Im trying to concatenate 4 different datasets onto pandas python. I can concatenated them but it results in several of the same column names. How do I only produce only one column of the same name, then multiples?
concatenated_dataframes = pd.concat(
[
dice.reset_index(drop=True),
json.reset_index(drop=True),
flexjobs.reset_index(drop=True),
indeed.reset_index(drop=True),
simply.reset_index(drop=True),
],
axis=1,
ignore_index=True,
)
concatenated_dataframes_columns = [
list(dice.columns),
list(json.columns),
list(flexjobs.columns),
list (indeed.columns),
list(simply.columns)
]
flatten = lambda nested_lists: [item for sublist in nested_lists for item in sublist]
concatenated_dataframes.columns = flatten(concatenated_dataframes_columns)
df= concatenated_dataframes
This results in
UNNAMED: 0 TITLE COMPANY DESCRIPTION LOCATION TITLE JOBLOCATION POSTEDDATE DETAILSPAGEURL COMPANYPAGEURL COMPANYLOGOURL SALARY CLIENTBRANDID COMPANYNAME EMPLOYMENTTYPE SUMMARY SCORE EASYAPPLY EMPLOYERTYPE WORKFROMHOMEAVAILABILITY ISREMOTE UNNAMED: 0 TITLE SALARY JOBTYPE LOCATION DESCRIPTION UNNAMED: 0 TITLE SALARY JOBTYPE DESCRIPTION LOCATION UNNAMED: 0 COMPANY DESCRIPTION LOCATION SALARY TITLE
Again, how do i combined all the 'titles' in one column, all the 'location' in one column, and so on? Instead of have multiple of them.
I think we can get away with making a blank dataframe that just has the columns we will want at the end and then concat() everything onto it.
import numpy as np
import pandas as pd
all_columns = list(dice.columns) + list(json.columns) + list(flexjobs.columns) + list(indeed.columns) + list(simply.columns)
all_unique_columns = np.unique(np.array(all_columns)) # this will, as the name suggests, give an end list of just the unique columns. You could run print(all_unique_columns) to make sure it has what you want
df = pd.DataFrame(columns=all_unique_columns)
df = pd.concat([dice, json, flexjobs, indeed, simply],axis=0)
It's a little tricky not having reproducible examples of the dataframes that you have. I tested this on a small mock-up example I put together, but let me know if it works for your more complex example.
Related
I've joined or concatenated two series into a dataframe. However one of the issues I'm not facing is that I have no column headings on the actual data that would help me do a sort
hist_a = pd.crosstab(category_a, category, normalize=True)
hist_b = pd.crosstab(category_b, category, normalize=True)
counts_a = pd.Series(np.diag(hist_a), index=[hist_a.index])
counts_b = pd.Series(np.diag(hist_b), index=[hist_b.index])
df_plots = pd.concat([counts_a, counts_b], axis=1).fillna(0)
The data looks like the following:
0 1
category
0017817703277 0.000516 5.384341e-04
0017817703284 0.000516 5.384341e-04
0017817731348 0.000216 2.856169e-04
0017817731355 0.000216 2.856169e-04
and I'd like to do a sort, but there are no proper column headings
df_plots = df_plots.sort_values(by=['0?'])
But the dataframe seems to be in two parts. How could I better structure the dataframe to have 'proper' columns such as '0' or 'plot a' rather than being indexable by an integer, which seems to be hard to work with.
category plot a plot b
0017817703277 0.000516 5.384341e-04
0017817703284 0.000516 5.384341e-04
0017817731348 0.000216 2.856169e-04
0017817731355 0.000216 2.856169e-04
Just rename the columns of the dataframe, for example:
df = pd.DataFrame({0:[1,23]})
df = df.rename(columns={0:'new name'})
If you have a lot of columns you rename all of them at once like:
df = pd.DataFrame({0:[1,23]})
rename_dict = {key: f'Col {key}' for key in df.keys() }
df = df.rename(columns=rename_dict)
You can also define the series with the name, so you avoid changing the name afterwards:
counts_a = pd.Series(np.diag(hist_a), index=[hist_a.index], name = 'counts_a')
counts_b = pd.Series(np.diag(hist_b), index=[hist_b.index], name = 'counts_b')
I am trying to match names from two dataframes (in the name columns) using fuzzywuzzy with process. The result should be df1(dfdum) with the best matching name from df2(dfpep) and the similarity score. This is going very well with the code below but besides the matching name and score I want to append more columns from df2 to df1 in the result. The dates of birth and countries of residence from df2 belonging to the matching name should also be added to df1. I cannot simply merge on names because there are duplicates.
Can anyone help me to amend the code so that i can add the extra info from the matching names from df2? I thus want to add two extra columns to df1 with relating information from the matching name from df2.
pep_name = []
sim_name = []
for i in dfdum.NAME:
ratio = process.extract(i, dfpep.NAME, limit=1,scorer=fuzz.token_set_ratio)
pep_name.append(ratio[0][0])
sim_name.append(ratio[0][1])
dfdum['pep_name'] = pd.Series(pep_name)
dfdum['sim_name'] = pd.Series(sim_name)
You could find the index of the best match in dfpep.NAME, and use that to retrieve the corresponding values of the other two columns.
