git_authorship.py 文件源码

def normalized_usage_by_package(self, package_usage_frame: pd.DataFrame,
                                    drop_package_prefix: str = None):
        scaler = MinMaxScaler()
        df = package_usage_frame.drop('package', 1)
        df_scaled = pd.DataFrame(scaler.fit_transform(df), columns=df.columns)
        if drop_package_prefix:
            df_scaled['package'] = package_usage_frame['package'].apply(
                lambda text: text[text.startswith(drop_package_prefix)
                                  and len(drop_package_prefix):])
        else:
            df_scaled['package'] = package_usage_frame['package']
        df_sorted = df_scaled.sort_values('user_count').reset_index()
        del df_sorted['index']
        return df_sorted