.95574346e-02,   1.29732716e-05,\n",
@@ -664,7 +647,7 @@
        ]
       }
      ],
-     "prompt_number": 9
+     "prompt_number": 8
     },
     {
      "cell_type": "markdown",
@@ -684,11 +667,11 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 10
+     "prompt_number": 9
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "step"
      ],
@@ -698,13 +681,13 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 11,
+       "prompt_number": 10,
        "text": [
         " 0.092879819"
        ]
       }
      ],
-     "prompt_number": 11
+     "prompt_number": 10
     },
     {
      "cell_type": "markdown",
@@ -715,7 +698,7 @@
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "float(step) * rate"
      ],
@@ -725,17 +708,17 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 12,
+       "prompt_number": 11,
        "text": [
         "2048.00000895"
        ]
       }
      ],
-     "prompt_number": 12
+     "prompt_number": 11
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "vamp.vampyhost.RealTime.to_frame(step, rate)"
      ],
@@ -745,13 +728,13 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 13,
+       "prompt_number": 12,
        "text": [
         "2048"
        ]
       }
      ],
-     "prompt_number": 13
+     "prompt_number": 12
     },
     {
      "cell_type": "markdown",
@@ -762,7 +745,7 @@
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "plt.imshow(chroma.transpose(), origin=\"lower\")"
      ],
@@ -772,9 +755,9 @@
       {
        "metadata": {},
        "output_type": "pyout",
-       "prompt_number": 14,
+       "prompt_number": 13,
        "text": [
-        "<matplotlib.image.AxesImage at 0x7fb0c6a80710>"
+        "<matplotlib.image.AxesImage at 0x7fc227986710>"
        ]
       },
       {
@@ -782,16 +765,186 @@
        "output_type": "display_data",
        "png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAABOCAYAAADvu5czAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
        "text": [
-        "<matplotlib.figure.Figure at 0x7fb0c6d57cd0>"
+        "<matplotlib.figure.Figure at 0x7fc227c5acd0>"
        ]
       }
      ],
+     "prompt_number": 13
+    },
+    {
+     "cell_type": "markdown",
+     "metadata": {},
+     "source": [
+      "### How did we know which parameters were available?\n",
+      "A bit further back, we set up some plugin parameters in a Python dictionary:"
+     ]
+    },
+    {
+     "cell_type": "code",
+     "collapsed": true,
+     "input": [
+      "plugin_params = { \"tuningmode\": 1, \"s\": 0.9, \"chromanormalize\": 3 }"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [],
      "prompt_number": 14
     },
     {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
+      "We also included `output = chroma` in the `vamp.collect` call to make sure we got the right output from the plugin. How did we know which parameters and outputs were available?\n",
+      "\n",
+      "Outputs are pretty simple, there's a `vamp.get_outputs_of` call that returns them directly."
+     ]
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "vamp.get_outputs_of(\"nnls-chroma:nnls-chroma\")"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "metadata": {},
+       "output_type": "pyout",
+       "prompt_number": 15,
+       "text": [
+        "['logfreqspec',\n",
+        " 'tunedlogfreqspec',\n",
+        " 'semitonespectrum',\n",
+        " 'chroma',\n",
+        " 'basschroma',\n",
+        " 'bothchroma']"
+       ]
+      }
+     ],
+     "prompt_number": 15
+    },
+    {
+     "cell_type": "markdown",
+     "metadata": {},
+     "source": [
+      "Querying details of a plugin's parameters is a bit more tricky with the current version of the module. You have to use a lower-level interface for this. It looks like the following."
+     ]
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "plug = vamp.vampyhost.load_plugin(\"nnls-chroma:nnls-chroma\", 44100, vamp.vampyhost.ADAPT_NONE)\n",
+      "plug.parameters"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "metadata": {},
+       "output_type": "pyout",
+       "prompt_number": 16,
+       "text": [
+        "[{'defaultValue': 1.0,\n",
+        "  'description': 'Toggles approximate transcription (NNLS).',\n",
+        "  'identifier': 'useNNLS',\n",
+        "  'isQuantized': True,\n",
+        "  'maxValue': 1.0,\n",
+        "  'minValue': 0.0,\n",
+        "  'name': 'use approximate transcription (NNLS)',\n",
+        "  'quantizeStep': 1.0,\n",
+        "  'unit': ''},\n",
+        " {'defaultValue': 0.0,\n",
+        "  'description': 'Consider the cumulative energy spectrum (from low to high frequencies). All bins below the first bin whose cumulative energy exceeds the quantile [bass noise threshold] x [total energy] will be set to 0. A threshold value of 0 means that no bins will be changed.',\n",
+        "  'identifier': 'rollon',\n",
+        "  'isQuantized': True,\n",
+        "  'maxValue': 5.0,\n",
+        "  'minValue': 0.0,\n",
+        "  'name': 'bass noise threshold',\n",
+        "  'quantizeStep': 0.5,\n",
+        "  'unit': '%'},\n",
+        " {'defaultValue': 0.0,\n",
+        "  'description': 'Tuning can be performed locally or on the whole extraction segment. Local tuning is only advisable when the tuning is likely to change over the audio, for example in podcasts, or in a cappella singing.',\n",
+        "  'identifier': 'tuningmode',\n",
+        "  'isQuantized': True,\n",
+        "  'maxValue': 1.0,\n",
+        "  'minValue': 0.0,\n",
+        "  'name': 'tuning mode',\n",
+        "  'quantizeStep': 1.0,\n",
+        "  'unit': '',\n",
+        "  'valueNames': ['global tuning', 'local tuning']},\n",
+        " {'defaultValue': 1.0,\n",
+        "  'description': 'Spectral whitening: no whitening - 0; whitening - 1.',\n",
+        "  'identifier': 'whitening',\n",
+        "  'isQuantized': False,\n",
+        "  'maxValue': 1.0,\n",
+        "  'minValue': 0.0,\n",
+        "  'name': 'spectral whitening',\n",
+        "  'unit': ''},\n",
+        " {'defaultValue': 0.699999988079071,\n",
+        "  'description': 'Determines how individual notes in the note dictionary look: higher values mean more dominant higher harmonics.',\n",
+        "  'identifier': 's',\n",
+        "  'isQuantized': False,\n",
+        "  'maxValue': 0.8999999761581421,\n",
+        "  'minValue': 0.5,\n",
+        "  'name': 'spectral shape',\n",
+        "  'unit': ''},\n",
+        " {'defaultValue': 0.0,\n",
+        "  'description': 'How shall the chroma vector be normalized?',\n",
+        "  'identifier': 'chromanormalize',\n",
+        "  'isQuantized': True,\n",
+        "  'maxValue': 3.0,\n",
+        "  'minValue': 0.0,\n",
+        "  'name': 'chroma normalization',\n",
+        "  'quantizeStep': 1.0,\n",
+        "  'unit': '',\n",
+        "  'valueNames': ['none', 'maximum norm', 'L1 norm', 'L2 norm']}]"
+       ]
+      }
+     ],
+     "prompt_number": 16
+    },
+    {
+     "cell_type": "markdown",
+     "metadata": {},
+     "source": [
+      "From this we can work out that (for example) to set the chroma vector normalisation to \"L2 norm\" we should supply something like `parameters = { \"chromanormalize\" : 3 }` in the call to `vamp.collect`.\n",
+      "\n",
+      "(Strictly speaking we're supposed to call"
+     ]
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "plug.unload()"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "metadata": {},
+       "output_type": "pyout",
+       "prompt_number": 17,
+       "text": [
+        "True"
+       ]
+      }
+     ],
+     "prompt_number": 17
+    },
+    {
+     "cell_type": "markdown",
+     "metadata": {},
+     "source": [
+      "after doing the above, as well; otherwise a bit of memory is wasted.)"
+     ]
+    },
+    {
+     "cell_type": "markdown",
+     "metadata": {},
+     "source": [
       "### Exporting the result to a CSV file\n",
       "This is a pretty standard use of Python's `csv` module."
      ]
@@ -809,7 +962,7 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 15
+     "prompt_number": 18
     },
     {
      "cell_type": "markdown",
@@ -850,11 +1003,11 @@
      "language": "python",
      "metadata": {},
      "outputs": [],
-     "prompt_number": 16
+     "prompt_number": 19
     },
     {
      "cell_type": "code",
-     "collapsed": true,
+     "collapsed": false,
      "input": [
       "for file in glob.glob(\"data/Music/*.flac\"):\n",
       "    extract_chroma(file)"
@@ -870,7 +1023,7 @@
        ]
       }
      ],
-     "prompt_number": 17
+     "prompt_number": 20
     },
     {
      "cell_type": "markdown",