Skip to content

Instantly share code, notes, and snippets.

@bmcfee
Created July 10, 2013 16:38
Show Gist options
  • Save bmcfee/5967897 to your computer and use it in GitHub Desktop.
Save bmcfee/5967897 to your computer and use it in GitHub Desktop.
A demonstration of buffered generators for online learning
Display the source blob
Display the rendered blob
Raw
{
"metadata": {
"name": "Demo - buffered generators"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Buffered generators for online learning\n",
"\n",
"Several of the algorithms provided by `sklearn` are implemented as online learning or stochastic gradient descent methods, which only operate on a small portion of data at any given time. As concrete examples, see `sklearn.linear_model.SGDClassifier`, `sklearn.cluster.MiniBatchKMeans`, or `sklearn.decomposition.MiniBatchDictionaryLearning`.\n",
"\n",
"In principle, these methods should support learning from an endless stream of data. However, in practice, the implementations require contiguous memory allocation (via `numpy.array`) for data storage. This means that all data must be allocated up-front as a contiguous array, and we cannot simply have a `generator` supply data examples as needed. As a result, we would be limited by the amount of memory on the machine, even though the learning algorithm only ever operates on a small subset at any given time. As discussed in detail on the [Machined Learnings blog](http://www.machinedlearnings.com/2013/05/data-has-mass-code-accordingly.html), it would be better if data can live out of core until it is needed.\n",
"\n",
"To resolve this issue, we can implement a thin wrapper class that buffers the output from a data generator, and once the buffer is full, feeds it to the estimator via `partial_fit`.\n",
"\n",
"This notebook provides a brief demonstration of the idea by running k-means on a large collection of audio samples. \n",
"\n",
"The dependencies are as follows:\n",
"\n",
" * [sklearn](http://scikit-learn.org/stable/) - for obvious reasons\n",
" * [datastream](https://github.com/bmcfee/ml_scraps/blob/master/datastream.py), [muxerator](https://github.com/bmcfee/ml_scraps/blob/master/muxerator.py) - a particular data generator for randomly multiplexing over files\n",
" * [librosa](https://github.com/bmcfee/librosa) - an audio processing library\n",
" * [BufferedEstimator](https://github.com/bmcfee/ml_scraps/blob/master/BufferedEstimator.py) - the generator buffer class\n",
" * To run this demo script, you will need to start ipython notebook in `--pylab` mode.\n",
"\n",
"--- [Brian McFee](http://cosmal.ucsd.edu/~bmcfee/), 2013-07-10."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Our core estimator class\n",
"from sklearn.cluster import MiniBatchKMeans\n",
"\n",
"# Our feature extraction library\n",
"import librosa\n",
"\n",
"# Our data generator class\n",
"import glob\n",
"from datastream import datastream\n",
"\n",
"# Finally, our generator buffer\n",
"from BufferedEstimator import BufferedEstimator"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 2
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This `mapper` function is specific to this example. It takes as input a filename (ie, a path to an audio file), and generates `n` randomly selected samples from the audio file's spectrogram."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def mapper(filename, n=20):\n",
" \"\"\"Audio frame generator.\n",
" Given an input audio file, this generator will yield a random selection of spectrogram frames.\n",
"\n",
" :parameters:\n",
" - filename : str\n",
" Path to the audio file\n",
" - n : int>0\n",
" Maximum number of frames to generate\n",
"\n",
" \"\"\"\n",
"\n",
" SR = 22050\n",
" N_FFT = 2048\n",
" HOP_LENGTH = 512\n",
" N_MELS = 128\n",
" F_MAX = 8000\n",
"\n",
" def loadspec():\n",
" y, sr = librosa.load(filename, sr=SR)\n",
" S = librosa.feature.melspectrogram(y, sr, n_fft=N_FFT, hop_length=HOP_LENGTH, n_mels=N_MELS, fmax=F_MAX)\n",
" return S / S.max()\n",
"\n",
" S = loadspec()\n",
" \n",
" for i in range(n):\n",
" # Grab a random frame\n",
" t = np.random.randint(0, S.shape[1])\n",
" \n",
" yield S[:, t]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 3
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The list `files` will be processed by the `datastream` class as follows:\n",
"\n",
" * `datastream` will maintain a pool of `k=10` active files.\n",
" * Each active filename `f` is converted to a generator by the above `mapper` function.\n",
" * `datastream` then randomly multiplexes over the `k` generators.\n",
" * When the pool is exhausted (all generators are empty), it is replenished by activating another `k` random files (without replacement).\n",
" * This process repeats until all data has been processed.\n",
"\n",
"If random multiplexing isn't your thing, `itertools.chain` operates similarly, but exhausts each file sequentially. While chaining is much simpler, it is probably not what you want for stochastic optimization as it introduces a great deal of temporal dependence between the generated examples.\n",
"\n",
"If you're really interested, these wav files can be obtained [here](http://smc.inescporto.pt/research/data-2/)."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"files = sorted(glob.glob('data/SMC_Mirex/SMC_MIREX_Audio/SMC_0*.wav'))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 4
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data_generator = datastream(mapper, files, k=10)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 5
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Our estimator is a simple `MiniBatchKMeans` object. We'll set `k=32` for demonstration purposes.\n",
"\n",
"Any `sklearn.base.BaseEstimator` class will work here, as long as it implements the `partial_fit()` method.\n",
"\n",
"If your estimator is unsupervised (as we have here), then the generator should yield a single feature vector `x` at each step.\n",
"\n",
"If your estimator is supervised (eg, `SGDClassifier`), then the generator should yield a tuple `(x, y)` at each step, where `x` is a feature vector and `y` is its label."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"estimator = MiniBatchKMeans(k=32)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 6
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The buffered estimator object takes the k-means estimator we constructed above, and a batch_size for the buffer.\n",
"\n",
"Here, we will perform each `partial_fit` update on a batch of size at most 256."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"buf_est = BufferedEstimator(estimator, batch_size=256)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 7
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"To train the model, we simply call the `fit()` method on the data generator."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"buf_est.fit(data_generator)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 8
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"After training, we can visualize the results:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"figure(figsize=(18,6))\n",
"codebook = estimator.cluster_centers_.T\n",
"\n",
"imshow(codebook ** 0.33333333, aspect='auto', interpolation='none', origin='lower')\n",
"\n",
"vlines(np.arange(-.5, codebook.shape[1]-.5), -.5, codebook.shape[0]-.5, colors='w', linestyles='--')\n",
"colorbar()\n",
"title('Codebook, k=32')\n",
"pass"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAA5wAAAF4CAYAAAA4+PZsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n"
}
],
"prompt_number": 10
}
],
"metadata": {}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment