数据分析S104：分析用户发言在每小时每天等粒度上的分布

python

[python]
def s104(usrid=None, name=None):
if not usrid:
usrid=common.get_id(name)
conn = sqlite3.connect(r’resource/chat.db’)
df =pd.read_sql(‘SELECT * FROM chatdb WHERE usrid =?’,conn,params=(usrid,),parse_dates=[‘date’],index_col=[‘date’])

# print(df[['usrid','name'],['name']].groupby(df.usrid).count().head(5))
df.insert(0,'count',df['id'])
df.groupby(lambda x:x.dayofweek).count()[['count']].plot(kind='bar')
df[['count']].resample("M", how="count").plot();
df2 =pd.read_sql('SELECT * FROM chatdb WHERE usrid =?',conn,params=(usrid,),parse_dates=['time'],index_col=['time'])
print(df2)
df2.insert(0,'count',df2['id'])
df2.groupby(lambda x:x.hour).count()[['count']].plot(kind='bar')
# df['count'].resample("M", how="count").plot(lw=2);
# df[['count']].groupby(df.date).count().plot(kind='bar')
plt.show()

[/python]