further enhance

alphamind/benchmarks/benchmarks.py

@@ -16,9 +16,9 @@ from alphamind.benchmarks.portfolio.rankbuild import benchmark_build_rank_with_g
 
 if __name__ == '__main__':
 
-    benchmark_neutralize(3000, 10, 1000)
-    benchmark_neutralize(30, 10, 50000)
-    benchmark_neutralize(50000, 50, 20)
+    # benchmark_neutralize(3000, 10, 1000)
+    # benchmark_neutralize(30, 10, 50000)
+    # benchmark_neutralize(50000, 50, 20)
     # benchmark_standardize(3000, 10, 1000)
     # benchmark_standardize_with_group(3000, 10, 1000, 30)
     # benchmark_standardize(30, 10, 50000)
@@ -31,9 +31,9 @@ if __name__ == '__main__':
     # benchmark_winsorize_normal_with_group(30, 10, 5000, 5)
     # benchmark_winsorize_normal(50000, 50, 20)
     # benchmark_winsorize_normal_with_group(50000, 50, 20, 50)
-    # benchmark_build_rank(3000, 1000, 300)
-    # benchmark_build_rank_with_group(3000, 1000, 10, 30)
-    # benchmark_build_rank(30, 50000, 3)
-    # benchmark_build_rank_with_group(30, 50000, 1, 3)
-    # benchmark_build_rank(50000, 20, 3000)
-    # benchmark_build_rank_with_group(50000, 20, 10, 300)
+    benchmark_build_rank(3000, 1000, 300)
+    benchmark_build_rank_with_group(3000, 1000, 10, 30)
+    benchmark_build_rank(30, 50000, 3)
+    benchmark_build_rank_with_group(30, 50000, 1, 3)
+    benchmark_build_rank(50000, 20, 3000)
+    benchmark_build_rank_with_group(50000, 20, 10, 300)

alphamind/portfolio/rankbuilder.py

@@ -7,28 +7,27 @@ Created on 2017-4-26
 
 import numpy as np
 from numpy import zeros
-from numpy import max
 
 
 def rank_build(er: np.ndarray, use_rank: int, groups: np.ndarray=None) -> np.ndarray:
+    length = len(er)
     neg_er = -er
-    masks = zeros(len(er), dtype=bool)
-    ordering = neg_er.argsort()
+    masks = zeros(length, dtype=bool)
+    weights = zeros(length)
 
     if groups is not None:
-        max_g = max(groups)
-        index_range = np.arange(len(er))
-
-        for i in range(max_g + 1):
+        max_g = groups.max()
+        index_range = np.arange(length)
+        for i in range(max_g+1):
             current_mask = groups == i
             current_index = index_range[current_mask]
             current_ordering = neg_er[current_mask].argsort()
             masks[current_index[current_ordering[:use_rank]]] = True
+        weights[masks] = 1. / masks.sum()
     else:
+        ordering = neg_er.argsort()
         masks[ordering[:use_rank]] = True
-
-    weights = zeros(len(er))
-    weights[masks] = 1. / use_rank
+        weights[masks] = 1. / use_rank
     return weights
 
 
@@ -36,12 +35,12 @@ if __name__ == '__main__':
 
     import datetime as dt
 
-    x = np.random.randn(3000)
+    x = np.random.randn(4)
 
-    groups = np.random.randint(30, size=3000)
+    groups = np.random.randint(2, size=4)
 
     start = dt.datetime.now()
     for i in range(10000):
-        weights = rank_build(x, 30, groups)
+        weights = rank_build(x, 1, groups)
     print(dt.datetime.now() - start)