This code (with some mock data) should give you the desired result; it assumes that dfpep.NAME has only unique values, though.
Please note that I'm far from a pandas expert so this solution is by no means the fastest or most elegant, but it should do the job :)
Also, I feel like there should be a way to do this without the for loop; maybe someone here has an idea for that.
import pandas as pd
from fuzzywuzzy import process, fuzz
dfdum = pd.DataFrame(["Johnny", "Peter", "Ben"])
dfdum.columns = ["NAME"]
dfpep = pd.DataFrame(["Pete", "John", "Bennie"])
dfpep.columns = ["NAME"]
dfpep["dob"] = pd.Series(["1990", "1991", "1992"])
dfpep["cor"] = pd.Series(["USA", "UK", "Germany"])
pep_name = []
sim_name = []
dob = []
cor = []
for i in dfdum.NAME:
ratio = process.extract(i, dfpep.NAME, limit=1, scorer=fuzz.token_set_ratio)
pep_name.append(ratio[0][0])
sim_name.append(ratio[0][1])
j = dfpep.index[dfpep.NAME == ratio[0][0]].tolist()[0]
dob.append(dfpep['dob'][j])
cor.append(dfpep['cor'][j])
dfdum['pep_name'] = pd.Series(pep_name)
dfdum['sim_name'] = pd.Series(sim_name)
dfdum['dob'] = pd.Series(dob)
dfdum['cor'] = pd.Series(cor)
print(dfdum)
I have a data frame like as shown below
df = pd.DataFrame({'subject_id':[11,11,11,12,12,12],
'test_date':['02/03/2012 10:24:21','05/01/2019 10:41:21','12/13/2011 11:14:21','10/11/1992 11:14:21','02/23/2002 10:24:21','07/19/2005 10:24:21'],
'original_enc':['A742','B963','C354','D563','J323','G578']})
hash_file = pd.DataFrame({'source_enc':['A742','B963','C354','D563','J323','G578'],
'hash_id':[1,2,3,4,5,6]})
cols = ["subject_id","test_date","enc_id","previous_enc_id"]
test_df = pd.DataFrame(columns=cols)
test_df.head()
I would like to do two things here
Map original_enc to their corresponding hash_id and store it in enc_id
Find the previous hash_id for each subject based on their current hash_id and store it in previous_enc_id
I tried the below
test_df['subject_id'] = df['subject_id']
test_df['test_date'] = df['test_date']
test_df['enc_id'] = df['original_enc'].map(hash_file)
test_df = test_df.sort_values(['subject_id','test_date'],ascending=True)
test_df['previous_enc_id'] = test_df.groupby(['subject_id','test_date'])['enc_id'].shift(1)
However, I don't get the expected output for the previous_enc_id column as it is all NA.
I expect my output to be like as shown below. You see NA in the expected row for the 1st row of every subject because that's their 1st encounter. There is no info to look back to.
Use only one column for groupby:
test_df['previous_enc_id'] = test_df.groupby('subject_id')['enc_id'].shift()
I have 2 dataframes :
df1 is a list of mailboxes and email ids
df2 shows a list of approved domains
I read both the dataframes from an excel sheet
xls = pd.ExcelFile(input_file_shared_mailbox)
df = pd.read_excel(xls, sheet_name = sheet_name_shared_mailbox)
i want to only keep records in df1 where df1[Email_Id] contains df2[approved_domain]
print(df1)
Mailbox Email_Id
0 mailbox1 abc#gmail.com
1 mailbox2 def#yahoo.com
2 mailbox3 ghi#msn.com
print(df2)
approved_domain
0 msn.com
1 gmail.com
and i want df3 which basically shows
print (df3)
Mailbox Email_Id
0 mailbox1 abc#gmail.com
1 mailbox3 ghi#msn.com
this is the code i have right now which i think is close but i can't figure out the exact problem in the syntax
df3 = df1[df1['Email_Id'].apply(lambda x: [item for item in x if item in df2['Approved_Domains'].tolist()])]
But get this error
TypeError: unhashable type: 'list'
i spent a lot of time researching the forum for a solution but could not find what i was looking for. appreciate all the help.
So these are the steps you will need to follow to do what you want done for your two data frames
1.Split your email_address column into two separate columns
df1['add'], df1['domain'] = df1['email_address'].str.split('#', 1).str
2.Then drop your add column to keep your data frame clean
df1 = df1.drop('add',axis =1)
3.Get a new Data Frame with only values you want by not selecting any value in the 'domain' column that doesn't match 'approved_doman' column
df_new = df1[~df1['domain'].isin(df2['approved_domain'])]
4. Drop the 'domain' column in df_new
df_new = df_new.drop('domain',axis = 1)
This is what the result will be
mailbox email_address
1 mailbox2 def#yahoo.com
2 mailbox3 ghi#msn.com
You can use dynamically created regular expression to search for the valid domain in the list and eventually filtering them out.
