Hi and Welcome
If you want to chose Data Science or Bioinformatics as a career, fetching, cleaning & organizing the data in a specific shape will be your day to day task before applying any analysis technique or machine learning algorithms. Here in this post, I'll be sharing few quick and basic pandas codes which will help you get your data in shape.
Pandas library in python is built on NumPy; It is a fundamental library in python for data analysis, pre-processing , e.t.c. it is rich in methods used for data munging & a quick Exploratory Data Analysis (EDA). There are two main objects in pandas, they are pandas Series & DataFrame.
You can see it as a database whose primary purpose is not for storing data, but easily manipulate and query data, which are features of every database management system.
The pandas spells covered in this article are :
- Importing Pandas
- Reading Datasets from files
- Exporting Data
- Create test objects
- Viewing/Inspecting Data
- Selection / Filtering a specific data
- Data Cleaning / Wrangling
- Filter, Sort & Groupby
- Join & Combine
- Statistics
Enough! Let's get down to business.
1. Importing Pandas
import pandas as pd
2. Reading Datasets from files
Basic file types pandas can handle are CSV, Excel file, html file and the list goes on ...!
- Reading CSV file
data = pd.read_csv(".../FilePath/file.csv") - Reading TSV file
data = pd.read_table(".../FilePath/file.tsv") - Reading Excel file
data = pd.read_excel(".../FilePath/file.xls") - Read from a SQL table/database
data = pd.read_sql(query, connection_object) - Read from a JSON formatted string, URL or file
data = pd.read_json(json_string) - From a dict, keys for columns names, values for data as lists
data = pd.DataFrame(dict) - Takes the contents of your clipboard and passes it to read_table()
data = pd.read_clipboard() - Parses an html URL, string or file and extracts tables to a list of dataframes
data = pd.read_html(url)
3. Exporting Data
Use these commands to export a DataFrame to CSV, .xlsx, SQL, or JSON.
[Here df is a DataFrame fetched from a dataset or a file ]
- Write to a CSV file
df.to_csv("filename.csv") - Write to an Excel file
df.to_excel("filename.xls") - Write to a SQL table
df.to_sql(table_name, connection_object) - Write to a file in JSON format
df.to_json("filename.json")
4. Create test objects
Before we proceed let's import one more essential module in Python
import numpy as np
These commands can be useful for creating test segments
For when you're in a mood to play
- Create 5 columns and 20 rows of random floats
pd.DataFrame(np.random.rand(20,5)) - Create a series from an iterable my_list
pd.Series(my_list) - Add a date index
df.index = pd.date_range('1900/1/30', periods=df.shape[0])
5. Viewing/Inspecting Data
Use these commands to take a look at specific sections of your pandas DataFrame or Series instead of scrolling all the way top just to look at your data.
Rule of thumb: Don't be shy
- First n rows of the DataFrame
df.head(n) - Last n rows of the DataFrame
df.tail(n) - Number of rows and columns
df.shape - Index, Datatype and Memory information
df.info() - Summary statistics for numerical columns
df.describe() - View unique values and counts
s.value_counts() - Unique values and counts for all columns
df.apply(pd.Series.value_counts)
6. Selection / Filtering a specific data
Use these commands to select a specific subset of your data.
Nobody is asking you to not be picky
[Here s implies that it's a Series]
- Returns column with label col as Series
df[col] - Returns columns as a new DataFrame
df[[col1, col2]] - Selection by index
s.loc['index_one'] - Selection by position
s.iloc[0] - First row
df.iloc[0,:] - First element of first column
df.iloc[0,0]
An easy way to remember how loc & iloc work is that i in iloc stands for integer i.e. iloc takes in integers as row & column values while loc accepts row number & column names.
Trust me,
ilochas a crush on integer
7. Data Cleaning / Wrangling
Use these commands to perform a variety of data cleaning tasks.
Oh.. GOD I love these ...
