#The features identified for removal can be accessed through the ops dictionary of the FeatureSelector object. missing_features = fs.ops['missing'] print(missing_features[:20]) fs.plot_missing() #在每一个画图的后面加上plt.show即可 plt.show() print(fs.missing_stats.head(20)) # 2 Single Unique Value fs.identify_single_unique() single_unique = fs.ops['single_unique'] print(single_unique) fs.plot_unique() #画图都不好用 plt.show() print(fs.unique_stats.sample(5)) # 3 Collinear (highly correlated) Feature fs.identify_collinear(correlation_threshold=0.975) correlated_features = fs.ops['collinear'] correlated_features[:5] fs.plot_collinear() plt.show() fs.plot_collinear(plot_all=True) plt.show()
if TEST_FEATURE_SELECTION:
fs.identify_missing(missing_threshold=0.6)
missing_features = fs.ops['missing']
missing_features[:10]
fs.plot_missing()
fs.missing_stats.head(10)
fs.identify_single_unique()
single_unique = fs.ops['single_unique']
single_unique
fs.plot_unique()
fs.identify_collinear(correlation_threshold=0.975)
correlated_features = fs.ops['collinear']
correlated_features[:5]
fs.plot_collinear()
fs.record_collinear.head()
fs.identify_zero_importance(task = 'classification', eval_metric = 'auc',
n_iterations = 10, early_stopping = True)
one_hot_features = fs.one_hot_features
base_features = fs.base_features
print('There are %d original features' % len(base_features))
print('There are %d one-hot features' % len(one_hot_features))
from feature_selector import FeatureSelector
# Features are in train and labels are in train_labels
fs = FeatureSelector(data=train, labels=train_labels)
#缺失值统计
fs.identify_missing(0.5)
df_miss_value = fs.missing_stats.sort_values('missing_fraction',
ascending=False)
print('df_miss_value', df_miss_value.head(15))
missing_features = fs.ops['missing']
print('missing_features to remove', missing_features[:20])
#单值特征统计
fs.identify_single_unique()
print('fs.plot_unique()', fs.plot_unique())
fs.identify_collinear(0.95)
print('plot_collinear()', fs.plot_collinear())
# list of collinear features to remove
collinear_features = fs.ops['collinear']
print('collinear_features', collinear_features)
# dataframe of collinear features
df_collinear_features = fs.record_collinear.sort_values('corr_value',
ascending=False)
print('df_collinear_features', df_collinear_features.head(50))
#零重要度特征统计
# Pass in the appropriate parameters