I have a pandas data frame with several thousand observations and I would like to create "leakage-free" variables in Python. So I am looking for a way to calculate e.g. a group-specific mean of a variable without the single observation in row i.
For example:
| Group | Price | leakage-free Group Mean |
-------------------------------------------
| 1 | 20 | 25 |
| 1 | 40 | 15 |
| 1 | 10 | 30 |
| 2 | ... | ... |
I would like to do that with several variables and I would like to create mean, median and variance in such a way, so a computationally fast method might be good. If a group has only one row I would like to enter 0s in the leakage-free Variable.
As I am rather a beginner in Python, some piece of code might be very helpful. Thank You!!
With one-liner:
df = pd.DataFrame({'Group': [1,1,1,2], 'Price':[20,40,10,30]})
df['lfgm'] = df.groupby('Group').transform(lambda x: (x.sum()-x)/(len(x)-1)).fillna(0)
print(df)
Output:
Group Price lfgm
0 1 20 25.0
1 1 40 15.0
2 1 10 30.0
3 2 30 0.0
Update:
For median and variance (not one-liners unfortunately):
df = pd.DataFrame({'Group': [1,1,1,1,2], 'Price':[20,100,10,70,30]})
def f(x):
for i in x.index:
z = x.loc[x.index!=i, 'Price']
x.at[i, 'mean'] = z.mean()
x.at[i, 'median'] = z.median()
x.at[i, 'var'] = z.var()
return x[['mean', 'median', 'var']]
df = df.join(df.groupby('Group').apply(f))
print(df)
Output:
Group Price mean median var
0 1 20 60.000000 70.0 2100.000000
1 1 100 33.333333 20.0 1033.333333
2 1 10 63.333333 70.0 1633.333333
3 1 70 43.333333 20.0 2433.333333
4 2 30 NaN NaN NaN
Use:
grp = df.groupby('Group')
n = grp['Price'].transform('count')
mean = grp['Price'].transform('mean')
df['new_col'] = (mean*n - df['Price'])/(n-1)
print(df)
Group Price new_col
0 1 20 25.0
1 1 40 15.0
2 1 10 30.0
Note: This solution will be faster than using apply, you can test using %%timeit followed by the codes.
Related
I need to create a new column as Billing and Non-Billing based on the Billable column. If the Billable is 'Yes' then i should create a new column as Billing and if its 'No' then need to create a new column as 'Non-Billable' and need to calculate it. Calculation should be in row axis.
Calculation for Billing in row:
Billing = df[Billing] * sum/168 * 100
Calculation for Non-Billing in row:
Non-Billing = df[Non-Billing] * sum/ 168 * 100
Data
Employee Name | Java | Python| .Net | React | Billable|
----------------------------------------------------------------
|Priya | 10 | | 5 | | Yes |
|Krithi | | 10 | 20 | | No |
|Surthi | | 5 | | | yes |
|Meena | | 20 | | 10 | No |
|Manju | 20 | 10 | 10 | | Yes |
Output
I have tried using insert statement but i cannot keep on inserting it. I tried append also but its not working.
Bill_amt = []
Non_Bill_amt = []
for i in df['Billable']:
if i == "Yes" or i == None:
Bill_amt = (df[Bill_amt].sum(axis=1)/168 * 100).round(2)
df.insert (len( df.columns ), column='Billable Amount', value=Bill_amt )#inserting the column and it name
#CANNOT INSERT ROW AFTER IT AND CANNOT APPEND IT TOO
else:
Non_Bill_amt = (DF[Non_Bill_amt].sum ( axis=1 ) / 168 * 100).round ( 2 )
df.insert ( len ( df.columns ), column='Non Billable Amount', value=Non_Bill_amt ) #inserting the column and its name
#CANNOT INSERT ROW AFTER IT.
Use .sum(axis=1) and then np.where() to put the values in respective columns. For example:
x = df.loc[:, "Java":"React"].sum(axis=1) / 168 * 100
df["Bill"] = np.where(df["Billable"].str.lower() == "yes", x, "")
df["Non_Bill"] = np.where(df["Billable"].str.lower() == "no", x, "")
print(df)
Prints:
Employee_Name Java Python .Net React Billable Bill Non_Bill
0 Priya 10.0 NaN 5.0 NaN Yes 8.928571428571429
1 Krithi NaN 10.0 20.0 NaN No 17.857142857142858
2 Surthi NaN 5.0 NaN NaN yes 2.976190476190476
3 Meena NaN 20.0 NaN 10.0 No 17.857142857142858
4 Manju 20.0 10.0 10.0 NaN Yes 23.809523809523807
I have a dataset given below:
weekid type amount
1 A 10
1 B 20
1 C 30
1 D 40
1 F 50
2 A 70
2 E 80
2 B 100
I am trying to convert it to another panda frame based on total number of type values defined with:
import pandas as pd
import numpy as np
df=pd.read_csv(INPUT_FILE)
for type in df["type"].unique():
//todo
My aim is to get a data given below:
weekid type_A type_B type_C type_D type_E type_F
1 10 20 30 40 0 50
2 70 100 0 0 80 0
Is there any specific function that convert unique values as a column and fills the missing values as 0 for each weekId groups? I am wondering that how this conversion can be done efficiently?
You can use the following:
df = df.pivot(columns=['type'], values=['amount'])
df.fillna(0)
dfp.columns = dfp.columns.droplevel(0)
Given your input this yields:
type A B C D F
weekid
1 10.0 20.0 30.0 40.0 50.0
2 70.0 80.0 100.0 0.0 0.0
I have a dataframe that looks like this:
ID EVENT DATE
1 1 142
1 5 167
1 3 245
2 1 54
2 5 87
3 3 165
3 2 178
And I would like to generate something like this:
EVENT_1 EVENT_2 COUNT
1 5 2
5 3 1
3 2 1
The idea is how many items (ID) go from one event to the next one. Don't care about previous states, I just want to consider the next state from the current state (e.g.: for ID 1, I don't want to count a transition from 1 to 3 because first, it goes to event 5 and then to 3).
The date format is the number of days from a specific date (sort of like SAS format).
Is there a clean way to achieve this?
Let's try this:
(df.groupby([df['EVENT'].rename('EVENT_1'),
df.groupby('ID')['EVENT'].shift(-1).rename('EVENT_2')])['ID']
.count()).rename('COUNT').reset_index().astype(int)
Output:
| | EVENT_1 | EVENT_2 | COUNT |
|---:|----------:|----------:|--------:|
| 0 | 1 | 5 | 2 |
| 1 | 3 | 2 | 1 |
| 2 | 5 | 3 | 1 |
Details: Groupby on 'EVENT' and shifted 'EVENT' within each ID, then count.
You could use groupby and shift. We'll also use rename_axis and reset_index to tidy up the final output:
(pd.concat([f.groupby([f['EVENT'], f['EVENT'].shift(-1).astype('Int64')]).size()
for _, f in df.groupby('ID')])
.groupby(level=[0, 1]).sum()
.rename_axis(['EVENT_1', 'EVENT_2']).reset_index(name='COUNT'))
[out]
EVENT_1 EVENT_2 COUNT
0 1 5 2
1 3 2 1
2 5 3 1
This question already has answers here:
How do I create a new column from the output of pandas groupby().sum()?
(4 answers)
Closed 3 years ago.
I want to create a new column in python dataframe based on other column values in multiple rows.
For example, my python dataframe df:
A | B
------------
10 | 1
20 | 1
30 | 1
10 | 1
10 | 2
15 | 3
10 | 3
I want to create variable C that is based on the value of variable A with condition from variable B in multiple rows. When the value of variable B in row i,i+1,..., the the value of C is the sum of variable A in those rows. In this case, my output data frame will be:
A | B | C
--------------------
10 | 1 | 70
20 | 1 | 70
30 | 1 | 70
10 | 1 | 70
10 | 2 | 10
15 | 3 | 25
10 | 3 | 25
I haven't got any idea the best way to achieve this. Can anyone help?
