Skip to main content
  • Home
  • Development
  • Documentation
  • Donate
  • Operational login
  • Browse the archive

swh logo
SoftwareHeritage
Software
Heritage
Archive
Features
  • Search

  • Downloads

  • Save code now

  • Add forge now

  • Help

Revision 1db48f3a735eb0fba06a7d503f080a7ead512604 authored by Artem Artemev on 11 July 2018, 12:50:44 UTC, committed by GitHub on 11 July 2018, 12:50:44 UTC
Update version.py file to 1.2.0 (#812)
1 parent 707b195
  • Files
  • Changes
  • 2109064
  • /
  • doc
  • /
  • source
  • /
  • notebooks
  • /
  • regression.ipynb
Raw File Download

To reference or cite the objects present in the Software Heritage archive, permalinks based on SoftWare Hash IDentifiers (SWHIDs) must be used.
Select below a type of object currently browsed in order to display its associated SWHID and permalink.

  • revision
  • directory
  • content
revision badge
swh:1:rev:1db48f3a735eb0fba06a7d503f080a7ead512604
directory badge
swh:1:dir:5e7ecf0b802d3e60cfac66660aeb4416a50ef469
content badge
swh:1:cnt:578b64a7282647a97d7728b01b0104de79f7c7aa

This interface enables to generate software citations, provided that the root directory of browsed objects contains a citation.cff or codemeta.json file.
Select below a type of object currently browsed in order to generate citations for them.

