Comparison of loading JSON into pandas dataframe

json_pd_test.py

import pandas as pd
from time import time
from pprint import pprint
import simplejson
import json
import ujson


def _loads(data, serializer):
    t_start = time()
    serialized_data = serializer.loads(data)
    t_serialization = time() - t_start
    df = pd.DataFrame.from_dict(serialized_data)
    t_end = time() - t_start
    return {
        "duration": t_end,
        "columns": df.columns,
        "shape": df.shape,
        "serialization": t_serialization
    }

with open("big.json", "r") as infile:
    data = infile.read()

for method in [simplejson, json, ujson]:
    result = _loads(data, method)
    print "method: " + method.__name__
    print "total: " + str(result["duration"])
    print "serialization: " + str(result["serialization"])