learner.py 文件源码

def distributions(self, records=None):
        logging.info("Analysing %s ...", self.database)
        try:
            if records is not None and isinstance(records, pd.DataFrame):
                max_mut = np.max(records['MUT'])
                self.n_samples = records.shape[0]
            else:
                # load from file
                max_mut, self.n_samples = io.get_max_mut(self.database)

            lin = np.linspace(0, max_mut, min(self.n_samples / 15., 12))
            sets = [(0, 0)] + zip(lin[:-1], lin[1:])
            if len(sets) == 1:
                # no correction needs to be applied
                return None
            out_muts = [self.intra_donor_distance(
                records, i, j) for i, j in zip(sets, sets)]
        except StandardError as msg:
            logging.critical(msg)
            out_muts = []

        my_dict = dict()
        for f, m in out_muts:
            my_dict.setdefault(m, []).append(f)
        return my_dict