for key, value in datafiles.items():
if key in list(ubuntus['Filename']):
plt.hist([size for size in value['st_size'] if size > 1], bins, normed=True, log=True, alpha=.5, label=key);
plt.xlabel('File size')
plt.ylabel('Frequency (logarithmic)')
plt.title('Distribution of file sizes on Ubuntu')
plt.show()
# <codecell>
# try and draw Seaborn KDE plots
with sns.palette_context("husl"):
f, (ax1, ax2) = plt.subplots(2, 1, figsize=(8, 20), sharex=False)
for key, value in datafiles.items():
if key in list(macs['Filename']):
path = key.replace('-', '/').split('.')[0]
ax1 = sns.kdeplot(double([size for size in value['st_size'] if size > 1]), shade=True, label=path, legend=True, ax=ax1)
ax1.set_xscale('log')
for key, value in datafiles.items():
if key in list(ubuntus['Filename']):
path = key.replace('-', '/').split('.')[0]
ax2 = sns.kdeplot(double([size for size in value['st_size'] if size > 1]), shade=True, label=path, legend=True, ax=ax2)
from scipy import stats
from pylab import *
import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline
# <codecell>
# Plot the relationship between time and number of data items
consumers = [2,4,6,8,10,12]
with sns.palette_context('husl'):
plt.figure(figsize=(8,8))
ax = plt.gca()
for count in consumers:
segment = means.loc[count].reset_index()
segment.plot(y='time', x='data', label='{0} consumers'.format(count))
plt.xlabel('Number of data items')
plt.ylabel('Time (seconds)')
plt.title('Time vs. Number of Data Items for Varying Numbers of Consumers')
plt.legend()
plt.savefig(figpath + 'timevsdatas.pdf')
plt.show()
