I am trying to create a function that creates a dataframe based on different lists of words that come up in a certain column of another dataframe.
In my example, I want a dataframe created on the basis of the words "chandos" and "electronics" coming up in the "description" column of the "uncategorised" dataframe.
The point of the function is that I want to be able to run this on different lists of words so I end up with different dataframes containing just the words I want.
words_Telephone = ["tfl", "electronics"]
df_Telephone = pd.DataFrame(columns=['date','description','paid out'])
def categorise(word_list, df_name):
""" takes the denoted terms from the "uncategorised" df and puts it into new df"""
for word in word_list:
df_name = uncategorised[uncategorised['description'].str.contains(word)]
return(df_name)
#apply the function
categorise(words_Telephone, df_Telephone)
I am expecting a dataframe that contains:
d = {'date': {0: '05/04/2017',
1: '06/04/2017',
2: '08/04/2017',
3: '08/04/2017',
4: '08/04/2017',
5: '10/04/2017',
6: '10/04/2017',
7: '10/04/2017'},
'description': {0: 'tfl',
1: 'tfl',
2: 'tfl',
3: 'tfl',
4: 'ac electronics ',
5: 'ac electronics ',},
'index': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'paid out': {0: 3.0,
1: 4.3,
2: 6.1,
3: 1.5,
4: 16.39,
5: 20.4,}}
Reproducible df:
d = {'date': {0: '05/04/2017',
1: '06/04/2017',
2: '06/04/2017',
3: '08/04/2017',
4: '08/04/2017',
5: '08/04/2017',
6: '10/04/2017',
7: '10/04/2017',
8: '10/04/2017'},
'description': {0: 'tfl',
1: 'mu subscription',
2: 'tfl',
3: 'tfl',
4: 'tfl',
5: 'ac electronics ',
6: 'itunes',
7: 'ac electronics ',
8: 'google adwords'},
'index': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'paid out': {0: 3.0,
1: 16.9,
2: 4.3,
3: 6.1,
4: 1.5,
5: 16.39,
6: 12.99,
7: 20.4,
8: 39.68}}
SOLUTION:
def categorise(word_list):
""" takes the denoted terms from the "uncategorised" df and puts it into new df then deletes from the uncategorised df"""
global uncategorised
new_dfs = []
for word in word_list:
new_dfs.append(uncategorised[uncategorised['description'].str.contains(word)])
uncategorised= uncategorised[ ~uncategorised['description'].str.contains(word)]
return (uncategorised)
return (pd.concat(new_dfs).reset_index())
#apply the function
df_Telephone = categorise(words_Telephone)
df_Telephone
words_Telephone = ["tfl", "electronics"]
original_df = pd.DataFrame().from_dict({'date': ['05/04/2017','06/04/2017','06/04/2017','08/04/2017','08/04/2017','08/04/2017','10/04/2017','10/04/2017','10/04/2017'], 'description': ['tfl','mu subscription','tfl','tfl','tfl','ac electronics','itunes','ac electronics','google adwords'], 'paid out' :[ 3.0,16.9, 4.3,6.1,1.5,16.39,12.99,20.4,39.68]})
def categorise(word_list, original_df):
""" takes the denoted terms from the "uncategorised" df and puts it into new df"""
new_dfs = []
for word in word_list:
new_dfs.append(original_df[original_df['description'].str.contains(word)])
return pd.concat(new_dfs).reset_index()
#apply the function
df_Telephone = categorise(words_Telephone, original_df)
print(df_Telephone)
date description paid out
0 05/04/2017 tfl 3.00
1 06/04/2017 tfl 4.30
2 08/04/2017 tfl 6.10
3 08/04/2017 tfl 1.50
4 08/04/2017 ac electronics 16.39
5 10/04/2017 ac electronics 20.40
Related
I have the following dataframes (this is just test data), in real samples, I have index values that are repeated a few times inside dataframe 1 and dataframe 2 - this causes the repeated/duplicate rows inside final dataframe.
