In [ ]:

    

%pylab inline
import json

def filternan(x):
    return x[isnan(x) != True]


    

In [1]:

    

with open("out.json") as f:
  rows = [json.loads(row) for row in f]

keys = set()
for row in rows: keys.update(set(row.keys()))
    
d = zeros(len(rows), dtype=[(str(name), 'f4') for name in keys])

for idx, row in enumerate(rows):
    for key in keys:
        d[idx][key] = nan
        try:
            d[idx][key] = float(row[key])
        except:
            pass


    

    




Populating the interactive namespace from numpy and matplotlib

In [27]:

    

histfig = figure(figsize=(20,5))
subplot = histfig.add_subplot(111)
subplot.hist(d['measure_new_score'], bins=200, normed=False, color='b', alpha=0.5, label="fishy score")
subplot.hist(d['measure_coursestddev'], bins=200, normed=False, color='r', alpha=0.5, label="course std")
subplot.hist(d['measure_speedstddev'], bins=200, normed=False, color='g', alpha=0.5, label="speed std")
subplot.hist(d['measure_speedavg'], bins=200, normed=False, color='y', alpha=0.5, label="speed avg")
show()


    

In [37]:

    

histfig = figure(figsize=(20,5))
subplot = histfig.add_subplot(111)
subplot.hist(filternan(d['speed']), range=(0, 25), bins=200, normed=False, color='b', alpha=0.5, label="raw momentary speed")
legend()
show()


    

In [42]:

    

histfig = figure(figsize=(20,5))
subplot = histfig.add_subplot(111)
subplot.hist(filternan(d['measure_speed']), bins=100, normed=False, color='b', alpha=0.5, label="normalized momentary speed")
legend()
show()


    

In [41]:

    

histfig = figure(figsize=(20,5))
subplot = histfig.add_subplot(111)
subplot.hist(filternan(d['measure_speedavg']), bins=200, normed=False, color='b', alpha=0.5, label="measure_speedavg")
legend()
show()


    

In [ ]: