Click here for the imageI m trying to create a list from 3 different series which will be of the shape "({A} {B} {C})" where A denotes the 1st element from series 1, B is for 1st element from series 2, C is for 1st element from series 3 and this way it should create a list containing 600 element.
List 1 List 2 List 3
u_p0 1 v_p0 2 w_p0 7
u_p1 21 v_p1 11 w_p1 45
u_p2 32 v_p2 25 w_p2 32
u_p3 45 v_p3 76 w_p3 49
... .... ....
u_p599 56 v_p599 78 w_599 98
Now I want the output list as follows
(1 2 7)
(21 11 45)
(32 25 32)
(45 76 49)
.....
These are the 3 series I created from a dataframe
r1=turb_1.iloc[qw1] #List1
r2=turb_1.iloc[qw2] #List2
r3=turb_1.iloc[qw3] #List3
Pic of the seriesFor the output I think formatted string python method will be useful but I m quite not sure how to proceed.
turb_3= ["({A} {B} {C})".format(A=i,B=j,C=k) for i in r1 for j in r2 for k in r3]
Any kind of help will be useful.
Use pandas.DataFrame.itertuples with str.format:
# Sample data
print(df)
col1 col2 col3
0 1 2 7
1 21 11 45
2 32 25 32
3 45 76 49
fmt = "({} {} {})"
[fmt.format(*tup) for tup in df[["col1", "col2", "col3"]].itertuples(False, None)]
Output:
['(1 2 7)', '(21 11 45)', '(32 25 32)', '(45 76 49)']
Related
I have a pandas dataframe as below:
import pandas as pd
import numpy as np
df = pd.DataFrame({'ORDER':["A", "A", "B", "B"], 'var1':[2, 3, 1, 5],'a1_bal':[1,2,3,4], 'a1c_bal':[10,22,36,41], 'b1_bal':[1,2,33,4], 'b1c_bal':[11,22,3,4], 'm1_bal':[15,2,35,4]})
df
ORDER var1 a1_bal a1c_bal b1_bal b1c_bal m1_bal
0 A 2 1 10 1 11 15
1 A 3 2 22 2 22 2
2 B 1 3 36 33 3 35
3 B 5 4 41 4 4 4
I want to create new columns as below:
a1_final_bal = sum(a1_bal, a1c_bal)
b1_final_bal = sum(b1_bal, b1c_bal)
m1_final_bal = m1_bal (since we only have m1_bal field not m1c_bal, so it will renain as it is)
I don't want to hardcode this step because there might be more such columns as "c_bal", "m2_bal", "m2c_bal" etc..
My final data should look something like below
ORDER var1 a1_bal a1c_bal b1_bal b1c_bal m1_bal a1_final_bal b1_final_bal m1_final_bal
0 A 2 1 10 1 11 15 11 12 15
1 A 3 2 22 2 22 2 24 24 2
2 B 1 3 36 33 3 35 38 36 35
3 B 5 4 41 4 4 4 45 8 4
You could try something like this. I am not sure if its exactly what you are looking for, but I think it should work.
dfforgroup = df.set_index(['ORDER','var1']) #Creates MultiIndex
dfforgroup.columns = dfforgroup.columns.str[:2] #Takes first two letters of remaining columns
df2 = dfforgroup.groupby(dfforgroup.columns,axis=1).sum().reset_index().drop(columns =
['ORDER','var1']).add_suffix('_final_bal') #groups columns by their first two letters and sums the columns up
df = pd.concat([df,df2],axis=1) #concatenates new columns to original df
I have a data frame with numbers in multiple columns listed by date, what I'm trying to do is find out the most frequently occurring numbers across the whole data set, also grouped by date.
import pandas as pd
import glob
def lotnorm(pdobject) :
# clean up special characters in the column names and make the date column the index as a date type.
pdobject["Date"] = pd.to_datetime(pdobject["Date"])
pdobject = pdobject.set_index('Date')
for column in pdobject:
if '#' in column:
pdobject = pdobject.rename(columns={column:column.replace('#','')})
return pdobject
def lotimport() :
lotret = {}
# list files in data directory with csv filename
for lotpath in [f for f in glob.glob("data/*.csv")]:
lotname = lotpath.split('\\')[1].split('.')[0]
lotret[lotname] = lotnorm(pd.read_csv(lotpath))
return lotret
print(lotimport()['ozlotto'])
------------- Output ---------------------
1 2 3 4 5 6 7 8 9
Date
2020-07-07 4 5 7 9 12 13 32 19 35
2020-06-30 1 17 26 28 38 39 44 14 41
2020-06-23 1 3 9 13 17 20 41 28 45
2020-06-16 1 2 13 21 22 27 38 24 33
2020-06-09 8 11 26 27 31 38 39 3 36
... .. .. .. .. .. .. .. .. ..
2005-11-15 7 10 13 17 30 32 41 20 14
2005-11-08 12 18 22 28 33 43 45 23 13
2005-11-01 1 3 11 17 24 34 43 39 4
2005-10-25 7 16 23 29 36 39 42 19 43
2005-10-18 5 9 12 30 33 39 45 7 19
The output I am aiming for is
Number frequency
45 201
32 195
24 187
14 160
48 154
--------------- Updated with append experiment -----------
I tried using append to create a single series from the dataframe, which worked for individual lines of code but got a really odd result when I ran it inside a for loop.
temp = lotimport()['ozlotto']['1']
print(temp)
temp = temp.append(lotimport()['ozlotto']['2'], ignore_index=True, verify_integrity=True)
print(temp)
temp = temp.append(lotimport()['ozlotto']['3'], ignore_index=True, verify_integrity=True)
print(temp)
lotcomb = pd.DataFrame()
for i in (lotimport()['ozlotto'].columns.tolist()):
print(f"{i} - {type(i)}")
lotcomb = lotcomb.append(lotimport()['ozlotto'][i], ignore_index=True, verify_integrity=True)
print(lotcomb)
This solution might be the one you are looking for.
freqvalues = np.unique(df.to_numpy(), return_counts=True)
df2 = pd.DataFrame(index=freqvalues[0], data=freqvalues[1], columns=["Frequency"])
df2.index.name = "Numbers"
df2
Output:
Frequency
Numbers
1 6
2 5
3 5
5 8
6 4
7 7
8 2
9 7
10 3
11 4
12 2
13 8
14 1
15 4
16 4
17 6
18 4
19 5
20 9
21 3
22 4
23 2
24 4
25 5
26 4
27 6
28 1
29 6
30 3
31 3
... ...
70 6
71 6
72 5
73 5
74 2
75 8
76 5
77 3
78 3
79 2
80 3
81 4
82 6
83 9
84 5
85 4
86 1
87 3
88 4
89 3
90 4
91 4
92 3
93 5
94 1
95 4
96 6
97 6
98 1
99 6
97 rows × 1 columns
df.max(axis=0)
for columns
df.max(axis=1)
for index
Ok so the final answer I came up with was a mix of a few things including some of the great input from people in this thread. Essentially I do the following:
Pull in the CSV file and clean up the dates and the column names, then convert it to a pandas dataframe.
Then create a new pandas series and append each column to it ignoring dates to prevent conflicts.
Once I have the series, I use Vioxini's suggestion to use numpy to get counts of unique values and then turn the values into the index, after that sort the column by count in descending order and return the top 10 values.
Below is the resulting code, I hope it helps someone else.
import pandas as pd
import glob
import numpy as np
def lotnorm(pdobject) :
# clean up special characters in the column names and make the date column the index as a date type.
pdobject["Date"] = pd.to_datetime(pdobject["Date"])
pdobject = pdobject.set_index('Date')
for column in pdobject:
if '#' in column:
pdobject = pdobject.rename(columns={column:column.replace('#','')})
return pdobject
def lotimport() :
lotret = {}
# list files in data directory with csv filename
for lotpath in [f for f in glob.glob("data/*.csv")]:
lotname = lotpath.split('\\')[1].split('.')[0]
lotret[lotname] = lotnorm(pd.read_csv(lotpath))
return lotret
lotcomb = pd.Series([],dtype=object)
for i in (lotimport()['ozlotto'].columns.tolist()):
lotcomb = lotcomb.append(lotimport()['ozlotto'][i], ignore_index=True, verify_integrity=True)
freqvalues = np.unique(lotcomb.to_numpy(), return_counts=True)
lotop = pd.DataFrame(index=freqvalues[0], data=freqvalues[1], columns=["Frequency"])
lotop.index.name = "Numbers"
lotop.sort_values(by=['Frequency'],ascending=False).head(10)
As homework for IT lessons I need to write a script which will check for the highest power of 4 which is in modified input number, but I can use only 8MB of RAM. I used for this logarithmic function, so my code looks like this:
from math import log, floor
n = int(input())
numbers = []
for i in range (0, n):
numbers.append(floor(int(input()) / 10))
for i in numbers:
print(4 ** floor(log(i, 4)))
But I checked this script on my PC and it uses more than 8MB!
Partition of a set of 74690 objects. Total size = 8423721 bytes.
Index Count % Size % Cumulative % Kind (class / dict of class)
0 23305 31 2100404 25 2100404 25 str
1 19322 26 1450248 17 3550652 42 tuple
2 5017 7 724648 9 4275300 51 types.CodeType
3 9953 13 716915 9 4992215 59 bytes
4 742 1 632536 8 5624751 67 type
5 4618 6 628048 7 6252799 74 function
6 742 1 405720 5 6658519 79 dict of type
7 187 0 323112 4 6981631 83 dict of module
8 612 1 278720 3 7260351 86 dict (no owner)
9 63 0 107296 1 7367647 87 set
<197 more rows. Type e.g. '_.more' to view.>
On my phone, however, this script uses only 2.5MB:
Partition of a set of 35586 objects. Total size = 2435735 bytes.
Index Count % Size % Cumulative % Kind (class / dict of class)
0 9831 28 649462 27 649462 27 str
1 9014 25 365572 15 1015034 42 tuple
2 4669 13 261232 11 1276266 52 bytes
3 2357 7 198684 8 1474950 61 types.CodeType
4 436 1 166276 7 1641226 67 type
5 2156 6 155232 6 1796458 74 function
6 436 1 130836 5 1927294 79 dict of type
7 93 0 87384 4 2014678 83 dict of module
8 237 1 62280 3 2076958 85 dict (no owner) 9 1091 3 48004 2 2124962 87 types.WrapperDescriptorType
<115 more rows. Type e.g. '_.more' to view.>
I tried changing list to tuple, but it didn't make any difference.
Is there any possibility to decrease/limit RAM usage?
My current dataframe data is as follows:
df=pd.DataFrame([[1.4,3.5,4.6],[2.8,5.4,6.4],[7.8,6.5,5.8]],columns=['t','i','m'])
t i m
0 14 35 46
1 28 54 64
2 28 34 64
3 78 65 58
My goal is to apply a vectorized operations on a df with a conditions as follows (pseudo code):
New column of answer starts with value of 1.
For row in df.itertuples():
if (m > i) & (answer in row-1 is an odd number):
answer in row = answer in row-1 + m
elif (m > i):
answer in row = answer in row-1 - m
else:
answer in row = answer in row-1
The desired output is as follows:
t i m answer
0 14 35 46 1
1 28 54 59 60
2 78 12 58 2
3 78 91 48 2
Any elegant solution would be appreciated.
Based on a thorough and accurate response to this question, I am now faced with a new issue based on slightly different data.
Given this data frame:
df = pd.DataFrame({
('A', 'a'): [23,3,54,7,32,76],
('B', 'b'): [23,'n/a',54,7,32,76],
('possible','possible'):[100,100,100,100,100,100]
})
df
A B possible
a b possible
0 23 23 100
1 3 n/a 100
2 54 54 100
3 7 n/a 100
4 32 32 100
5 76 76 100
I'd like to subtract 4 from 'possible', per row, for any instance (column) where the value is 'n/a' for that row (and then change all 'n/a' values to 0).
A B possible
a b possible
0 23 23 100
1 3 n/a 96
2 54 54 100
3 7 n/a 96
4 32 32 100
5 76 76 100
Some conditions:
It may occur that a column is all floats (though they appear to be integers upon inspection). This was not factored into the original question.
It may also occur that a row contains two instances (columns) of 'n/a' values. This was addressed by the previous solution.
Here is the previous solution:
idx = pd.IndexSlice
df.loc[:, idx['possible', 'possible']] -= (df.loc[:, idx[('A','B'),:]] == 'n/a').sum(axis=1) * 4
df.replace({'n/a':0}, inplace=True)
It works, except for where a column (A or B) contains all floats (seemingly integers). When that's the case, this error occurs:
TypeError: Could not compare ['n/a'] with block values
I think you can add casting to string by astype to condition:
idx = pd.IndexSlice
df.loc[:, idx['possible', 'possible']] -=
(df.loc[:, idx[('A','B'),:]].astype(str) == 'n/a').sum(axis=1) * 4
df.replace({'n/a':0}, inplace=True)
print df
A B possible
a b possible
0 23 23 100
1 3 0 96
2 54 54 100
3 7 0 96
4 32 32 100
5 76 76 100