DataFrame 1:
pd.DataFrame({'id': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'first_name': {0: 'Jennee',
1: 'Dagny',
2: 'Correy',
3: 'Pall',
4: 'Julie',
5: 'Janene',
6: 'Lemmy',
7: 'Coleman',
8: 'Beck',
9: 'Che'},
'last_name': {0: 'Strelitzki',
1: 'Dunsire',
2: 'Wickrath',
3: 'Jopp',
4: 'Gheeraert',
5: 'Gawith',
6: 'Farrow',
7: 'Legging',
8: 'Beckwith',
9: 'Burgoin'},
'email': {0: 'jstrelitzki0#google.de',
1: 'ddunsire1#geocities.com',
2: 'cwickrath2#github.com',
3: 'pjopp3#infoseek.co.jp',
4: 'jgheeraert4#theatlantic.com',
5: 'jgawith5#sciencedirect.com',
6: 'lfarrow6#wikimedia.org',
7: 'clegging7#businessinsider.com',
8: 'bbeckwith8#zdnet.com',
9: 'cburgoin9#reference.com'},
'gender': {0: 'Male',
1: 'Female',
2: 'Female',
3: 'Female',
4: 'Female',
5: 'Female',
6: 'Male',
7: 'Female',
8: 'Polygender',
9: 'Male'},
'ip_address': {0: '8.99.68.120',
1: '188.238.129.48',
2: '87.159.243.249',
3: '66.37.174.94',
4: '233.77.128.104',
5: '190.202.131.98',
6: '84.175.231.196',
7: '140.178.100.5',
8: '81.211.179.167',
9: '31.219.69.206'},
'Boolean': {0: False,
1: False,
2: True,
3: True,
4: False,
5: True,
6: True,
7: False,
8: False,
9: False}})
DataFrame 2:
pd.DataFrame({'id': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'Model': {0: 2005,
1: 2007,
2: 2011,
3: 2003,
4: 1998,
5: 1992,
6: 1992,
7: 1992,
8: 2008,
9: 1996},
'Make': {0: 'Cadillac',
1: 'Lexus',
2: 'Dodge',
3: 'Dodge',
4: 'Oldsmobile',
5: 'Volkswagen',
6: 'Chevrolet',
7: 'Suzuki',
8: 'Ford',
9: 'Mazda'},
'Colour': {0: 'Red',
1: 'Red',
2: 'Crimson',
3: 'Red',
4: 'Purple',
5: 'Crimson',
6: 'Red',
7: 'Aquamarine',
8: 'Puce',
9: 'Maroon'}})
The two dataframes should be connected based on common Index values found in both dataframes only. Which means, any index values that don't match in those two dataframes; should not appear in the final combined/merged dataframe.
I want to ensure that the final dataframe is unique, and only captures combinations of columns, based on unique Index values.
When I try using the following code, the output is supposed to 'inner join' based on the unique index found in both dataframes.
final = pd.merge(df1, df2, left_index=True, right_index=True)
However, when I try applying the above merge technique on my larger (other) pandas dataframes, there are many rows being repeated/duplicated multiple times. When the merging happpens a few times with more dataframes, the rows gets repeated very frequently, with the same Index value.
I am expecting to see one Index value returned per row (with all the column combinations from each dataframe).
I am not sure why this happens. I can confirm that there is nothing wrong with the datasets.
Is there a better technique of merging those two dataframes, based on only common index values, and at the same time ensure that I don't repeat any rows (with the same index) in my final dataframe ? I often find that this merging often creates a giant final CSV file around 20GB in size too. The source files are only around 15MB into total.
Any help is much appreciated.
