How much unique data is there, put it all in a table - sql

Question

I would like to query in SQL how many unique values ​​there are and how many rows are there. In Python, I could do it like this. But how do I do this in SQL so that I get the result like at the bottom?
In Python I could do the following
d = {'sellerid': [1, 1, 1, 2, 2, 3, 3, 3], 'modelnumber': [85, 45, 85, 12 ,85, 74, 85, 12]
, 'modelgroup': [2, 3, 2, 1, 2, 3, 2, 1 ]}
df = pd.DataFrame(data=d)
display(df.head(10))
df['Dataframe']='df'
unique_sellerid = df['sellerid'].nunique()
print("unique_sellerid", unique_sellerid)
unique_modelnumber = df['modelnumber'].nunique()
print("unique_modelnumber", unique_modelnumber)
unique_modelgroup = df['modelgroup'].nunique()
print("unique_modelgroup", unique_modelgroup)
total_rows = df.shape[0]
print("total_rows", total_rows)
[OUT]
unique_sellerid 3
unique_modelnumber 4
unique_modelgroup 3
total_rows 8
I want a query like
Here is the dummy table
CREATE TABLE cars (
sellerid INT NOT NULL,
modelnumber INT NOT NULL,
modelgroup INT,
);
INSERT INTO cars
(sellerid , modelnumber, modelgroup )
VALUES
(1, 85, 2),
(1, 45, 3),
(1, 85, 2),
(2, 12, 1),
(2, 85, 2),
(3, 74, 3),
(3, 85, 2),
(3, 12, 1);

Answer 1

You could use the count(distinct column) aggregation function like :
select
count(distinct col1) as nunique_col1,
count(distinct col2) as nunique_col2,
count(1) as nb_rows
from database
Also in pandas, you can also apply the nunique() function on the dataset, rather than doing it on each column: df.nunique()