added linear model

alphamind/cyimpl.pyx

@@ -34,4 +34,4 @@ cpdef groupby(long[:] groups):
             group_ids[curr_tag] = [i]
         else:
             deref(it).second.push_back(i)
-    return [np.array(v) for v in group_ids.values()]
+    return {k: np.array(v) for k, v in group_ids.items()}
\ No newline at end of file

alphamind/data/neutralize.py

@@ -37,7 +37,7 @@ def neutralize(x: np.ndarray, y: np.ndarray, groups: np.ndarray=None, output_exp
 
         groups_ids = groupby(groups)
 
-        for curr_idx in groups_ids:
+        for curr_idx in groups_ids.values():
             curr_x = x[curr_idx]
             curr_y = y[curr_idx]
             b = ls_fit(curr_x, curr_y)

alphamind/model/linearmodel.py

+# -*- coding: utf-8 -*-
+"""
+Created on 2017-5-10
+
+@author: cheng.li
+"""
+
+from typing import Union
+import numpy as np
+from alphamind.cyimpl import groupby
+from alphamind.data.neutralize import ls_fit
+
+
+def _train(x: np.ndarray, y: np.ndarray, groups: np.ndarray=None) -> Union[np.ndarray, dict]:
+    if groups is None:
+        return ls_fit(x, y)
+    else:
+        groups_ids = groupby(groups)
+        res_beta = {}
+
+        for k, curr_idx in groups_ids.items():
+            curr_x = x[curr_idx]
+            curr_y = y[curr_idx]
+            res_beta[k] = ls_fit(curr_x, curr_y)
+
+        return res_beta
+
+
+if __name__ == '__main__':
+    x = np.random.randn(3000, 10)
+    y = np.random.randn(3000)
+    groups = np.random.randint(30, size=3000)
+
+    print(_train(x, y, groups))
\ No newline at end of file

alphamind/portfolio/percentbuilder.py

@@ -21,7 +21,7 @@ def percent_build(er: np.ndarray, percent: float, groups: np.ndarray=None) -> np
         weights = zeros((length, 1))
         if groups is not None:
             group_ids = groupby(groups)
-            for current_index in group_ids:
+            for current_index in group_ids.values():
                 current_ordering = neg_er[current_index].argsort()
                 current_ordering.shape = -1, 1
                 use_rank = int(percent * len(current_index))
@@ -37,7 +37,7 @@ def percent_build(er: np.ndarray, percent: float, groups: np.ndarray=None) -> np
 
         if groups is not None:
             group_ids = groupby(groups)
-            for current_index in group_ids:
+            for current_index in group_ids.values():
                 current_ordering = neg_er[current_index].argsort(axis=0)
                 use_rank = int(percent * len(current_index))
                 set_value(weights, current_index[current_ordering[:use_rank]], 1)

alphamind/portfolio/rankbuilder.py

@@ -21,7 +21,7 @@ def rank_build(er: np.ndarray, use_rank: int, groups: np.ndarray=None) -> np.nda
         weights = zeros((length, 1))
         if groups is not None:
             group_ids = groupby(groups)
-            for current_index in group_ids:
+            for current_index in group_ids.values():
                 current_ordering = neg_er[current_index].argsort()
                 current_ordering.shape = -1, 1
                 set_value(weights, current_index[current_ordering[:use_rank]], 1.)
@@ -35,7 +35,7 @@ def rank_build(er: np.ndarray, use_rank: int, groups: np.ndarray=None) -> np.nda
 
         if groups is not None:
             group_ids = groupby(groups)
-            for current_index in group_ids:
+            for current_index in group_ids.values():
                 current_ordering = neg_er[current_index].argsort(axis=0)
                 set_value(weights, current_index[current_ordering[:use_rank]], 1)
         else: