Skip to content

M
musicinformationretrieval-com
Commit d41ffadf authored by Steve Tjoa's avatar Steve Tjoa
Browse files
Options

removing knn

parent b92db02c
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 0 additions and 14942 deletions
index.html
View file @ d41ffadf
......@@ -11918,8 +11918,6 @@ div#notebook {
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<ol>
<li><a href="knn.html">K-Nearest Neighbor Classification</a> (<a href="knn.ipynb">ipynb</a>)</li>
<li><a href="knn_instrument_classification.html">Exercise: K-Nearest Neighbor Instrument Classification</a> (<a href="knn_instrument_classification.ipynb">ipynb</a>)</li>
<li><a href="kmeans.html">K-Means Clustering</a> (<a href="kmeans.ipynb">ipynb</a>)</li>
<li><a href="kmeans_instrument_classification.html">Exercise: Unsupervised Instrument Classification using K-Means</a> (<a href="kmeans_instrument_classification.ipynb">ipynb</a>)</li>
<li><a href="neural_networks.html">Neural Networks</a> (<a href="neural_networks.ipynb">ipynb</a>)</li>
......
index.ipynb
View file @ d41ffadf
......@@ -120,8 +120,6 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"1. [K-Nearest Neighbor Classification](knn.html) ([ipynb](knn.ipynb))\n",
"1. [Exercise: K-Nearest Neighbor Instrument Classification](knn_instrument_classification.html) ([ipynb](knn_instrument_classification.ipynb))\n",
"1. [K-Means Clustering](kmeans.html) ([ipynb](kmeans.ipynb))\n",
"1. [Exercise: Unsupervised Instrument Classification using K-Means](kmeans_instrument_classification.html) ([ipynb](kmeans_instrument_classification.ipynb))\n",
"1. [Neural Networks](neural_networks.html) ([ipynb](neural_networks.ipynb))\n",
......
knn.html deleted 100644 → 0
View file @ b92db02c
This diff is collapsed.
knn.ipynb deleted 100644 → 0
View file @ b92db02c
This diff is collapsed.
knn_instrument_classification.html deleted 100644 → 0
View file @ b92db02c
This source diff could not be displayed because it is too large. You can view the blob instead.
knn_instrument_classification.ipynb deleted 100644 → 0
View file @ b92db02c
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"%matplotlib inline\n",
"import seaborn\n",
"import numpy, scipy, matplotlib.pyplot as plt, sklearn, librosa, mir_eval, urllib, IPython.display, stanford_mir\n",
"plt.rcParams['figure.figsize'] = (14,5)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"[&larr; Back to Index](index.html)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Exercise: Instrument Classification using K-NN"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This exercise is loosely based upon \"Lab 1\" from previous MIR workshops ([2010](https://ccrma.stanford.edu/workshops/mir2010/Lab1_2010.pdf))."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For more on K-NN, see the [notebook on K-NN](knn.ipynb)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For help from a similar exercise, [follow the steps in the feature sonification exercise](feature_sonification.ipynb#Step-1:-Retrieve-Audio) first."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Goals"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"1. Extract spectral features from an audio signal.\n",
"2. Train a K-Nearest Neighbor classifier.\n",
"3. Use the classifier to classify beats in a drum loop."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 1: Retrieve Audio, Detect Onsets, and Segment"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Download the file `simple_loop.wav` onto your local machine."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"filename = 'simple_loop.wav'\n",
"urllib.urlretrieve?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load the audio file:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"librosa.load?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Play the audio file:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"IPython.display.Audio?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Detect onsets:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"librosa.onset.onset_detect?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Convert onsets from frames to seconds (and samples):"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"librosa.frames_to_time?"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"librosa.frames_to_samples?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Listen to a click track, with clicks located at each onset, plus the original audio:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"mir_eval.sonify.clicks?"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"IPython.display.Audio?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 2: Extract Features"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For each segment, compute the zero crossing rate and spectral centroid."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"librosa.zero_crossings?"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"librosa.feature.spectral_centroid?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Scale the features to be in the range [-1, 1]:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"sklearn.preprocessing.MinMaxScaler?"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"sklearn.preprocessing.MinMaxScaler.fit_transform?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 3: Train K-NN Classifier"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Use `stanford_mir.download_drum_samples` to download ten kick drum samples and ten snare drum samples. Each audio file contains a single drum hit at the beginning of the file."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"stanford_mir.download_drum_samples?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For each audio file, extract one feature vector. Concatenate all of these feature vectors into one feature table."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"numpy.concatenate?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 4: Run the Classifier"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Create a K-NN classifer model object:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"sklearn.neighbors.KNeighborsClassifier?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Train the classifier:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"sklearn.neighbors.KNeighborsClassifier.fit?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finally, run the classifier on the test input audio file:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"sklearn.neighbors.KNeighborsClassifier.predict?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 5: Sonify the Classifier Output"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Play a \"beep\" for each detected kick drum. Repeat for the snare drum."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"mir_eval.sonify.clicks?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## For Further Exploration"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In addition to the features used above, extract the following features (see librosa docs on [feature extraction](http://librosa.github.io/librosa/feature.html)):\n",
"\n",
"- spectral centroid\n",
"- spectral spread\n",
"- spectral skewness\n",
"- spectral kurtosis\n",
"- spectral rolloff\n",
"- MFCCs"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Re-train the classifier, and re-run the classifier over the test audio signal. Do the results change?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Repeat the steps above for more audio files."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"[&larr; Back to Index](index.html)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 2",
"language": "python",
"name": "python2"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.10"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment