In [6]:
import pandas as pd
df = pd.read_csv("../dataset/git_demo_timestamp_linux.csv", sep=";")
df.author.value_counts().head(10)
Out[6]:
In [8]:
%matplotlib inline
df.author.value_counts().head(10).plot(kind='pie')
Out[8]:
In [10]:
df['timestamp_local'] = pd.to_datetime(df['timestamp_local'])
df.describe()
Out[10]:
In [13]:
df.timestamp_local.dt.year.value_counts(sort=False).plot(kind='bar')
Out[13]:
In [15]:
df.timestamp_local.dt.hour.value_counts(sort=False).plot(kind='bar')
Out[15]: