In [ ]:

    




    

In [ ]:

    

import pandas as pd

comments = pd.read_csv('jung/jung.csv')
# comments['date'] = comments['date'].apply(lambda d: datetime.datetime.strptime(d, '%Y-%m-%d').date())
comments['date'] = pd.to_datetime(comments['date'])
score_comments = comments[['date', 'score']].sort(['date'])
# print(score_comments)

score_comments = score_comments.set_index('date')
gs = score_comments.groupby(lambda x:x.week)
# print(gs.head())
summary_comment = 
pd.concat([gs.std(), gs.mean(), gs.count()],axis=1, keys=['std', 'mean', 'count'])
# gs.mean().plot()
# gs.std().plot()


    

In [2]:

    

import pandas as pd
search_result = pd.read_csv('jung/jung_search.csv')
# print(df.head())
search_result['date'] = pd.to_datetime(search_result['date'])

search_result = search_result.set_index('date')
search_result = search_result.groupby(lambda x:x.week)
search_result = search_result.count()
print(type(search_result))
# dramaid = 63553
# ratings = df[df['dramaid']==dramaid].sort(['date'])[['date','ratings']]
# ratings = ratings.set_index('date')
# ratings = ratings.groupby(lambda x:x.week)
# ratings = ratings.aggregate(np.sum)

# # ratings.join(summary_comment, how='outer')
# ratings = ratings.join(summary_comment)
# print(ratings.corr())
# ratings.plot()


    

    




<class 'pandas.core.frame.DataFrame'>

In [30]:

    

import pandas as pd
df = pd.read_csv('jung/ratings.csv')
# print(df.head())
df['date'] = pd.to_datetime(df['date'])
dramaid = 63553
ratings = df[df['dramaid']==dramaid].sort(['date'])[['date','ratings']]
ratings = ratings.set_index('date')
ratings = ratings.groupby(lambda x:x.week)
ratings = ratings.aggregate(np.sum)

# ratings.join(summary_comment, how='outer')
ratings = ratings.join(summary_comment)
# ratings = ratings.join(search_result)
# print(ratings)
ratings2 = ratings.copy()
ratings['Y'] = ratings['ratings'].shift(-1) - ratings['ratings']
print(ratings)

# print(ratings2.corr())
# print(ratings.corr())
# ratings2['ratings']

# ratings.plot()


    

    




    ratings  (std, score)  (mean, score)  (count, score)    Y
6      13.2      3.248012       8.257143              35  2.4
7      15.6      2.282902       9.214286              28 -1.2
8      14.4      2.395785       9.161290              31  1.5
10     15.9      1.479611       9.548387              31  0.6
11     16.5      0.809760       9.742424              66 -0.8
12     15.7      2.490951       9.105263              38 -0.4
13     15.3      1.805547       9.480000              25  2.4
14     17.7      2.156235       9.428571              56 -0.6
15     17.1      3.092588       8.888889              27  0.8
17     17.9      3.597221       8.240000              25 -0.3
18     17.6      1.553417       9.518519              27  1.4
19     19.0      1.598575       9.343750              32 -0.8
20     18.2      2.115457       9.386364              44 -1.3
21     16.9      2.125928       9.512821              39 -0.3
22     16.6      0.751622       9.772727              22  1.5
23     18.1      1.320094       9.352941              17  0.5
24     18.6      0.832050       9.769231              13 -0.6
25     18.0      2.725541       8.500000               8 -0.2
26     17.8      0.701472       9.722222              36  NaN

[19 rows x 5 columns]

In [17]:

    




    

In [ ]: