trying to remove usage of mc inplace of pm

Author: CamDavidsonPilon

Chapter1_Introduction/Chapter1_Introduction.ipynb

@@ -6,7 +6,7 @@ def draw_sky( galaxies ):
     """adapted from Vishal Goklani"""
     size_multiplier = 45
     fig = plt.figure(figsize=(10,10))
-    fig.patch.set_facecolor("blue")
+    #fig.patch.set_facecolor("blue")
     ax = fig.add_subplot(111, aspect='equal')
     n = galaxies.shape[0]
     for i in xrange(n):

Chapter2_MorePyMC/MorePyMC.ipynb

@@ -322,9 +322,9 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "import pymc as mc\n",
+      "import pymc as pm\n",
       "\n",
-      "to_include = mc.Bernoulli( \"to_include\", 0.5, size= 200 )"
+      "to_include = pm.Bernoulli( \"to_include\", 0.5, size= 200 )"
      ],
      "language": "python",
      "metadata": {},
@@ -335,7 +335,7 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "coef = mc.Uniform( \"coefs\", 0, 1, size = 200 )"
+      "coef = pm.Uniform( \"coefs\", 0, 1, size = 200 )"
      ],
      "language": "python",
      "metadata": {},
@@ -346,7 +346,7 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "@mc.deterministic\n",
+      "@pm.deterministic\n",
       "def Z( coef = coef, to_include = to_include, data = training_data ):\n",
       "    ym = np.dot( to_include*training_data, coef )\n",
       "    return ym - ym.mean()"
@@ -360,7 +360,7 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "@mc.deterministic\n",
+      "@pm.deterministic\n",
       "def T( z = Z ):\n",
       "    return 0.45*(np.sign(z) + 1.1)"
      ],
@@ -373,10 +373,10 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "obs = mc.Bernoulli( \"obs\", T, value = training_labels, observed = True)\n",
+      "obs = pm.Bernoulli( \"obs\", T, value = training_labels, observed = True)\n",
       "\n",
-      "model = mc.Model( [to_include, coef, Z, T, obs] )\n",
-      "map_ = mc.MAP( model )\n",
+      "model = pm.Model( [to_include, coef, Z, T, obs] )\n",
+      "map_ = pm.MAP( model )\n",
       "map_.fit()"
      ],
      "language": "python",
@@ -396,7 +396,7 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "mcmc = mc.MCMC( model )"
+      "mcmc = pm.MCMC( model )"
      ],
      "language": "python",
      "metadata": {},

Chapter4_TheGreatestTheoremNeverTold/LawOfLargeNumbers.ipynb

@@ -1,4 +1,4 @@
-import pymc as mc
+import pymc as pm
 import numpy as np
 
 count_data = np.loadtxt("../../Chapter1_Introduction/data/txtdata.csv")
@@ -7,21 +7,21 @@
 alpha = 1.0/count_data.mean() #recall count_data is 
                               #the variable that holds our txt counts
 
-lambda_1 = mc.Exponential( "lambda_1",  alpha )
-lambda_2 = mc.Exponential( "lambda_2", alpha )
+lambda_1 = pm.Exponential( "lambda_1",  alpha )
+lambda_2 = pm.Exponential( "lambda_2", alpha )
 
-tau = mc.DiscreteUniform( "tau", lower = 0, upper = n_count_data )
+tau = pm.DiscreteUniform( "tau", lower = 0, upper = n_count_data )
 
-@mc.deterministic
+@pm.deterministic
 def lambda_( tau = tau, lambda_1 = lambda_1, lambda_2 = lambda_2 ):
     out = np.zeros( n_count_data )  
     out[:tau] = lambda_1 #lambda before tau is lambda1
     out[tau:] = lambda_2 #lambda after tau is lambda1
     return out
 
-observation = mc.Poisson( "obs", lambda_, value = count_data, observed = True)
-model = mc.Model( [observation, lambda_1, lambda_2, tau] )
+observation = pm.Poisson( "obs", lambda_, value = count_data, observed = True)
+model = pm.Model( [observation, lambda_1, lambda_2, tau] )
 
 
-mcmc = mc.MCMC(model)
+mcmc = pm.MCMC(model)
 mcmc.sample( 100000, 50000, 1 )

Chapter5_LossFunctions/LossFunctions.ipynb

@@ -3,7 +3,7 @@
 
 """
 
