[ Python 常見問題 ] Pandas - How to find a minimum value in a pandas dataframe column ?

 Source From Here

Question
Examples of how to find a minimum value in a pandas dataframe column.

HowTo

Create a dataframe
Lets create for example a simple data frame:

view plaincopy to clipboardprint?

1. import pandas as pd  
2.   
3. data = {'Name':['Ben','Anna','Zoe','Tom','John','Steve','Becky','Bob'],   
4.         'Age':[36,27,20,12,30,20,22,21]}  
5.   
6. df = pd.DataFrame(data)  
7. df  

Output:

view plaincopy to clipboardprint?

1.     Name  Age  
2. 0    Ben   36  
3. 1   Anna   27  
4. 2    Zoe   20  
5. 3    Tom   12  
6. 4   John   30  
7. 5  Steve   20  
8. 6  Becky   22  
9. 7    Bob   21  

Find the min value in the column Age
To find the minimum value in the column Age, a solution is to use the pandas function min:

view plaincopy to clipboardprint?

1. df['Age'].min()  # Output: 12  

Find the index corresponding to the min value in the column Age
It is also possible to find the index corresponding to the min value in the column Age using the pandas function called idxmin:

view plaincopy to clipboardprint?

1. df['Age'].idxmin()  # Output: 3  

Then using the index above:

view plaincopy to clipboardprint?

1. df.iloc[3,:]  

Output:

view plaincopy to clipboardprint?

1. Name    Tom  
2. Age      12  
3. Name: 3, dtype: object  

An example with multiple rows with a min value in the same column
Lets create a dataframe with two min values in the column Age:

view plaincopy to clipboardprint?

1. import pandas as pd  
2.   
3. data = {'Name':['Ben','Anna','Zoe','Tom','John','Steve','Becky','Bob'],   
4.         'Age':[12,27,20,12,30,20,22,21]}  
5.   
6. df = pd.DataFrame(data)  
7.   
8. print(df)  

Output:

view plaincopy to clipboardprint?

1.     Name  Age  
2. 0    Ben   12  
3. 1   Anna   27  
4. 2    Zoe   20  
5. 3    Tom   12  
6. 4   John   30  
7. 5  Steve   20  
8. 6  Becky   22  
9. 7    Bob   21  

Then the function min:

view plaincopy to clipboardprint?

1. df['Age'].min()  # Output: 12  

however idxmin:

view plaincopy to clipboardprint?

1. df['Age'].idxmin()  # Output: 0  

To get rows with a min value in the column Age a solution is to do:

view plaincopy to clipboardprint?

1. df[ df['Age'] == df['Age'].min() ]  

Output:

view plaincopy to clipboardprint?

1.   Name  Age  
2. 0  Ben   12  
3. 3  Tom   12  

and to get the indexes:

view plaincopy to clipboardprint?

1. df[ df['Age'] == df['Age'].min() ].index  

which return:

view plaincopy to clipboardprint?

1. Int64Index([0, 3], dtype='int64')  

[ Python 常見問題 ] Pandas - Set value for particular cell in pandas DataFrame using index

 Source From Here

Question
I've created a Pandas DataFrame

view plaincopy to clipboardprint?

1. import pandas as pd  
2.   
3. df = pd.DataFrame(index=['A','B','C'], columns=['x','y'])  
4. df  

Output:

view plaincopy to clipboardprint?

1.     x    y  
2. A  NaN  NaN  
3. B  NaN  NaN  
4. C  NaN  NaN  

Then I want to assign value to particular cell, for example for row 'C' and column 'x'. I've expected to get such result:Any suggestions?

view plaincopy to clipboardprint?

1.     x    y  
2. A  NaN  NaN  
3. B  NaN  NaN  
4. C  10  NaN  

Any suggestions?

HowTo
Going forward, the recommended method is .iat/.at. So the recommended approach is:

[ Python 常見問題 ] Pandas - Groupby: Count and mean combined

 Source From Here

Question
Working with PANDAS to try and summarise a dataframe as a count of certain categories, as well as the means sentiment score for these categories. There is table full of strings which 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) dataframe looks like this:

view plaincopy to clipboardprint?

1. import pandas as pd  
2. import numpy as np  
3.   
4. df = pd.DataFrame(data=[  
5.     ['bar', 'some string', 0.13],  
6.     ['foo', 'alt string',  -0.8],  
7.     ['bar', 'another str',  0.7],  
8.     ['foo', 'some text',   -0.2],  
9.     ['foo', 'more text',   -0.5]],  
10.     columns=['source', 'text', 'sent']  
11. )  

My expected output will look like this:

view plaincopy to clipboardprint?

1. source    count     mean_sent  
2. -----------------------------  
3. foo       3         -0.5  
4. bar       2         0.415  

HowTo
You can use groupby with aggregate:

view plaincopy to clipboardprint?

1. df.groupby('source') \  
2.        .agg({'text':'size', 'sent':'mean'}) \  
3.        .rename(columns={'text':'count','sent':'mean_sent'}) \  
4.        .reset_index()