My end output should look like this (please copy and use this as Pandas DF):
pd.DataFrame({'id': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'first_name': {0: 'Jennee',
1: 'Dagny',
2: 'Correy',
3: 'Pall',
4: 'Julie',
5: 'Janene',
6: 'Lemmy',
7: 'Coleman',
8: 'Beck',
9: 'Che'},
'last_name': {0: 'Strelitzki',
1: 'Dunsire',
2: 'Wickrath',
3: 'Jopp',
4: 'Gheeraert',
5: 'Gawith',
6: 'Farrow',
7: 'Legging',
8: 'Beckwith',
9: 'Burgoin'},
'email': {0: 'jstrelitzki0#google.de',
1: 'ddunsire1#geocities.com',
2: 'cwickrath2#github.com',
3: 'pjopp3#infoseek.co.jp',
4: 'jgheeraert4#theatlantic.com',
5: 'jgawith5#sciencedirect.com',
6: 'lfarrow6#wikimedia.org',
7: 'clegging7#businessinsider.com',
8: 'bbeckwith8#zdnet.com',
9: 'cburgoin9#reference.com'},
'gender': {0: 'Male',
1: 'Female',
2: 'Female',
3: 'Female',
4: 'Female',
5: 'Female',
6: 'Male',
7: 'Female',
8: 'Polygender',
9: 'Male'},
'ip_address': {0: '8.99.68.120',
1: '188.238.129.48',
2: '87.159.243.249',
3: '66.37.174.94',
4: '233.77.128.104',
5: '190.202.131.98',
6: '84.175.231.196',
7: '140.178.100.5',
8: '81.211.179.167',
9: '31.219.69.206'},
'Boolean': {0: False,
1: False,
2: True,
3: True,
4: False,
5: True,
6: True,
7: False,
8: False,
9: False},
'Model': {0: 2005,
1: 2007,
2: 2011,
3: 2003,
4: 1998,
5: 1992,
6: 1992,
7: 1992,
8: 2008,
9: 1996},
'Make': {0: 'Cadillac',
1: 'Lexus',
2: 'Dodge',
3: 'Dodge',
4: 'Oldsmobile',
5: 'Volkswagen',
6: 'Chevrolet',
7: 'Suzuki',
8: 'Ford',
9: 'Mazda'},
'Colour': {0: 'Red',
1: 'Red',
2: 'Crimson',
3: 'Red',
4: 'Purple',
5: 'Crimson',
6: 'Red',
7: 'Aquamarine',
8: 'Puce',
9: 'Maroon'}})
This is expected behavior with non-unique idx values. Since you have 3 ID1 rows in one df and 2 ID1 in the other, you end up with 6 ID1 rows in your merged df. If you add validate="one_to_one" to pd.merge() you will get this Error. MergeError: Merge keys are not unique in either left or right dataset; not a one-to-one mergeAll other validations fail except for many to many.
If it makes sense for your data, you can use the left_on, and right_on parameters to find unique combinations and give you a one-to-one if that's what you're after.
Edit after your new data:
Now that you have unique ids, this should work for you. Notice it doesn't throw a validation error.
final = pd.merge(df1, df2, left_on=['id'], right_on=['id'], validate='one_to_one')
I am trying to merge 2 dataframes and have a problem in figuring out how, as it is not straigh forward.
One data frame has match results for over 25000 games and looks like this.
The second one has team performance metrics but only for around 1500 games.
As I am not allowed to post pictures yet, here are the column names of interest:
df_match['date', 'home_team_api_id', 'away_team_api_id']
df_team_attributes['date', 'team_api_id']
Both data frames have additional columns with results or performance metrics.
To be able to merge correctly, I need to merge by date and by looking if the 'team_api_id' matches either 'home...' or 'away_team_api_id'
This is what I have tried until now:
df_team_performance = pd.merge(df_team_attributes, df_match,
how = 'left',
left_on = ['date', 'team_api_id', 'team_api_id'],
right_on = ['date', 'home_team_api_id', 'home_team_api_id'])
I have tried also with only 2 columns, but w/o succes.
What I would like to get is a new data frame with only the rows of the df_team_attributes and columns from both data frames.
Thank you in advance!
