| { |
| "cells": [ |
| { |
| "cell_type": "markdown", |
| "metadata": {}, |
| "source": [ |
| "# Binary Predictions with Negative Sampling\n", |
| "Example of a recommender system making binary predictions instead of predicting a rating.\n", |
| "Demonstrates use of `NegativeSamplingDataIter` to wrap an existing data iterator with `CosineLoss`.\n", |
| "See [BlackOut by Shihao Ji et al](https://arxiv.org/abs/1511.06909) for more on negative sampling.\n", |
| "\n", |
| "You need to have python package pandas and bokeh installed (pip install pandas bokeh)." |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": false |
| }, |
| "outputs": [], |
| "source": [ |
| "import mxnet as mx\n", |
| "from movielens_data import get_data_iter, max_id\n", |
| "from matrix_fact import train\n", |
| "import recotools" |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": false |
| }, |
| "outputs": [], |
| "source": [ |
| "# If MXNet is not compiled with GPU support (e.g. on OSX), set to [mx.cpu(0)]\n", |
| "# Can be changed to [mx.gpu(0), mx.gpu(1), ..., mx.gpu(N-1)] if there are N GPUs\n", |
| "ctx = [mx.gpu(0)]" |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": false |
| }, |
| "outputs": [], |
| "source": [ |
| "pos_train_data, pos_test_data = get_data_iter(batch_size=100)\n", |
| "max_user, max_item = max_id('./ml-100k/u.data')\n", |
| "(max_user, max_item)" |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": false |
| }, |
| "outputs": [], |
| "source": [ |
| "train_data = recotools.NegativeSamplingDataIter(pos_train_data, sample_ratio=3, positive_label=0, negative_label=1)\n", |
| "test_data = recotools.NegativeSamplingDataIter(pos_test_data, sample_ratio=3, positive_label=0, negative_label=1)\n", |
| "train_test_data = (train_data, test_data)" |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": false |
| }, |
| "outputs": [], |
| "source": [ |
| "def plain_net(k):\n", |
| " # input\n", |
| " user = mx.symbol.Variable('user')\n", |
| " item = mx.symbol.Variable('item')\n", |
| " label = mx.symbol.Variable('score')\n", |
| " # user feature lookup\n", |
| " user = mx.symbol.Embedding(data = user, input_dim = max_user, output_dim = k)\n", |
| " # item feature lookup\n", |
| " item = mx.symbol.Embedding(data = item, input_dim = max_item, output_dim = k)\n", |
| " # loss layer\n", |
| " pred = recotools.CosineLoss(a=user, b=item, label=label)\n", |
| " return pred\n", |
| "\n", |
| "net1 = plain_net(64)\n", |
| "mx.viz.plot_network(net1)" |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": false, |
| "scrolled": false |
| }, |
| "outputs": [], |
| "source": [ |
| "results1 = train(net1, train_test_data, num_epoch=20, learning_rate=0.02, ctx=ctx)" |
| ] |
| }, |
| { |
| "cell_type": "code", |
| "execution_count": null, |
| "metadata": { |
| "collapsed": true |
| }, |
| "outputs": [], |
| "source": [] |
| } |
| ], |
| "metadata": { |
| "anaconda-cloud": {}, |
| "kernelspec": { |
| "display_name": "Python [Root]", |
| "language": "python", |
| "name": "Python [Root]" |
| }, |
| "language_info": { |
| "codemirror_mode": { |
| "name": "ipython", |
| "version": 2 |
| }, |
| "file_extension": ".py", |
| "mimetype": "text/x-python", |
| "name": "python", |
| "nbconvert_exporter": "python", |
| "pygments_lexer": "ipython2", |
| "version": "2.7.12" |
| } |
| }, |
| "nbformat": 4, |
| "nbformat_minor": 1 |
| } |