  • revision
  • directory
  • content
(requires biblatex-software package)
Generating citation ...
(requires biblatex-software package)
Generating citation ...
(requires biblatex-software package)
Generating citation ...
regression.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "GP Regression with GPflow\n",
    "--\n",
    "\n",
    "*James Hensman, 2015, 2016*\n",
    "\n",
    "GP regression (with Gaussian noise) is the most straightforward GP model in GPflow. Because of the conjugacy of the latent process and the noise, the marginal likelihood $p(\\mathbf y\\,|\\,\\theta)$ can be computed exactly.\n",
    "\n",
    "This notebook shows how to build a GPR model and estimate the parameters $\\theta$ by both maximum likelihood and MCMC. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:50.141046Z",
     "start_time": "2018-06-19T08:59:49.132677Z"
    }
   },
   "outputs": [],
   "source": [
    "import gpflow\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "%matplotlib inline\n",
    "matplotlib.rcParams['figure.figsize'] = (12, 6)\n",
    "plt = matplotlib.pyplot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First build a simple data set. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:50.291500Z",
     "start_time": "2018-06-19T08:59:50.141992Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa521aecc88>]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa523b2b278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "N = 12\n",
    "X = np.random.rand(N,1)\n",
    "Y = np.sin(12*X) + 0.66*np.cos(25*X) + np.random.randn(N,1)*0.1 + 3\n",
    "plt.plot(X, Y, 'kx', mew=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model construction\n",
    "\n",
    "A GPflow model is created by instantiating one of the GPflow model classes, in this case GPR. We'll make a kernel `k` and instantiate a GPR object using the generated data and the kernel. We'll set the variance of the likelihood to a sensible initial guess, too. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:50.402776Z",
     "start_time": "2018-06-19T08:59:50.292900Z"
    },
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:gpflow.logdensities:Shape of x must be 2D at computation.\n"
     ]
    }
   ],
   "source": [
    "k = gpflow.kernels.Matern52(1, lengthscales=0.3)\n",
    "m = gpflow.models.GPR(X, Y, kern=k)\n",
    "m.likelihood.variance = 0.01"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction\n",
    "\n",
    "GPflow models have several prediction methods:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " - `m.predict_f` returns the mean and variance of the latent function (f) at the points Xnew. \n",
    "\n",
    " - `m.predict_f_full_cov` additionally returns the full covariance matrix of the prediction.\n",
    "\n",
    " - `m.predict_y` returns the mean and variance of a new data point (i.e. includes the noise varaince). In the case of non-Gaussian likelihoods, the variance is computed by (numerically) integrating the non-Gaussian likelihood. \n",
    "\n",
    " - `m.predict_f_samples` returns samples of the latent function\n",
    "\n",
    " - `m.predict_density` returns the log-density of the points Ynew at Xnew. \n",
    " \n",
    "We'll use `predict_y` to make a simple plotting function. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:50.640000Z",
     "start_time": "2018-06-19T08:59:50.404378Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa520031f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot(m):\n",
    "    xx = np.linspace(-0.1, 1.1, 100).reshape(100, 1)\n",
    "    mean, var = m.predict_y(xx)\n",
    "    plt.figure(figsize=(12, 6))\n",
    "    plt.plot(X, Y, 'kx', mew=2)\n",
    "    plt.plot(xx, mean, 'C0', lw=2)\n",
    "    plt.fill_between(xx[:,0],\n",
    "                     mean[:,0] - 2*np.sqrt(var[:,0]),\n",
    "                     mean[:,0] + 2*np.sqrt(var[:,0]),\n",
    "                     color='C0', alpha=0.2)\n",
    "    plt.xlim(-0.1, 1.1)\n",
    "    \n",
    "plot(m)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using mean functions\n",
    "\n",
    "All GPflow models can have parameterized mean functions, some simple ones are provided in `gpflow.mean_functions`. Here's a model with a Linear mean function. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:50.774802Z",
     "start_time": "2018-06-19T08:59:50.641039Z"
    },
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:gpflow.logdensities:Shape of x must be 2D at computation.\n"
     ]
    }
   ],
   "source": [
    "k = gpflow.kernels.Matern52(1, lengthscales=0.3)\n",
    "meanf = gpflow.mean_functions.Linear(1.0, 0.0)\n",
    "m = gpflow.models.GPR(X, Y, k, meanf)\n",
    "m.likelihood.variance = 0.01"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Maximum Likelihood estimation of $\\theta$\n",
    "\n",
    "Getting the ML estimate of the parameters is a simple call to `minimize` method of the optimizer's instance, for e.g. `gpflow.train.ScipyOptimizer()`. By default, GPflow plugs into the L-BFGS-B algorithm via scipy. Here are the parameters before optimization: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:50.799532Z",
     "start_time": "2018-06-19T08:59:50.776372Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>prior</th>\n",
       "      <th>transform</th>\n",
       "      <th>trainable</th>\n",
       "      <th>shape</th>\n",
       "      <th>fixed_shape</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>GPR/kern/lengthscales</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/kern/variance</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/mean_function/b</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>(none)</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/mean_function/A</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>(none)</td>\n",
       "      <td>True</td>\n",
       "      <td>(1, 1)</td>\n",
       "      <td>True</td>\n",
       "      <td>[[1.0]]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/likelihood/variance</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             class prior transform  trainable   shape  \\\n",
       "GPR/kern/lengthscales    Parameter  None       +ve       True      ()   \n",