-import pymc as mc
+import pymc as pm
 
 #these two quantities are unknown to us.
 true_p_A = 0.05 
@@ -14,28 +14,28 @@
 N_B = 1000 
 
 #generate data
-observations_A = mc.rbernoulli( true_p_A, N_A )
-observations_B = mc.rbernoulli( true_p_B, N_B )
+observations_A = pm.rbernoulli( true_p_A, N_A )
+observations_B = pm.rbernoulli( true_p_B, N_B )
 
 
 
 #set up the pymc model. Again assume Uniform priors for p_A and p_B
 
-p_A = mc.Uniform("p_A", 0, 1)
-p_B = mc.Uniform("p_B", 0, 1)
+p_A = pm.Uniform("p_A", 0, 1)
+p_B = pm.Uniform("p_B", 0, 1)
 
 
 #define the deterministic delta function. This is our unknown of interest.
 
-@mc.deterministic
+@pm.deterministic
 def delta( p_A = p_A, p_B = p_B ):
     return p_A - p_B
 
 
 #set of observations, in this case we have two observation datasets. 
-obs_A = mc.Bernoulli( "obs_A", p_A, value = observations_A, observed = True )
-obs_B = mc.Bernoulli( "obs_B", p_B, value = observations_B, observed = True )
+obs_A = pm.Bernoulli( "obs_A", p_A, value = observations_A, observed = True )
+obs_B = pm.Bernoulli( "obs_B", p_B, value = observations_B, observed = True )
 
 #to be explained in chapter 3. 
-mcmc = mc.MCMC( [p_A, p_B, delta, obs_A, obs_B] )
+mcmc = pm.MCMC( [p_A, p_B, delta, obs_A, obs_B] )
 mcmc.sample( 20000, 1000)

Chapter5_LossFunctions/draw_sky2.py

@@ -1,15 +1,15 @@
-import pymc as mc
+import pymc as pm
 
-p = mc.Uniform( "freq_cheating", 0, 1) 
+p = pm.Uniform( "freq_cheating", 0, 1) 
 
-@mc.deterministic
+@pm.deterministic
 def p_skewed( p =  p ):
     return 0.5*p + 0.25
 
-yes_responses = mc.Binomial( "number_cheaters", 100, p_skewed, value = 35, observed = True )
+yes_responses = pm.Binomial( "number_cheaters", 100, p_skewed, value = 35, observed = True )
 
-model = mc.Model( [yes_responses, p_skewed, p ] )
+model = pm.Model( [yes_responses, p_skewed, p ] )
 
 ### To Be Explained in Chapter 3!
-mcmc = mc.MCMC(model)
+mcmc = pm.MCMC(model)
 mcmc.sample( 50000, 25000 )

Chapter7_BayesianMachineLearning/DontOverfit.ipynb

@@ -1,4 +1,4 @@
-import pymc as mc
+import pymc as pm
 
 
 challenger_data = np.genfromtxt("../../Chapter2_MorePyMC/data/challenger_data.csv", skip_header = 1, usecols=[1,2], missing_values="NA", delimiter=",")
@@ -9,20 +9,20 @@
 temperature = challenger_data[:,0]
 D = challenger_data[:,1] #defect or not?
 
-beta = mc.Normal( "beta", 0, 0.001, value = 0 )
-alpha = mc.Normal( "alpha", 0, 0.001, value = 0 )
+beta = pm.Normal( "beta", 0, 0.001, value = 0 )
+alpha = pm.Normal( "alpha", 0, 0.001, value = 0 )
 
-@mc.deterministic
+@pm.deterministic
 def p( temp = temperature, alpha = alpha, beta = beta):
     return 1.0/( 1. + np.exp( beta*temperature + alpha) ) 
 
 
-observed = mc.Bernoulli( "bernoulli_obs", p, value = D, observed=True)
+observed = pm.Bernoulli( "bernoulli_obs", p, value = D, observed=True)
 
-model = mc.Model( [observed, beta, alpha] )
+model = pm.Model( [observed, beta, alpha] )
 
 #mysterious code to be explained in Chapter 3
-map_ = mc.MAP(model)
+map_ = pm.MAP(model)
 map_.fit()
-mcmc = mc.MCMC( model )
+mcmc = pm.MCMC( model )
 mcmc.sample( 260000, 220000, 2 )