Here is the code for our reference.
# -*- coding: utf-8 -*-
import pandas as pd
import re
mailbox_list = [
['mailbox1', 'abc#gmail.com'],
['mailbox2', 'def#yahoo.com'],
['mailbox3', 'ghi#msn.com']]
valid_domains = ['msn.com', 'gmail.com']
df1 = pd.DataFrame(mailbox_list, columns=['Mailbox', 'EmailID'])
df2 = pd.DataFrame(valid_domains)
valid_list = []
for index, row in df1.iterrows():
for idx, record in df2.iterrows():
if re.search(rf"#{record[0]}", row[1], re.IGNORECASE):
valid_list.append([row[0], row[1]])
df3 = pd.DataFrame(valid_list, columns=['Mailbox', 'EmailID'])
print(df3)
The output of this is:
Mailbox EmailID
0 mailbox1 abc#gmail.com
1 mailbox3 ghi#msn.com
Solution
df1 = {'MailBox': ['mailbox1', 'mailbox2', 'mailbox3'], 'Email_Id': ['abc#gmail.com', 'def#yahoo.com', 'ghi#msn.com']}
df2 = {'approved_domain':['msn.com', 'gmail.com']}
mailboxes, emails = zip( # unzip the columns
*filter( # filter
lambda i: any([ # i = ('mailbox1', 'abc#gmail.com')
approved_domain in i[1] for approved_domain in df2['approved_domain']
]),
zip(df1['MailBox'], df1['Email_Id']) # zip the columns
)
)
df3 = {
'MailBox': mailboxes,
'Email_I': emails
}
print(df3)
Output:
> {'Email_ID': ('abc#gmail.com', 'ghi#msn.com'), 'MailBox': ('mailbox1', 'mailbox3')}
Some notes:
Big chunk of this code is basically just for parsing the data structure. The zipping and unzipping is only there to convert the list of columns to a list of rows and back. If you have a a list of rows already, you just have to do the filtering part
I have a lisit of DataFrames that come from the census api, i had stored each year pull into a list.
So at the end of my for loop i have a list with dataframes per year and a list of years to go along side the for loop.
The problem i am having is merging all the DataFrames in the list while also taging them with a list of years.
So i have tried using the reduce function, but it looks like it only taking 2 of the 6 Dataframes i have.
concat just adds them to the dataframe with out tagging or changing anything
# Dependencies
import pandas as pd
import requests
import json
import pprint
import requests
from census import Census
from us import states
# Census
from config import (api_key, gkey)
year = 2012
c = Census(api_key, year)
for length in range(6):
c = Census(api_key, year)
data = c.acs5.get(('NAME', "B25077_001E","B25064_001E",
"B15003_022E","B19013_001E"),
{'for': 'zip code tabulation area:*'})
data_df = pd.DataFrame(data)
data_df = data_df.rename(columns={"NAME": "Name",
"zip code tabulation area": "Zipcode",
"B25077_001E":"Median Home Value",
"B25064_001E":"Median Rent",
"B15003_022E":"Bachelor Degrees",
"B19013_001E":"Median Income"})
data_df = data_df.astype({'Zipcode':'int64'})
filtervalue = data_df['Median Home Value']>0
filtervalue2 = data_df['Median Rent']>0
filtervalue3 = data_df['Median Income']>0
cleandata = data_df[filtervalue][filtervalue2][filtervalue3]
cleandata = cleandata.dropna()
yearlst.append(year)
datalst.append(cleandata)
year += 1
so this generates the two seperate list one with the year and other with dataframe.
So my output came out to either one Dataframe with missing Dataframe entries or it just concatinated all without changing columns.
what im looking for is how to merge all within a list, but datalst[0] to be tagged with yearlst[0] when merging if at all possible
No need for year list, simply assign year column to data frame. Plus avoid incrementing year and have it the iterator column. In fact, consider chaining your process:
for year in range(2012, 2019):
c = Census(api_key, year)
data = c.acs5.get(('NAME', "B25077_001E","B25064_001E", "B15003_022E","B19013_001E"),
{'for': 'zip code tabulation area:*'})
cleandata = (pd.DataFrame(data)
.rename(columns={"NAME": "Name",
"zip code tabulation area": "Zipcode",
"B25077_001E": "Median_Home_Value",
"B25064_001E": "Median_Rent",
"B15003_022E": "Bachelor_Degrees",
"B19013_001E": "Median_Income"})
.astype({'Zipcode':'int64'})
.query('(Median_Home_Value > 0) & (Median_Rent > 0) & (Median_Income > 0)')
.dropna()
.assign(year_column = year)
)
datalst.append(cleandata)
final_data = pd.concat(datalst, ignore_index = True)