In [2]:
%matplotlib inline
import pandas as pd
In [3]:
from IPython.core.display import HTML
css = open('style-table.css').read() + open('style-notebook.css').read()
HTML('<style>{}</style>'.format(css))
Out[3]:
In [4]:
cast = pd.DataFrame.from_csv('data/cast.csv', index_col=None)
cast.head()
Out[4]:
In [12]:
release_dates = pd.DataFrame.from_csv('data/release_dates.csv', index_col=None,
parse_dates=['date'], infer_datetime_format=True)
release_dates.head()
Out[12]:
In [ ]:
In [23]:
r_d = release_dates[(release_dates.title.str.contains("Christmas")) & (release_dates.country == "USA")]
r_d.date.dt.month.value_counts().sort_index().plot(kind="bar")
Out[23]:
In [ ]:
In [27]:
r_d = release_dates[(release_dates.title.str.contains("The Hobbit")) & (release_dates.country == "USA")]
r_d
Out[27]:
In [25]:
r_d.date.dt.month.value_counts().sort_index().plot(kind="bar")
Out[25]:
In [30]:
r_d = release_dates[(release_dates.title.str.contains("Romance")) ]
r_d
Out[30]:
In [29]:
r_d.date.dt.dayofweek.value_counts().sort_index().plot(kind="bar")
Out[29]:
In [32]:
r_d = release_dates[(release_dates.title.str.contains("Action")) ]
r_d
Out[32]:
In [33]:
r_d.date.dt.dayofweek.value_counts().sort_index().plot(kind="bar")
Out[33]:
In [34]:
usa = release_dates[release_dates.country == 'USA']
c = cast
c = c[c.name == 'Judi Dench']
c = c[c.year // 10 * 10 == 1990]
c.merge(usa).sort('date')
Out[34]:
In [ ]:
In [35]:
c = cast
c = c[c.name == 'Judi Dench']
m = c.merge(usa).sort('date')
m.date.dt.month.value_counts().sort_index().plot(kind='bar')
Out[35]:
In [36]:
c = cast
c = c[c.name == 'Tom Cruise']
m = c.merge(usa).sort('date')
m.date.dt.month.value_counts().sort_index().plot(kind='bar')
Out[36]:
In [ ]:
In [ ]: