Pandas Multiple Column by Floating List Scalars

Question

Suppose i have a loop structure that looks like the following:

numlist = [0.1, 0.25, 0.5, 0.75, 0.90, 1.1, 1.25, 1.5]
numlist = np.array(numlist,dtype=float)
apcnt = np.zeros((4,8))
for x in range(5):
   for y in numlist:
     apcnt[x][y] = a.iloc[x][0]*numlist[y]
     print(apcnt)

Dataframe "a" looks like this:

           adjusted_power
YearMonth                
Q1           19388.321839
Q2           13435.865517
Q3           12579.123793
Q4           19238.458966

and I'm trying to obtain a final answer that looks like this:

YearMonth
Q1           19388*0.1   19388*0.25  19388*0.50  19388*0.75  19388*0.9   19388*1.1  19388*1.25  
19388*1.5
Q2           13436(same as above for Q1 example)
Q3           12579(same as above for Q1 example)
Q4           19238(same as above for Q1 example)

I get an index error in this form ( thank you for your help):

File "<ipython-input-40-e72f79d55da3>", line 7, in <module>
apcnt[x][y] = a.iloc[x][0]*numlist[y]

IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or 
boolean arrays are valid indices

score 2 · Accepted Answer · answered Jun 18 '20 at 00:01

We can do multiply.outer

pd.DataFrame(np.multiply.outer(a['adjusted_power'].values,numlist),index=a.index)
                     0            1  ...             6             7
YearMonth                            ...                            
Q1         1938.832184  4847.080460  ...  24235.402299  29082.482759
Q2         1343.586552  3358.966379  ...  16794.831896  20153.798275
Q3         1257.912379  3144.780948  ...  15723.904741  18868.685690
Q4         1923.845897  4809.614741  ...  24048.073707  28857.688449
[4 rows x 8 columns]

score 1 · Answer 2 · answered Jun 17 '20 at 23:50

1

You can use broadcasting:

out = pd.DataFrame(a.adjusted_power.values[:,None] * numlist,
                   index=a.index)

Output:

          0        1        2         3        4        5        6        7
--  -------  -------  -------  --------  -------  -------  -------  -------
Q1  1938.83  4847.08  9694.16  14541.2   17449.5  21327.2  24235.4  29082.5
Q2  1343.59  3358.97  6717.93  10076.9   12092.3  14779.5  16794.8  20153.8
Q3  1257.91  3144.78  6289.56   9434.34  11321.2  13837    15723.9  18868.7
Q4  1923.85  4809.61  9619.23  14428.8   17314.6  21162.3  24048.1  28857.7

answered Jun 17 '20 at 23:50

Quang Hoang

146,074
10
56
74

hi -thanks, it works great!! what is the meaning of the [: , None] after "values"? – user2100039 Jun 17 '20 at 23:58
Basically `arr = np.array([1,2,3])[:,None]` gives you `[[1],[2],[3]]` which then allows `arr * [4,5]`. More details [here](https://numpy.org/doc/stable/user/basics.broadcasting.html). you can also do `np.multiply.outer([1,2,3], [4,5])` – Quang Hoang Jun 18 '20 at 00:00

Pandas Multiple Column by Floating List Scalars

2 Answers2