NLP pre-processing - Remove unicode chars from text

fix_unicode.py

import glob
import pandas as pd

files = glob.glob('out-*.json')

def remove_unicode_char(file_name):
    f = open(file_name, 'rb').read()
    with open(file_name, 'w') as nf:
        nf.write(f.decode(encoding="ascii", errors="ignore"))
        print ("=> ", file_name)

def json_to_csv(file_name):
    df = pd.read_json(file_name)
    print ("=> ", file_name)
    df.to_csv("".join(file_name.split(".")[:-1]) + ".csv")