added kwargs

paulruvolo · Apr 26, 2015 · f3cfb3b · f3cfb3b
1 parent a7fed6f
commit f3cfb3b
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 32 deletions.
diff --git a/ELLA.ipynb b/ELLA.ipynb
@@ -12,6 +12,7 @@
      "collapsed": false,
      "input": [
       "%matplotlib inline\n",
+      "from sklearn.cross_validation import train_test_split\n",
       "\n",
       "def multi_task_train_test_split(Xs,Ys,train_size=0.5):\n",
       "    Xs_train = []\n",
@@ -66,32 +67,14 @@
      ],
      "language": "python",
      "metadata": {},
-     "outputs": [
-      {
-       "output_type": "stream",
-       "stream": "stdout",
-       "text": [
-        "Average explained variance score 0.998366292378\n",
-        "Average classification accuracy"
-       ]
-      },
-      {
-       "output_type": "stream",
-       "stream": "stdout",
-       "text": [
-        " 0.905\n"
-       ]
-      }
-     ],
-     "prompt_number": 9
+     "outputs": []
     },
     {
      "cell_type": "code",
      "collapsed": false,
      "input": [
       "from scipy.io import loadmat\n",
       "from sklearn.metrics import roc_auc_score\n",
-      "from sklearn.cross_validation import train_test_split\n",
       "\n",
       "data = loadmat('landminedata.mat')\n",
       "\n",
@@ -107,7 +90,7 @@
       "\n",
       "Xs_lm_train, Xs_lm_test, Ys_lm_train, Ys_lm_test = multi_task_train_test_split(Xs_lm,Ys_lm,train_size=0.5)    \n",
       "\n",
-      "model = ELLA(d,k,LogisticRegression,mu=1,lam=10**-5)\n",
+      "model = ELLA(d,k,LogisticRegression,{'C':10**-1},mu=1,lam=10**-5)\n",
       "for t in range(T):\n",
       "    model.fit(Xs_lm_train[t], Ys_lm_train[t], t)\n",
       "print \"Average AUC:\", np.mean([roc_auc_score(Ys_lm_test[t],\n",
@@ -116,16 +99,7 @@
      ],
      "language": "python",
      "metadata": {},
-     "outputs": [
-      {
-       "output_type": "stream",
-       "stream": "stdout",
-       "text": [
-        "Average AUC: 0.810932123838\n"
-       ]
-      }
-     ],
-     "prompt_number": 12
+     "outputs": []
     },
     {
      "cell_type": "code",

diff --git a/ELLA.py b/ELLA.py
@@ -11,7 +11,7 @@
 
 class ELLA(object):
     """ The ELLA model """
-    def __init__(self,d,k,base_learner,mu=1,lam=1):
+    def __init__(self,d,k,base_learner,base_learner_kwargs={},mu=1,lam=1):
     	""" Initializes a new model for the given base_learner.
     	    d: the number of parameters for the base learner
     	    k: the number of latent model components
@@ -39,6 +39,7 @@ def __init__(self,d,k,base_learner,mu=1,lam=1):
             raise Exception("Unsupported Base Learner")
 
         self.base_learner = base_learner
+        self.base_learner_kwargs = base_learner_kwargs
 
     def fit(self,X,y,task_id):
         """ Fit the model to a new batch of training data.  The task_id must
@@ -50,7 +51,7 @@ def fit(self,X,y,task_id):
             task_id: the id of the task
         """
         self.T += 1
-        single_task_model = self.base_learner(fit_intercept=False).fit(X,y)
+        single_task_model = self.base_learner(fit_intercept=False,**self.base_learner_kwargs).fit(X,y)
         D_t = self.get_hessian(single_task_model, X, y)
         D_t_sqrt = sqrtm(D_t)
         theta_t = single_task_model.coef_