Lasso in PyMC3

lasso.py

from pymc import *
from scipy.stats import norm
import pylab as plt

# Same model as the tutorial    
n = 1000    
x1 = norm.rvs(0, 1, size=n)
x2 = -x1 + norm.rvs(0, 10**-3, size=n)
x3 = norm.rvs(0, 1, size=n)

y = 10 * x1 + 10 * x2 + 0.1 * x3
  
with Model() as model:
    # Laplacian prior only works with Metropolis sampler
    coef1 = Laplace('x1', 0, b=1/sqrt(2))
    coef2 = Laplace('x2', 0, b=1/sqrt(2))
    coef3 = Laplace('x3', 0, b=1/sqrt(2))
    
    # Gaussian prior works with NUTS sampler
    #coef1 = Normal('x1', mu = 0, sd = 1)
    #coef2 = Normal('x2', mu = 0, sd = 1)
    #coef3 = Normal('x3', mu = 0, sd = 1)
    
    likelihood = Normal('y', mu= coef1 * x1 + coef2 * x2 + coef3 * x3, tau = 1, observed=y)
    
    #step = Metropolis() # Works just like PyMC2
    start = find_MAP() 
    step = NUTS(scaling = start)
    trace = sample(20000, step, start = start, progressbar=True) 
    trace = trace[1000::5]
    
plt.figure(figsize=(7, 7))
traceplot(trace)
plt.tight_layout()

autocorrplot(trace)
summary(trace)