       "GPR/kern/variance        Parameter  None       +ve       True      ()   \n",
       "GPR/mean_function/b      Parameter  None    (none)       True      ()   \n",
       "GPR/mean_function/A      Parameter  None    (none)       True  (1, 1)   \n",
       "GPR/likelihood/variance  Parameter  None       +ve       True      ()   \n",
       "\n",
       "                         fixed_shape    value  \n",
       "GPR/kern/lengthscales           True      0.3  \n",
       "GPR/kern/variance               True      1.0  \n",
       "GPR/mean_function/b             True      0.0  \n",
       "GPR/mean_function/A             True  [[1.0]]  \n",
       "GPR/likelihood/variance         True     0.01  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.as_pandas_table()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here are the parameters after optimization, and a new plot. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:51.377278Z",
     "start_time": "2018-06-19T08:59:50.800715Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Optimization terminated with:\n",
      "  Message: b'CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH'\n",
      "  Objective function value: 9.122512\n",
      "  Number of iterations: 21\n",
      "  Number of functions evaluations: 23\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Optimization terminated with:\n",
      "  Message: b'CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH'\n",
      "  Objective function value: 9.122512\n",
      "  Number of iterations: 21\n",
      "  Number of functions evaluations: 23\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>prior</th>\n",
       "      <th>transform</th>\n",
       "      <th>trainable</th>\n",
       "      <th>shape</th>\n",
       "      <th>fixed_shape</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>GPR/kern/lengthscales</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0602623675564972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/kern/variance</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.6050496883970332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/mean_function/b</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>(none)</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>3.6281027899993705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/mean_function/A</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>(none)</td>\n",
       "      <td>True</td>\n",
       "      <td>(1, 1)</td>\n",
       "      <td>True</td>\n",
       "      <td>[[-1.38007666436]]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/likelihood/variance</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>None</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.002893234099970401</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             class prior transform  trainable   shape  \\\n",
       "GPR/kern/lengthscales    Parameter  None       +ve       True      ()   \n",
       "GPR/kern/variance        Parameter  None       +ve       True      ()   \n",
       "GPR/mean_function/b      Parameter  None    (none)       True      ()   \n",
       "GPR/mean_function/A      Parameter  None    (none)       True  (1, 1)   \n",
       "GPR/likelihood/variance  Parameter  None       +ve       True      ()   \n",
       "\n",
       "                         fixed_shape                 value  \n",
       "GPR/kern/lengthscales           True    0.0602623675564972  \n",
       "GPR/kern/variance               True    0.6050496883970332  \n",
       "GPR/mean_function/b             True    3.6281027899993705  \n",
       "GPR/mean_function/A             True    [[-1.38007666436]]  \n",
       "GPR/likelihood/variance         True  0.002893234099970401  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa523b1cd68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gpflow.train.ScipyOptimizer().minimize(m)\n",
    "plot(m)\n",
    "m.as_pandas_table()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the local optimum found by Maximum Likelihood may not be the one you want, e.g. it may be overfitting or be oversmooth. This depends on the initial values of hyperparameters and is specific to each data set. This discussion is beyond the scope of GPflow; an introduction can be found, for example, at [A Practical Guide to Gaussian Processes](https://drafts.distill.pub/gp/)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### MCMC for $\\theta$\n",
    "\n",
    "Here's a quick demonstration of how to obtain posteriors over the hyper-parameters in GP regression. First, we'll set some priors on the kernel parameters, then we'll run MCMC and see how much posterior uncertainty there is in the parameters.\n",
    "\n",
    "First we'll choose rather arbitrary priors. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T08:59:51.602372Z",
     "start_time": "2018-06-19T08:59:51.378393Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:gpflow.logdensities:Shape of x must be 2D at computation.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>prior</th>\n",
       "      <th>transform</th>\n",
       "      <th>trainable</th>\n",
       "      <th>shape</th>\n",
       "      <th>fixed_shape</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>GPR/kern/lengthscales</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>Ga([ 1.],[ 1.])</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.0602623675564972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/kern/variance</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>Ga([ 1.],[ 1.])</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.6050496883970332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/mean_function/b</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>N([ 0.],[ 10.])</td>\n",
       "      <td>(none)</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>3.6281027899993705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/mean_function/A</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>N([ 0.],[ 10.])</td>\n",
       "      <td>(none)</td>\n",
       "      <td>True</td>\n",
       "      <td>(1, 1)</td>\n",