Added to request by Correlien:
output of print(df_match[['date', 'home_team_api_id', 'away_team_api_id', 'win_home', 'win_away', 'draw', 'win']].head(10).to_dict())
{'date': {0: '2008-08-17 00:00:00', 1: '2008-08-16 00:00:00', 2: '2008-08-16 00:00:00', 3: '2008-08-17 00:00:00', 4: '2008-08-16 00:00:00', 5: '2008-09-24 00:00:00', 6: '2008-08-16 00:00:00', 7: '2008-08-16 00:00:00', 8: '2008-08-16 00:00:00', 9: '2008-11-01 00:00:00'}, 'home_team_api_id': {0: 9987, 1: 10000, 2: 9984, 3: 9991, 4: 7947, 5: 8203, 6: 9999, 7: 4049, 8: 10001, 9: 8342}, 'away_team_api_id': {0: 9993, 1: 9994, 2: 8635, 3: 9998, 4: 9985, 5: 8342, 6: 8571, 7: 9996, 8: 9986, 9: 8571}, 'win_home': {0: 0, 1: 0, 2: 0, 3: 1, 4: 0, 5: 0, 6: 0, 7: 0, 8: 1, 9: 1}, 'win_away': {0: 0, 1: 0, 2: 1, 3: 0, 4: 1, 5: 0, 6: 0, 7: 1, 8: 0, 9: 0}, 'draw': {0: 1, 1: 1, 2: 0, 3: 0, 4: 0, 5: 1, 6: 1, 7: 0, 8: 0, 9: 0}, 'win': {0: 0, 1: 0, 2: 1, 3: 1, 4: 1, 5: 0, 6: 0, 7: 1, 8: 1, 9: 1}}
output for print(df_team_attributes[['date', 'team_api_id', 'buildUpPlaySpeed', 'buildUpPlaySpeedClass']].head(10).to_dict())
{'date': {0: '2010-02-22 00:00:00', 1: '2014-09-19 00:00:00', 2: '2015-09-10 00:00:00', 3: '2010-02-22 00:00:00', 4: '2011-02-22 00:00:00', 5: '2012-02-22 00:00:00', 6: '2013-09-20 00:00:00', 7: '2014-09-19 00:00:00', 8: '2015-09-10 00:00:00', 9: '2010-02-22 00:00:00'}, 'team_api_id': {0: 9930, 1: 9930, 2: 9930, 3: 8485, 4: 8485, 5: 8485, 6: 8485, 7: 8485, 8: 8485, 9: 8576}, 'buildUpPlaySpeed': {0: 60, 1: 52, 2: 47, 3: 70, 4: 47, 5: 58, 6: 62, 7: 58, 8: 59, 9: 60}, 'buildUpPlaySpeedClass': {0: 'Balanced', 1: 'Balanced', 2: 'Balanced', 3: 'Fast', 4: 'Balanced', 5: 'Balanced', 6: 'Balanced', 7: 'Balanced', 8: 'Balanced', 9: 'Balanced'}}
Have you tried casting the your date columns into the correct format and then attempting the merge? The following worked for me based on the example that you provided -
# Casting to date
df_match["date"] = pd.to_datetime(df_match["date"])
df_team_attributes["date"] = pd.to_datetime(df_match["date"])
# Merging on the date field alone
df_team_performance = pd.merge(df_team_attributes, df_match,
how = 'left',
on = 'date')
# Filtering out the required rows
result = df_team_performance.query("(team_api_id == home_team_api_id) | (team_api_id == away_team_api_id)")
Please let me know if my understanding of your question is correct.
I have a data set that I need to reformat so that I can plot and work with it further. It is sort of an transpose action but I am struggling to not overwrite the data in the new dataframe. I sorted out the headings using dictionaries and it maps the fields from the original df to the new output df correctly. It is just overwriting the first entry and not adding a new POLY/POLY_NAME
Input dataframe:
Output dataframe:
Below is my code so far:
import pandas as pd
fractions = {"A": 1.35, "B": 1.40, "C": 1.45}
quality = {"POLY_NAME":"POLY", "AS":"Ash", "CV":"CV","FC":"FC","MS":"Moist","TS":"Tots","VM":"Vols","YL":"Yield"}
frac = list(fractions.values())
headers = list(quality.values())
df = pd.DataFrame(columns=headers, index=frac)
wash_dic = {'POLY_NAME': {0: 'Asset 1', 1: 'Asset 2', 2: 'Asset 3'},
'RD': {0: 1.63, 1: 1.63, 2: 1.57},
'SEAMTH': {0: 3.02, 1: 3.02, 2: 3.37},
'AAS': {0: 7.76, 1: 7.34, 2: 7.24},
'ACV': {0: 28.98, 1: 29.18, 2: 29.27},
'AFC': {0: 54.95, 1: 53.55, 2: 52.38},
'AMS': {0: 4.22, 1: 4.26, 2: 4.63},
'ATS': {0: 0.97, 1: 1.09, 2: 1.23},
'AVM': {0: 33.07, 1: 34.85, 2: 35.75},
'AYL': {0: 0.4, 1: 0.95, 2: 0.75},
'BAS': {0: 9.28, 1: 9.27, 2: 9.58},
'BCV': {0: 28.17, 1: 28.33, 2: 28.09},
'BFC': {0: 56.21, 1: 54.39, 2: 52.11},
'BMS': {0: 4.25, 1: 4.25, 2: 4.61},
'BTS': {0: 0.84, 1: 1.01, 2: 1.22},
'BVM': {0: 30.25, 1: 32.08, 2: 33.7},
'BYL': {0: 3.11, 1: 5.44, 2: 4.36},
'CAS': {0: 11.01, 1: 10.96, 2: 11.25},
'CCV': {0: 27.31, 1: 27.53, 2: 27.39},
'CFC': {0: 58.09, 1: 56.0, 2: 53.43},
'CMS': {0: 4.41, 1: 4.38, 2: 4.62},
'CTS': {0: 0.63, 1: 0.83, 2: 0.98},
'CVM': {0: 26.5, 1: 28.66, 2: 30.71},
'CYL': {0: 13.45, 1: 16.11, 2: 12.94}}
wash = pd.DataFrame(wash_dic)
wash
for label, content in wash.items():
print('fraction:', fractions.get(label[0]), ' quality:', quality.get(label[-2:]))
for c in content:
try:
df.loc[fractions.get(label[0]), quality.get(label[-2:])] = c
except:
pass
I have tried to add another for loop but the logic is escaping me currently.
Required outcome as dictionary
outcome = {'Unnamed: 0': {0: 1.35,
1: 1.4,
2: 1.45,
3: 1.35,
4: 1.4,
5: 1.45,
6: 1.35,
7: 1.4,
8: 1.45},
'POLY': {0: 'Asset 1',
1: 'Asset 1',
2: 'Asset 1',
3: 'Asset 2',
4: 'Asset 2',
5: 'Asset 2',
6: 'Asset 3',
7: 'Asset 3',
8: 'Asset 3'},
'Ash': {0: 7.76,
1: 9.28,
2: 11.01,
3: 7.34,
4: 9.27,
5: 10.96,
6: 7.24,
7: 9.58,
8: 11.25},
'CV': {0: 28.98,
1: 28.17,
2: 27.31,
3: 29.18,
4: 28.33,
5: 27.53,
6: 29.27,
7: 28.09,
8: 27.39},
'FC': {0: 54.95,
1: 56.21,
2: 58.09,
3: 53.55,
4: 54.39,
5: 56.0,
6: 52.38,
7: 52.11,
8: 53.43},
'Moist': {0: 4.22,
1: 4.25,
2: 4.41,
3: 4.26,
4: 4.25,
5: 4.38,
6: 4.63,
7: 4.61,
8: 4.62},
'Tots': {0: 0.97,
1: 0.84,
2: 0.63,
3: 1.09,
4: 1.01,
5: 0.83,
6: 1.23,
7: 1.22,
8: 0.98},
'Vols': {0: 33.07,
1: 30.25,
2: 26.5,
3: 34.85,
4: 32.08,
5: 28.66,
6: 35.75,
7: 33.7,
8: 30.71},
'Yiels': {0: 0.4,
1: 3.11,
2: 13.45,
3: 0.95,
4: 5.44,
5: 16.11,
6: 0.75,
7: 4.36,
8: 12.94}}
Regards
I resolved to duplicate/overwriting of the values by first grouping the original wash DF and then in the for loop and the data of each loop into a blank DF and at the end of the loop append it to the Final DF. Just for neatness I made the index column a normal column and reordered the columns.