notebook.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:33e74e04f0501867e2241ee2beb65b2b6297307b216314e6114c5de9773bb90a"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "\n",
      "from pymc import *\n",
      "from scipy.stats import norm\n",
      "import pylab as plt\n",
      "\n",
      "# Same model as the tutorial    \n",
      "n = 1000    \n",
      "x1 = norm.rvs(0, 1, size=n)\n",
      "x2 = -x1 + norm.rvs(0, 10**-3, size=n)\n",
      "x3 = norm.rvs(0, 1, size=n)\n",
      "\n",
      "y = 10 * x1 + 10 * x2 + 0.1 * x3\n",
      "  \n",
      "with Model() as model:\n",
      "    # Laplacian prior only works with Metropolis sampler\n",
      "    coef1 = Laplace('x1', 0, b=1/sqrt(2))\n",
      "    coef2 = Laplace('x2', 0, b=1/sqrt(2))\n",
      "    coef3 = Laplace('x3', 0, b=1/sqrt(2))\n",
      "    \n",
      "    # Gaussian prior works with NUTS sampler\n",
      "    #coef1 = Normal('x1', mu = 0, sd = 1)\n",
      "    #coef2 = Normal('x2', mu = 0, sd = 1)\n",
      "    #coef3 = Normal('x3', mu = 0, sd = 1)\n",
      "    \n",
      "    likelihood = Normal('y', mu= coef1 * x1 + coef2 * x2 + coef3 * x3, tau = 1, observed=y)\n",
      "    \n",
      "    #step = Metropolis() # Works just like PyMC2\n",
      "    start = find_MAP() # Doesn't help\n",
      "    step = NUTS(scaling = start) # Doesn't work\n",
      "    trace = sample(10000, step, start = start, progressbar=True) \n",
      "    trace = trace[1000::5]\n",
      "    \n",
      "plt.figure(figsize=(7, 7))\n",
      "traceplot(trace)\n",
      "plt.tight_layout()\n",
      "\n",
      "autocorrplot(trace)\n",
      "summary(trace)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [                  0%                  ] 88 of 10000 complete in 0.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [                  2%                  ] 223 of 10000 complete in 1.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-                 3%                  ] 337 of 10000 complete in 1.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-                 4%                  ] 427 of 10000 complete in 2.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--                5%                  ] 542 of 10000 complete in 2.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--                6%                  ] 646 of 10000 complete in 3.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--                7%                  ] 747 of 10000 complete in 3.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---               8%                  ] 870 of 10000 complete in 4.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---               9%                  ] 974 of 10000 complete in 4.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [----             11%                  ] 1117 of 10000 complete in 5.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [----             12%                  ] 1226 of 10000 complete in 5.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----            13%                  ] 1333 of 10000 complete in 6.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----            14%                  ] 1459 of 10000 complete in 6.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----            15%                  ] 1567 of 10000 complete in 7.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [------           16%                  ] 1699 of 10000 complete in 7.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [------           18%                  ] 1811 of 10000 complete in 8.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-------          19%                  ] 1935 of 10000 complete in 8.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-------          20%                  ] 2060 of 10000 complete in 9.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--------         21%                  ] 2174 of 10000 complete in 9.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--------         22%                  ] 2284 of 10000 complete in 10.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------        24%                  ] 2431 of 10000 complete in 10.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------        25%                  ] 2549 of 10000 complete in 11.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [----------       26%                  ] 2671 of 10000 complete in 11.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [----------       27%                  ] 2795 of 10000 complete in 12.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------      29%                  ] 2912 of 10000 complete in 12.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------      30%                  ] 3065 of 10000 complete in 13.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [------------     31%                  ] 3190 of 10000 complete in 13.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [------------     33%                  ] 3326 of 10000 complete in 14.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-------------    34%                  ] 3428 of 10000 complete in 14.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-------------    35%                  ] 3551 of 10000 complete in 15.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--------------   36%                  ] 3699 of 10000 complete in 15.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--------------   38%                  ] 3839 of 10000 complete in 16.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------------  39%                  ] 3963 of 10000 complete in 16.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------------  40%                  ] 4097 of 10000 complete in 17.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------------- 42%                  ] 4234 of 10000 complete in 17.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------------- 43%                  ] 4348 of 10000 complete in 18.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------------- 44%                  ] 4458 of 10000 complete in 18.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------45%                  ] 4597 of 10000 complete in 19.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------47%                  ] 4704 of 10000 complete in 19.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------48%                  ] 4850 of 10000 complete in 20.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------49%                  ] 4984 of 10000 complete in 20.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------51%                  ] 5111 of 10000 complete in 21.1 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------52%                  ] 5253 of 10000 complete in 21.6 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------53%                  ] 5392 of 10000 complete in 22.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------55%                  ] 5522 of 10000 complete in 22.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------56%-                 ] 5655 of 10000 complete in 23.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------57%-                 ] 5788 of 10000 complete in 23.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------59%--                ] 5916 of 10000 complete in 24.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------60%--                ] 6039 of 10000 complete in 24.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------61%---               ] 6112 of 10000 complete in 25.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------62%---               ] 6245 of 10000 complete in 25.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------63%----              ] 6346 of 10000 complete in 26.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------64%----              ] 6477 of 10000 complete in 26.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------65%----              ] 6577 of 10000 complete in 27.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------66%-----             ] 6689 of 10000 complete in 27.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------68%-----             ] 6811 of 10000 complete in 28.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------69%------            ] 6916 of 10000 complete in 28.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------70%------            ] 7006 of 10000 complete in 29.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------71%-------           ] 7110 of 10000 complete in 29.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------72%-------           ] 7229 of 10000 complete in 30.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------73%-------           ] 7322 of 10000 complete in 30.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------74%--------          ] 7475 of 10000 complete in 31.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------75%--------          ] 7590 of 10000 complete in 31.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------76%---------         ] 7697 of 10000 complete in 32.2 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------77%---------         ] 7799 of 10000 complete in 32.7 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------79%----------        ] 7941 of 10000 complete in 33.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------80%----------        ] 8061 of 10000 complete in 33.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------81%-----------       ] 8185 of 10000 complete in 34.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------83%-----------       ] 8303 of 10000 complete in 34.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------84%------------      ] 8426 of 10000 complete in 35.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------85%------------      ] 8553 of 10000 complete in 35.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------86%------------      ] 8680 of 10000 complete in 36.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------88%-------------     ] 8803 of 10000 complete in 36.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------89%-------------     ] 8947 of 10000 complete in 37.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------90%--------------    ] 9089 of 10000 complete in 37.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------92%--------------    ] 9209 of 10000 complete in 38.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------93%---------------   ] 9324 of 10000 complete in 38.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------94%---------------   ] 9433 of 10000 complete in 39.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------95%----------------  ] 9577 of 10000 complete in 39.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------96%----------------  ] 9697 of 10000 complete in 40.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------98%----------------- ] 9813 of 10000 complete in 40.8 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------99%----------------- ] 9956 of 10000 complete in 41.3 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------100%-----------------] 10000 of 10000 complete in 41.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "x1:\n",
        " \n",
        "  Mean             SD               MC Error         95% HPD interval\n",
        "  -------------------------------------------------------------------\n",
        "  \n",
        "  0.011            0.491            0.017            [-0.917, 1.180]\n",
        "\n",
        "  Posterior quantiles:\n",
        "  2.5            25             50             75             97.5\n",
        "  |--------------|==============|==============|--------------|\n",
        "  \n",
        "  -0.995         -0.246         -0.001         0.250          1.120\n",
        "\n",
        "\n",
        "x2:\n",
        " \n",
        "  Mean             SD               MC Error         95% HPD interval\n",
        "  -------------------------------------------------------------------\n",
        "  \n",
        "  0.011            0.491            0.017            [-0.985, 1.116]\n",
        "\n",
        "  Posterior quantiles:\n",
        "  2.5            25             50             75             97.5\n",
        "  |--------------|==============|==============|--------------|\n",
        "  \n",
        "  -0.990         -0.243         -0.001         0.248          1.113\n",
        "\n",
        "\n",
        "x3:\n",
        " \n",
        "  Mean             SD               MC Error         95% HPD interval\n",
        "  -------------------------------------------------------------------\n",
        "  \n",
        "  0.100            0.032            0.001            [0.035, 0.162]\n",
        "\n",
        "  Posterior quantiles:\n",
        "  2.5            25             50             75             97.5\n",
        "  |--------------|==============|==============|--------------|\n",
        "  \n",
        "  0.036          0.079          0.101          0.120          0.164\n",
        "\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "<matplotlib.figure.Figure at 0x7f1a3a0c7bd0>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAAGqCAYAAAD5mI6tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nOQ8AAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f1a3726f3d0>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEPCAYAAAC6Kkg/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxywAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f1a36f5c050>"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}