preprocessing.py 文件源码

def get_journal_features(cacheddatapath, rawdatapath):
    try:
        with gzip.open(os.path.join(cacheddatapath, "journal_features.pklz"), "rb") as infile:
            journ_raw = pickle.load(infile)
    except:
        df = get_preprocessed_df(cacheddatapath, rawdatapath)
        journ_raw = make_journal_features(df)
        with gzip.open(os.path.join(cacheddatapath, "journal_features.pklz"), "wb") as outfile:
            pickle.dump(journ_raw, outfile, protocol=4)
    return journ_raw