Pandas Groupby：计数和平均值相结合

Question

Working with pandas to try and summarise a data frame as a count of certain categories, as well as the means sentiment score for these categories.

There is a table full of strings that have different sentiment scores, and I want to group each text source by saying how many posts they have, as well as the average sentiment of these posts.

My (simplified) data frame looks like this:

source    text              sent
--------------------------------
bar       some string       0.13
foo       alt string        -0.8
bar       another str       0.7
foo       some text         -0.2
foo       more text         -0.5

The output from this should be something like this:

source    count     mean_sent
-----------------------------
foo       3         -0.5
bar       2         0.415

The answer is somewhere along the lines of:

df['sent'].groupby(df['source']).mean()

Yet only gives each source and it's mean, with no column headers.

Answer 1

You can use groupby with aggregate :

df = df.groupby('source') \
       .agg({'text':'size', 'sent':'mean'}) \
       .rename(columns={'text':'count','sent':'mean_sent'}) \
       .reset_index()
print (df)
  source  count  mean_sent
0    bar      2      0.415
1    foo      3     -0.500

Answer 2

In newer versions of Panda you don't need the rename anymore, just use named parameters:

df = df.groupby('source') \
       .agg(count=('text', 'size'), mean_sent=('sent', 'mean')) \
       .reset_index()

print (df)
  source  count  mean_sent
0    bar      2      0.415
1    foo      3     -0.500

Answer 3

下面一个应该可以正常工作：

df[['source','sent']].groupby('source').agg(['count','mean'])

Answer 4

A shorter version to achieve this is:

df.groupby('source')['sent'].agg(count='size', mean_sent='mean').reset_index()

The nice thing about this is that you can extend it if you want to take the mean of multiple variables but only count once. In this case you will have to pass a dictionary:

df.groupby('source')['sent1', 'sent2'].agg({'count': 'size', 'means': 'mean'}).reset_index()

Answer 5

对于那些正在寻找超过两列聚合的人（就像我一样）：只需将它们添加到“agg”。

df = df.groupby(['id']).agg({'texts': 'size', 'char_num': 'mean', 'bytes': 'mean'}).reset_index()

Answer 6

我认为这应该提供您想要的输出：

result = pd.DataFrame(df.groupby('source').size())

results['mean_score'] = df.groupby('source').sent.mean()