Thanks in advance
recreate the data:
import pandas as pd
A = [10,20,30,10,10,15,10]
B = [1,1,1,1,2,3,3]
df = pd.DataFrame({'A':A, 'B':B})
df
A B
0 10 1
1 20 1
2 30 1
3 10 1
4 10 2
5 15 3
6 10 3
and then i'll create a lookup Series from the df:
lookup = df.groupby('B')['A'].sum()
lookup
A
B
1 70
2 10
3 25
and then i'll use that lookup on the df using apply
df.loc[:,'C'] = df.apply(lambda row: lookup[lookup.index == row['B']].values[0], axis=1)
df
A B C
0 10 1 70
1 20 1 70
2 30 1 70
3 10 1 70
4 10 2 10
5 15 3 25
6 10 3 25
You have to use groupby() method, to group the rows on B and sum() on A.
df['C'] = df.groupby('B')['A'].transform(sum)
I have data in the following format:
| | Measurement 1 | | Measurement 2 | |
|------|---------------|------|---------------|------|
| | Mean | Std | Mean | Std |
| Time | | | | |
| 0 | 17 | 1.10 | 21 | 1.33 |
| 1 | 16 | 1.08 | 21 | 1.34 |
| 2 | 14 | 0.87 | 21 | 1.35 |
| 3 | 11 | 0.86 | 21 | 1.33 |
I am using the following code to generate a matplotlib line graph from this data, which shows the standard deviation as a filled in area, see below:
def seconds_to_minutes(x, pos):
minutes = f'{round(x/60, 0)}'
return minutes
fig, ax = plt.subplots()
mean_temperature_over_time['Measurement 1']['mean'].plot(kind='line', yerr=mean_temperature_over_time['Measurement 1']['std'], alpha=0.15, ax=ax)
mean_temperature_over_time['Measurement 2']['mean'].plot(kind='line', yerr=mean_temperature_over_time['Measurement 2']['std'], alpha=0.15, ax=ax)
ax.set(title="A Line Graph with Shaded Error Regions", xlabel="x", ylabel="y")
formatter = FuncFormatter(seconds_to_minutes)
ax.xaxis.set_major_formatter(formatter)
ax.grid()
ax.legend(['Mean 1', 'Mean 2'])
Output:
This seems like a very messy solution, and only actually produces shaded output because I have so much data. What is the correct way to produce a line graph from the dataframe I have with shaded error regions? I've looked at Plot yerr/xerr as shaded region rather than error bars, but am unable to adapt it for my case.
What's wrong with the linked solution? It seems pretty straightforward.
Allow me to rearrange your dataset so it's easier to load in a Pandas DataFrame
Time Measurement Mean Std
0 0 1 17 1.10
1 1 1 16 1.08
2 2 1 14 0.87
3 3 1 11 0.86
4 0 2 21 1.33
5 1 2 21 1.34
6 2 2 21 1.35
7 3 2 21 1.33
for i, m in df.groupby("Measurement"):
ax.plot(m.Time, m.Mean)
ax.fill_between(m.Time, m.Mean - m.Std, m.Mean + m.Std, alpha=0.35)
And here's the result with some random generated data:
EDIT
Since the issue is apparently iterating over your particular dataframe format let me show how you could do it (I'm new to pandas so there may be better ways). If I understood correctly your screenshot you should have something like:
Measurement 1 2
Mean Std Mean Std
Time
0 17 1.10 21 1.33
1 16 1.08 21 1.34
2 14 0.87 21 1.35
3 11 0.86 21 1.33
df.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 4 entries, 0 to 3
Data columns (total 4 columns):
(1, Mean) 4 non-null int64
(1, Std) 4 non-null float64
(2, Mean) 4 non-null int64
(2, Std) 4 non-null float64
dtypes: float64(2), int64(2)
memory usage: 160.0 bytes
df.columns
MultiIndex(levels=[[1, 2], [u'Mean', u'Std']],
labels=[[0, 0, 1, 1], [0, 1, 0, 1]],
names=[u'Measurement', None])
And you should be able to iterate over it with and obtain the same plot:
for i, m in df.groupby("Measurement"):
ax.plot(m["Time"], m['Mean'])
ax.fill_between(m["Time"],
m['Mean'] - m['Std'],
m['Mean'] + m['Std'], alpha=0.35)
Or you could restack it to the format above with
(df.stack("Measurement") # stack "Measurement" columns row by row
.reset_index() # make "Time" a normal column, add a new index
.sort_values("Measurement") # group values from the same Measurement
.reset_index(drop=True)) # drop sorted index and make a new one