python字典到大 pandas 数据帧与多列
问题描述：
我有以下的python字典：
I have the following python dictionary:
d= {'data' : Counter({ 'important' : 2,
'very' : 3}),
'analytics' : Counter({ 'boring' : 5,
'sleep' : 3})
}
我想以下列格式将其保存为大熊猫数据框：
I want to save it as a pandas dataframe in the following format:
category | word | count
data | important | 2
data | very | 3
analytics | boring | 5
analytics | sleep | 3
我尝试了以下操作，但没有一个工作
I tried the following, but none of it worked
df = pd.DataFrame(d.items())
df = pd.DataFrame.from_dict(d, orient='index').reset_index()
df = pd.DataFrame(data)
答
您可以使用 stack ：
You can use stack:
df = pd.DataFrame(d).stack().reset_index()
df.columns = ['word','category','count']
print(df)
word category count
0 boring analytics 5.0
1 important data 2.0
2 sleep analytics 3.0
3 very data 3.0
df = pd.DataFrame.from_dict(d, orient='index').stack().reset_index()
df.columns = ['category','word','count']
print(df)
category word count
0 analytics boring 5.0
1 analytics sleep 3.0
2 data important 2.0
3 data very 3.0
嵌套列表理解的另一个解决方案：
Another solution with nested list comprehension:
df = pd.DataFrame([(key,key1,val1) for key,val in d.items() for key1,val1 in val.items()])
df.columns = ['category','word','count']
print(df)
category word count
0 analytics boring 5
1 analytics sleep 3
2 data important 2
3 data very 3