def foo(self, actual, prediction): _, _, gt_data_for_sum = create_target_table(self.sub_gcd_res, actual) _, _, actual_data_for_sum = create_target_table(self.sub_gcd_res, prediction[:, 1]) all_accuracies = dict([ [rep, accuracy_by_repetition(actual_data_for_sum, gt_data_for_sum, number_of_repetition=rep)] for rep in range(1,11)]) print ", ".join([ "acc {}:{}".format(k, v) for k, v in all_accuracies.iteritems()]) return all_accuracies
# plt.show() auc_score = roc_auc_score(test_target_gcd, test_prediction[:, 1]) print "auc_score:{0}".format(auc_score) sub_gcd_res = create_data_for_compare_by_repetition(file_name) # sub_gcd_res = dict(train_trial=gcd_res['train_trial'][gcd_res['train_mode'] != 1], # train_block=gcd_res['train_block'][gcd_res['train_mode'] != 1], # stimulus=gcd_res['stimulus'][gcd_res['train_mode'] != 1]) _, _, gt_data_for_sum = create_target_table(sub_gcd_res, test_target_gcd) _, _, actual_data_for_sum = create_target_table( sub_gcd_res, test_prediction[:, 1]) print "accuracy_by_repetition {0}".format( accuracy_by_repetition(actual_data_for_sum, gt_data_for_sum, number_of_repetition=10)) results.append( dict(subject_name=subject_name, test_prediction=test_prediction, auc_score=auc_score)) break # In[4]: import keras # In[20]: # In[73]: