added sample weights to all estimators

added preprocessing for sparse X
reorganized the opt module
added sample weights to opt module

Merge branch 'sample-weights'

Author: idc9

docs/concave_penalty_two_stage_estimators.ipynb

@@ -139,7 +139,7 @@
     {
      "data": {
       "text/plain": [
-       "array([ 0.00000000e+00,  0.00000000e+00, -1.36847349e-07,  0.00000000e+00,\n",
+       "array([ 0.00000000e+00,  0.00000000e+00, -1.36847748e-07,  0.00000000e+00,\n",
        "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
        "        0.00000000e+00,  0.00000000e+00])"
       ]

docs/linear_regression_broad_overview.ipynb

@@ -18,7 +18,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -85,8 +85,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "CPU times: user 762 ms, sys: 11.5 ms, total: 773 ms\n",
-      "Wall time: 776 ms\n"
+      "CPU times: user 778 ms, sys: 13.4 ms, total: 791 ms\n",
+      "Wall time: 889 ms\n"
      ]
     }
    ],
@@ -117,8 +117,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "CPU times: user 2.21 s, sys: 41.8 ms, total: 2.26 s\n",
-      "Wall time: 2.26 s\n"
+      "CPU times: user 2.3 s, sys: 63.7 ms, total: 2.37 s\n",
+      "Wall time: 2.41 s\n"
      ]
     }
    ],
@@ -161,10 +161,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Lasso L2 to truth 1.1222397223387948\n",
-      "ENet L2 to truth 0.9661888417457567\n",
-      "Adaptive Lasso L2 to truth 0.5178048099475633\n",
-      "SCAD L2 to truth 0.8400704078733395\n"
+      "Lasso L2 to truth 1.1211348334707714\n",
+      "ENet L2 to truth 0.9630396203100647\n",
+      "Adaptive Lasso L2 to truth 0.5147703900551543\n",
+      "SCAD L2 to truth 0.8401140167810839\n"
      ]
     }
    ],
@@ -244,7 +244,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 8,
    "metadata": {
     "scrolled": false
    },
@@ -253,15 +253,15 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Entrywise lasso L2 to truth 0.911922407158712\n",
+      "Entrywise lasso L2 to truth 0.9119237212284282\n",
       "\n",
-      "Multi-task Lasso L2 to truth 0.8684933954494701\n",
-      "Adaptive multi-task lasso L2 to truth 0.7141523201068535\n",
-      "FCP multi-task with LLA L2 to truth 0.7502828771190455\n",
+      "Multi-task Lasso L2 to truth 0.8684943342448098\n",
+      "Adaptive multi-task lasso L2 to truth 0.7141545319969675\n",
+      "FCP multi-task with LLA L2 to truth 0.7502835459738096\n",
       "\n",
-      "Nuclear norm L2 to truth 0.709992352578054\n",
-      "Adaptive nuclear norm L2 to truth 0.4736246352155898\n",
-      "FCP nuclear nrom with LLA L2 to truth 0.5079065745103277\n"
+      "Nuclear norm L2 to truth 0.7099944507829721\n",
+      "Adaptive nuclear norm L2 to truth 0.47362674050743064\n",
+      "FCP nuclear nrom with LLA L2 to truth 0.507907042256749\n"
      ]
     }
    ],

docs/linear_regression_deeper_look.ipynb

@@ -13,10 +13,12 @@ def solve_glm(X, y,
               loss_func='lin_reg',
               loss_kws={},
               fit_intercept=True,
+              sample_weight=None,
 
               lasso_pen=None,
               lasso_weights=None,
 
+
               # groups=None,
               # L1to2=False,
               # nuc=False,
@@ -31,6 +33,9 @@ def solve_glm(X, y,
               p0=10, verbose=0, tol=1e-4, prune=0,
               return_n_iter=False):
 
+    if sample_weight is not None:
+        raise NotImplementedError("need to add")
+
     X = check_array(X, 'csc', dtype=[np.float64, np.float32],
                     order='F', copy=False, accept_large_sparse=False)
     y = check_array(y, 'csc', dtype=X.dtype.type, order='F', copy=False,

docs/loss_plus_penalty_combo.ipynb

@@ -14,6 +14,7 @@ def solve_glm(X, y,
               loss_func='lin_reg',
               loss_kws={},
               fit_intercept=True,
+              sample_weight=None,
               lasso_pen=None,
               lasso_weights=None,
               groups=None,
@@ -28,6 +29,9 @@ def solve_glm(X, y,
               solver=None,
               cp_kws={}):
 
+    if sample_weight is not None:
+        raise NotImplementedError("need to add")
+
     start_time = time()
     ######################
     # objective function #

ya_glm/backends/andersoncd/glm_solver.py

@@ -32,6 +32,7 @@ def solve(self, L1_weights, opt_init=None, opt_init_upv=None):
                            loss_func=self.glm_loss,
 
                            loss_kws=self.loss_kws,
+                           sample_weight=self.sample_weight,
                            fit_intercept=self.fit_intercept,
                            lasso_pen=1,
                            lasso_weights=L1_weights,