data.py 文件源码

def _sample_testset(self, data):
        test_sample = self.test_sample
        if not isinstance(test_sample, int):
            return data

        userid, feedback = self.fields.userid, self.fields.feedback
        if test_sample > 0:
            sampled = (data.groupby(userid, sort=False, group_keys=False)
                            .apply(random_choice, test_sample, self.random_state or np.random))
        elif test_sample < 0: #leave only the most negative feedback from user
            idx = (data.groupby(userid, sort=False)[feedback]
                        .nsmallest(-test_sample).index.get_level_values(1))
            sampled = data.loc[idx]
        else:
            sampled = data

        return sampled