preprocessing.py 文件源码

def get_raw(filename):
    with open(filename) as infile:
        raw = infile.read()
        # the next line needs rewriting as soon as the zenodo-dump conforms to 'records'-format
        # [{k:v}, {k:v},...]
        rawfacts = pd.read_json('[%s]' % ','.join(raw.splitlines()), orient='records')
    return rawfacts


### functions for ingesting from CProject



### functions for preprocessing