groups = wash.groupby("POLY_NAME")
df_final = pd.DataFrame(columns=headers)
for name, group in groups:
df = pd.DataFrame(columns=headers)
for label, content in group.items():
if quality.get(label[-2:]) in headers:
#print(label)
#print(name)
#print(label, content)
for c in content:
try:
df.loc[fractions.get(label[0]), "POLY"] = name
df.loc[fractions.get(label[0]), quality.get(label[-2:])] = c
#print('Poly:', name, ' fraction:', fractions.get(label[0]), ' quality:', quality.get(label[-2:]))
except:
pass
df_final = df_final.append(df)
df_final = df_final.reset_index().rename({'index':'FLOAT'}, axis = 'columns')
df_final = df_final.reindex(columns=["POLY","FLOAT","Ash","CV","FC","Moist","Tots","Vols","Yield"])
Might not be the neatest or fastest method but it gives the required results.
There is a huge df with multiple columns but want to read only specific column that is interested to me:
in the below data, I would like to read only the column 'Type 1'
import numpy as np
import pandas as pd
data = {'Type 1': {0: 1, 1: 3, 2: 5, 3: 'HH', 4: 9, 5: 11, 6: 13, 7: 15, 8: 17},
'Type 2': {0: 'AA',
1: 'BB',
2: 'np.NaN',
3: '55',
4: '3.14',
5: '-96',
6: 'String',
7: 'FFFFFF',
8: 'FEEE'},
'Type 3': {0: 0, 1: 0, 2: 0, 3: 0, 4: 0, 5: 0, 6: 0, 7: 0, 8: 0},
'Type 4': {0: '23',
1: 'fefe',
2: 'abcd',
3: 'dddd',
4: 'dad',
5: 'cfe',
6: 'cf42',
7: '321',
8: '0'},
'Type 5': {0: -120,
1: -120,
2: -120,
3: -120,
4: -120,
5: -120,
6: -120,
7: -120,
8: -120}}
df = pd.DataFrame(data)
df
int_count = df['Type 1'].count(0,numeric_only = True) # should count only cells that contain integers and return 8
total_count = df['Type 1'].count(0,numeric_only = False) # should count all the cells and return 9
I want something like count only the numeric values in particular column
eg: df['Type 1'].count(0,numeric_only = True) should return 8 (exclude counting the string 'HH' in Type 1 column)
df['Type 1'].count(0,numeric_only = False) should return 9 (total number of cells in the particular column)
but "df['Type 1'].count(0,numeric_only = True/False)" this is not working as I expect...
I would suggest the below:
int_count = len(df.loc[df['Type 1'].astype(str).str.isnumeric()])
total_count = len(df)
There are multiple columns in the df, out of which only selected columns has to be converted from hexa decimal to decimal
Selected column names are stored in a list A = ["Type 2", "Type 4"]
{'Type 1': {0: 1, 1: 3, 2: 5, 3: 7, 4: 9, 5: 11, 6: 13, 7: 15, 8: 17},
'Type 2': {0: 'AA',
1: 'BB',
2: 'CC',
3: '55',
4: '88',
5: '96',
6: 'FF',
7: 'FFFFFF',
8: 'FEEE'},
'Type 3': {0: 0, 1: 0, 2: 0, 3: 0, 4: 0, 5: 0, 6: 0, 7: 0, 8: 0},
'Type 4': {0: '23',
1: 'fefe',
2: 'abcd',
3: 'dddd',
4: 'dad',
5: 'cfe',
6: 'cf42',
7: '321',
8: '0'},
'Type 5': {0: -120,
1: -120,
2: -120,
3: -120,
4: -120,
5: -120,
6: -120,
7: -120,
8: -120}}
Say, you have the string "AA" in hex.
You can convert hex to decimal like this:
str(int("AA", 16))
Similarly, for a dataframe column that has hexadecimal values, you can use a lambda function.
df['Type2'] = df['Type2'].apply(lambda x: str(int(str(x), 16)))
Assuming, df is the name of the imported dataframe.
You can use pandas.DataFrame.applymap to cast element-wise:
>>> df[["Type 2", "Type 4"]].applymap(lambda n: int(n, 16))
Type 2 Type 4
0 170 35
1 187 65278
2 204 43981
3 85 56797
4 136 3501
5 150 3326
6 255 53058
7 16777215 801
8 65262 0