Skip to content

Instantly share code, notes, and snippets.

@jllanfranchi
Last active November 1, 2016 15:14
Show Gist options
  • Save jllanfranchi/193ac0b2f5c0032a60e6b7ddef2479cd to your computer and use it in GitHub Desktop.
Save jllanfranchi/193ac0b2f5c0032a60e6b7ddef2479cd to your computer and use it in GitHub Desktop.
Display the source blob
Display the rendered blob
Raw
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Populating the interactive namespace from numpy and matplotlib\n"
]
}
],
"source": [
"%pylab inline"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"## Import custom modules"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# KDE module: obtain from\n",
"# https://github.com/jllanfranchi/kde\n",
"from kde.pykde import gaussian_kde, bootstrap_kde\n",
"\n",
"# My personal histogram plotter\n",
"from histogramTools import stepHist"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"## Load data"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"raw = np.loadtxt('./HESE-MuonGun-SplineMPE-weights-llhratio.txt')\n",
"data = raw[:,0]\n",
"weights = raw[:,1]"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"## Histogram for reference"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Note that this is normed to have area of 1."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"hist, edges = np.histogramdd(sample=data, weights=weights,\n",
" bins=50, normed=True)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"## Adaptive-bandwidth KDE"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"### Nominal KDE"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"kern_wts = gaussian_kde(data, weights=weights, bw_method='silverman',\n",
" adaptive=True, weight_adaptive_bw=False)\n",
"\n",
"# Derive points at which to sample the KDE\n",
"dmin = min(data)\n",
"dmax = max(data)\n",
"x = np.linspace(dmin, dmax, 1000)\n",
"\n",
"# Sample the KDE\n",
"nom_dens = kern_wts(x)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"### Bootstrapping to approximate expected estimator variance"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"boot_wts = bootstrap_kde(data, weights=weights, bw_method='silverman',\n",
" adaptive=True, weight_adaptive_bw=False,\n",
" niter=1000)\n",
"\n",
"# Sample the bootstrap: mean and stddev\n",
"boot_mean, boot_err = boot_wts(x)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"## Save results to disk"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Construct a Pandas DataFrame to make this task easier, including the sample x-values, the nominal KDE samples, and samples of the bootstrap mean and error."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>x</th>\n",
" <th>nominal_kde</th>\n",
" <th>boot_mean</th>\n",
" <th>boot_err</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>57.70000</td>\n",
" <td>0.000455</td>\n",
" <td>0.000415</td>\n",
" <td>0.000151</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>64.08999</td>\n",
" <td>0.000484</td>\n",
" <td>0.000440</td>\n",
" <td>0.000159</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>70.47998</td>\n",
" <td>0.000514</td>\n",
" <td>0.000467</td>\n",
" <td>0.000167</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>76.86997</td>\n",
" <td>0.000545</td>\n",
" <td>0.000496</td>\n",
" <td>0.000175</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>83.25996</td>\n",
" <td>0.000578</td>\n",
" <td>0.000525</td>\n",
" <td>0.000184</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" x nominal_kde boot_mean boot_err\n",
"0 57.70000 0.000455 0.000415 0.000151\n",
"1 64.08999 0.000484 0.000440 0.000159\n",
"2 70.47998 0.000514 0.000467 0.000167\n",
"3 76.86997 0.000545 0.000496 0.000175\n",
"4 83.25996 0.000578 0.000525 0.000184"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import pandas as pd\n",
"from collections import OrderedDict\n",
"\n",
"df = pd.DataFrame(OrderedDict((\n",
" ('x', x),\n",
" ('nominal_kde', nom_dens),\n",
" ('boot_mean', boot_mean),\n",
" ('boot_err', boot_err))\n",
"))\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"Save the dataframe to a CSV file."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"df.to_csv('kde_results.csv', sep=' ', index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": false
},
"source": [
"## Plot results"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAI6CAYAAABy9NwVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nvN6npwAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f10b1358090>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, axes = plt.subplots(2, 1, figsize=(6,8))\n",
"\n",
"for ax in axes:\n",
" ax.cla()\n",
" stepHist(edges[0], hist, ax=ax, color='b', label='Histogram')\n",
" ax.fill_between(\n",
" x, boot_mean-boot_err, boot_mean+boot_err,\n",
" lw=0, color='r', alpha=0.5,\n",
" label='Bootstrapped KDE variance est'\n",
" )\n",
" ax.plot(x, nom_dens, 'k-', lw=1, label='Nominal KDE')\n",
" ax.plot(x, boot_mean, 'r-', lw=1, label='Bootstrapped KDE mean')\n",
"\n",
"axes[0].set_title('Linear scale')\n",
"a0ylim = axes[0].get_ylim()\n",
"axes[0].set_ylim(0, a0ylim[1])\n",
"axes[1].set_yscale('log')\n",
"axes[1].set_title('Log scale');\n",
"axes[0].legend(loc='best')\n",
"fig.tight_layout();\n",
"fig.savefig('kde.pdf');"
]
}
],
"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": "2.7.11"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment