DFS（枝刈り込み）による統計的識別手法の最良組み合わせ抽出

statistical_identify_dfs.ipynb

Timer unit: 1e-06 s

Total time: 0.007946 s
File:
Function: auc at line 1

Line # Hits Time Per Hit % Time Line Contents

 1                                           def auc(y, X):
 2         1         3283   3283.0     41.3      pred = np.mean(X, axis = 1)
 3         1         4498   4498.0     56.6      fpr, tpr, thresholds = metrics.roc_curve(y, pred)
 4         1          165    165.0      2.1      return metrics.auc(fpr, tpr)

To reduce total processing time by using multiprocessing

import multiprocessing as mp

def roc(v):
    """ calculate one pair, return (index, auc) """
    i, true, pred = v
    fpr, tpr, thresholds = metrics.roc_curve(true, pred, drop_intermediate=True)
    auc = metrics.auc(fpr, tpr)
    return i, auc

pool = mp.Pool(3) 
result = pool.map_async(roc, ((i, true[i], pred[i]) for i in range(2)))
pool.close()
pool.join()
print result.get()

Python is not supported tail call optimization, therforer, recursive call function should transform loop function.
my god!