def trellis_plot_histogram():
""" Trellis Plot arranges data in a rectangular grid by values of certain attributes using a histogram """
plot =rplot.RPlot(TIPS_DATA, x='total_bill', y='tip')
plot.add(rplot.TrellisGrid(['sex', 'smoker']))
plot.add(rplot.GeomHistogram())
plot.render(plt.gcf())
def cond_hists(df, plot_cols, grid_col):
import matplotlib.pyplot as plt
import pandas.tools.rplot as rplot
## Loop over the list of columns
for col in plot_cols:
## Define figure
fig = plt.figure(figsize=(14, 4))
fig.clf()
ax = fig.gca()
## Setup plot and grid and plot the data
plot = rplot.RPlot(df, x = col,
y = '.')
plot.add(rplot.TrellisGrid(['.', grid_col]))
plot.add(rplot.GeomHistogram())
ax.set_title('Histograms of ' + col + ' conditioned by ' + grid_col + '\n')
plot.render()
return grid_col
def main():
projects_all = pd.read_csv(open('../../dataset/projects.csv', 'r'))
outcomes = pd.read_csv(open('../../dataset/outcomes.csv', 'r'))
projects = pd.merge(projects_all, outcomes, on='projectid')
projects1 = pd.DataFrame(
projects.fillna(''),
columns=['projectid', 'primary_focus_area', 'is_exciting'])
projects1['cat_primary_focus_area'] = pd.factorize(
projects1.primary_focus_area)[0]
print(projects1)
plt.figure()
plot = rplot.RPlot(
projects1,
x='cat_primary_focus_area',
)
plot.add(rplot.TrellisGrid(['.', 'is_exciting']))
plot.add(rplot.GeomHistogram())
plot.render(plt.gcf())
pylab.show()
plt.pyplot.pie(tweetdf.tweet_type.value_counts())
crosstabs = pd.crosstab(tweetdf.tweet_day, tweetdf.tweet_type)
crosstabs.plot(kind='bar', stacked=True)
df['col'].hist(bins=25)
#splom - diagonal can be 'hist'
pd.scatter_matrix(trans_data, diagonal = 'kde', color = 'k', alpha=0.3)
#trellis
tips_data = pd.read_csv('tips.csv')
import pandas.tools.rplot as rplot
plt.figure()
plot = rplot.RPlot(tips_data, x='total_bill', y='tip')
plot.add(rplot.TrellisGrid(['sex', 'smoker']))
plot.add(rplot.GeomHistogram())
plot.render(plt.gcf())
#---------------------------------------------------------------------------
# Stats - from scipy
#some useful functions
np.random.randn(4, 3)
years = range(1880,2011)
shape()
#check that series values are close to but not exactly 1
np.allclose(x, 1)