- Rename columns
df.columns = ['a','b','c'] - Checks for null Values, Returns Boolean Arrray
pd.isnull() - Opposite of pd.isnull()
pd.notnull() - Drop all rows that contain null values
df.dropna() - Drop all columns that contain null values
df.dropna(axis=1) - Drop all rows have have less than n non null values
df.dropna(axis=1,thresh=n) - Replace all null values with x
df.fillna(x) - Replace all null values with the mean (mean can be replaced with almost any function from the statistics module)
s.fillna(s.mean()) - Convert the datatype of the series to float
s.astype(float) - Replace all values equal to 1 with 'one'
s.replace(1,'one') - Replace all 1 with 'one' and 3 with 'three'
s.replace([1,3],['one','three']) - Mass renaming of columns
df.rename(columns=lambda x: x + 1) - Selective renaming
df.rename(columns={'old_name': 'new_ name'}) - Change the index
df.set_index('column_one') - Mass renaming of index
df.rename(index=lambda x: x + 1)
8. Filter, Sort & Groupby
Use these commands to filter, sort, and group your data.
You see Data doesn't like to be lonely, neither with a lot of it's own kind ! pretty, selectively social.
- Rows where the column col is greater than 0.5
df[df[col] > 0.5] - Rows where 0.7 > col > 0.5
df[(df[col] > 0.5) & (df[col] < 0.7)] - Sort values by col1 in ascending order
df.sort_values(col1) - Sort values by col2 in descending order
df.sort_values(col2,ascending=False) - Sort values by col1 in ascending order then col2 in descending order
df.sort_values([col1,col2],ascending=[True,False]) - Returns a groupby object for values from one column
df.groupby(col) - Returns groupby object for values from multiple columns
df.groupby([col1,col2]) - Returns the mean of the values in col2, grouped by the values in col1 (mean can be replaced with almost any function from the statistics module)
df.groupby(col1)[col2] - Create a pivot table that groups by col1 and calculates the mean of col2 and col3
df.pivot_table(index=col1,values=[col2,col3],aggfunc=mean) - Find the average across all columns for every unique col1 group
df.groupby(col1).agg(np.mean) - Apply the function np.mean() across each column
df.apply(np.mean) - Apply the function np.max() across each row
nf.apply(np.max,axis=1)
9. Join & Combine
Use these commands to combine multiple dataframes into a single one.
Urgh...! Why ? Data Why?...
- Add the rows in df1 to the end of df2 (columns should be identical)
df1.append(df2) - Add the columns in df1 to the end of df2 (rows should be identical)
pd.concat([df1, df2],axis=1) - SQL-style join the columns in df1 with the columns on df2 where the rows for col have identical values.
howcan be one ofleft,right,outer,inner. [Useful with dataframes constitute rows of varying number]df1.join(df2,on=col1,how='inner')
10. Statistics
Use these commands to perform various statistical tests. (These work on series as well.)
Short & Sweet
- Summary statistics for numerical columns
df.describe() - Returns the mean of all columns
df.mean() - Returns the correlation between columns in a DataFrame
df.corr() - Returns the number of non-null values in each DataFrame column
df.count() - Returns the highest value in each column
df.max() - Returns the lowest value in each column
df.min() - Returns the median of each column
df.median() - Returns the standard deviation of each column
df.std()
Conclusion : Obtaining the Super Powers of the pandas library in Python is a must for one to be very successful at data manipulation, especially if you want to reduce the time taken in data cleansing.
Hooraay...! Finally ...
that was a lot I know, but this article isn't going anywhere, take your time learn them slowly. P.s. make sure to bookmark it.
The best way to learn these is to learn these.
Okay fineeee.... this was the last one. If you have made it till here, You're AWESOME... ! Checkout how I have used pandas to visualize relationship between COVID-19 & World Happiness on Kaggle. Warning [cute COVID picture in the above article]
Support me by hitting the subscribe button on my YouTube channel CodeBeast