       "      <td>True</td>\n",
       "      <td>[[-1.38007666436]]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GPR/likelihood/variance</th>\n",
       "      <td>Parameter</td>\n",
       "      <td>Ga([ 1.],[ 1.])</td>\n",
       "      <td>+ve</td>\n",
       "      <td>True</td>\n",
       "      <td>()</td>\n",
       "      <td>True</td>\n",
       "      <td>0.002893234099970401</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             class            prior transform  trainable  \\\n",
       "GPR/kern/lengthscales    Parameter  Ga([ 1.],[ 1.])       +ve       True   \n",
       "GPR/kern/variance        Parameter  Ga([ 1.],[ 1.])       +ve       True   \n",
       "GPR/mean_function/b      Parameter  N([ 0.],[ 10.])    (none)       True   \n",
       "GPR/mean_function/A      Parameter  N([ 0.],[ 10.])    (none)       True   \n",
       "GPR/likelihood/variance  Parameter  Ga([ 1.],[ 1.])       +ve       True   \n",
       "\n",
       "                          shape  fixed_shape                 value  \n",
       "GPR/kern/lengthscales        ()         True    0.0602623675564972  \n",
       "GPR/kern/variance            ()         True    0.6050496883970332  \n",
       "GPR/mean_function/b          ()         True    3.6281027899993705  \n",
       "GPR/mean_function/A      (1, 1)         True    [[-1.38007666436]]  \n",
       "GPR/likelihood/variance      ()         True  0.002893234099970401  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.clear()\n",
    "m.kern.lengthscales.prior = gpflow.priors.Gamma(1., 1.)\n",
    "m.kern.variance.prior = gpflow.priors.Gamma(1., 1.)\n",
    "m.likelihood.variance.prior = gpflow.priors.Gamma(1., 1.)\n",
    "m.mean_function.A.prior = gpflow.priors.Gaussian(0., 10.)\n",
    "m.mean_function.b.prior = gpflow.priors.Gaussian(0., 10.)\n",
    "m.compile()\n",
    "m.as_pandas_table()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T09:00:01.967192Z",
     "start_time": "2018-06-19T08:59:51.603429Z"
    },
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:gpflow.logdensities:Shape of x must be 2D at computation.\n",
      "WARNING:gpflow.logdensities:Shape of x must be 2D at computation.\n",
      "WARNING:gpflow.logdensities:Shape of x must be 2D at computation.\n"
     ]
    }
   ],
   "source": [
    "sampler = gpflow.train.HMC()\n",
    "samples = sampler.sample(m, num_samples=gpflow.test_util.notebook_niter(500), epsilon=0.05, lmin=10, lmax=20, logprobs=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T09:00:02.138487Z",
     "start_time": "2018-06-19T09:00:01.969028Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'parameter value')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa4d19341d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "for i, col in samples.iteritems():\n",
    "    plt.plot(col, label=col.name)\n",
    "plt.legend(loc=0)\n",
    "plt.xlabel('hmc iteration')\n",
    "plt.ylabel('parameter value')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the sampler runs in unconstrained space (so that positive parameters remain positive, parameters that are not trainable are ignored), but GPflow returns a dataframe with values in the true units.\n",
    "\n",
    "For serious analysis you most certainly want to run the sampler longer, with multiple chains and convergence checks. This will do for illustration though!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T09:00:02.375331Z",
     "start_time": "2018-06-19T09:00:02.139514Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'signal_variance')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa4d1884ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, axs = plt.subplots(1,3, figsize=(12,4))\n",
    "\n",
    "axs[0].plot(samples['GPR/likelihood/variance'],\n",
    "            samples['GPR/kern/variance'], 'k.', alpha = 0.15)\n",
    "axs[0].set_xlabel('noise_variance')\n",
    "axs[0].set_ylabel('signal_variance')\n",
    "\n",
    "axs[1].plot(samples['GPR/likelihood/variance'],\n",
    "            samples['GPR/kern/lengthscales'], 'k.', alpha = 0.15)\n",
    "axs[1].set_xlabel('noise_variance')\n",
    "axs[1].set_ylabel('lengthscale')\n",
    "\n",
    "axs[2].plot(samples['GPR/kern/lengthscales'],\n",
    "            samples['GPR/kern/variance'], 'k.', alpha = 0.1)\n",
    "axs[2].set_xlabel('lengthscale')\n",
    "axs[2].set_ylabel('signal_variance')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To plot the posterior of predictions, we'll iterate through the samples and set the model state with each sample. Then, for that state (set of hyper-parameters) we'll draw some samples from the prediction function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-19T09:00:03.022960Z",
     "start_time": "2018-06-19T09:00:02.376340Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa500fcd860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plot the function posterior\n",
    "xx = np.linspace(-0.1, 1.1, 100)[:,None]\n",
    "plt.figure(figsize=(12, 6))\n",
    "for i, s in samples.iloc[::20].iterrows():\n",
    "    f = m.predict_f_samples(xx, 1, initialize=False, feed_dict=m.sample_feed_dict(s))\n",
    "    plt.plot(xx, f[0,:,:], 'C0', lw=2, alpha=0.1)\n",
    "    \n",
    "plt.plot(X, Y, 'kx', mew=2)\n",
    "_ = plt.xlim(xx.min(), xx.max())\n",
    "_ = plt.ylim(0, 6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note the bimodal posterior: the data can be explained by a very smooth function (with lots of noise) or as a rough, short lengthscale function."
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
The diff you're trying to view is too large. Only the first 1000 changed files have been loaded.
Showing with 0 additions and 0 deletions (0 / 0 diffs computed)
swh spinner

Computing file changes ...

back to top

Software Heritage — Copyright (C) 2015–2026, The Software Heritage developers. License: GNU AGPLv3+.
The source code of Software Heritage itself is available on our development forge.
The source code files archived by Software Heritage are available under their own copyright and licenses.
Terms of use: Archive access, API— Content policy— Contact— JavaScript license information— Web API