Hi,
I have below df and want to move multiple columns to initial position.
df:
wat tey step low tool tech category subj name
1 2 3 4 5 6 7 8 9
6 2 7 4 8 6 7 8 9
name subj tool category wat tey step low tech
9 8 5 7 1 2 3 4 6
9 8 8 7 6 2 7 4 6
I use below code:
first_column = ['name',subj','tool','category ']
df_out = df[first_column].shift[-1]
But it didn't work. Can anyone please suggest how to achive that. first_column, other column order does not matter.
I don't think shift does what you think it does. You are creating a dataframe that only contains 'name', subj', 'tool' and 'category and shifting the rows up 1 so the new dataframe still has 2 rows, but the bottom row is NaN.
You cancreate a new dataframe from the old with all the columns rearranged.
import pandas as pd
df = pd.DataFrame({"A":[1, 2, 3], "B":[4, 5, 6], "C":[7, 8, 9]})
print(df)
df = df[["C", "A", "B"]]
print(df)Output:
A B C
0 1 4 7
1 2 5 8
2 3 6 9
C A B
0 7 1 4
1 8 2 5
2 9 3 6
You can use pop() and insert() to modify the existing dataframe.
import pandas as pd
df = pd.DataFrame({"A":[1, 2, 3], "B":[4, 5, 6], "C":[7, 8, 9]})
print(df)
column = df.pop("C")
df.insert(0, "C", column)
print(df)Output:
A B C
0 1 4 7
1 2 5 8
2 3 6 9
C A B
0 7 1 4
1 8 2 5
2 9 3 6
I do pop, but it only allow one column at a time. Can we do pop() for multiple columns?
(Jul-02-2022, 03:02 AM)deanhystad Wrote: [ -> ]I don't think shift does what you think it does. You are creating a dataframe that only contains 'name', subj', 'tool' and 'category and shifting the rows up 1 so the new dataframe still has 2 rows, but the bottom row is NaN.
You cancreate a new dataframe from the old with all the columns rearranged.
import pandas as pd
df = pd.DataFrame({"A":[1, 2, 3], "B":[4, 5, 6], "C":[7, 8, 9]})
print(df)
df = df[["C", "A", "B"]]
print(df)Output:
A B C
0 1 4 7
1 2 5 8
2 3 6 9
C A B
0 7 1 4
1 8 2 5
2 9 3 6
You can use pop() and insert() to modify the existing dataframe.
import pandas as pd
df = pd.DataFrame({"A":[1, 2, 3], "B":[4, 5, 6], "C":[7, 8, 9]})
print(df)
column = df.pop("C")
df.insert(0, "C", column)
print(df)Output:
A B C
0 1 4 7
1 2 5 8
2 3 6 9
C A B
0 7 1 4
1 8 2 5
2 9 3 6
I also tried create new data frame, but the problem is I do not now some time coulmns are more and some time less. but the columns I want to shift lways exists. Other columns are random.
(Jul-02-2022, 03:02 AM)deanhystad Wrote: [ -> ]I don't think shift does what you think it does. You are creating a dataframe that only contains 'name', subj', 'tool' and 'category and shifting the rows up 1 so the new dataframe still has 2 rows, but the bottom row is NaN.
You cancreate a new dataframe from the old with all the columns rearranged.
import pandas as pd
df = pd.DataFrame({"A":[1, 2, 3], "B":[4, 5, 6], "C":[7, 8, 9]})
print(df)
df = df[["C", "A", "B"]]
print(df)Output:
A B C
0 1 4 7
1 2 5 8
2 3 6 9
C A B
0 7 1 4
1 8 2 5
2 9 3 6
You can use pop() and insert() to modify the existing dataframe.
import pandas as pd
df = pd.DataFrame({"A":[1, 2, 3], "B":[4, 5, 6], "C":[7, 8, 9]})
print(df)
column = df.pop("C")
df.insert(0, "C", column)
print(df)Output:
A B C
0 1 4 7
1 2 5 8
2 3 6 9
C A B
0 7 1 4
1 8 2 5
2 9 3 6
The documentation for pop says the "item" is label of the column to be popped, not a list of labels. Trying to pop a list results in an index error.
https://pandas.pydata.org/pandas-docs/st...e.pop.html
But you can pop multiple items one at a time.
import pandas as pd
df = pd.DataFrame({
"A":[1, 2, 3],
"B":[4, 5, 6],
"C":[7, 8, 9],
"D":[10, 11, 12]})
for label in ["B", "D"]:
df.insert(0, label, df.pop(label))
print(df)Output:
D B A C
0 10 4 1 7
1 11 5 2 8
2 12 6 3 9
DataFrame.columns() returns an iterator that you can use to get all the column labels. Here I use this to create a reorganized column list.
import pandas as pd
df = pd.DataFrame({
"A":[1, 2, 3],
"B":[4, 5, 6],
"C":[7, 8, 9],
"D":[10, 11, 12]})
labels = ["D", "B"]
extras = [col for col in df.columns if col not in labels]
df = df[labels + extras]
print(df)Output:
D B A C
0 10 4 1 7
1 11 5 2 8
2 12 6 3 9