dataprep 0.1.0a2

Dataprep is a library to help data scientist accomplish all the tasks using one library before building the predictive model.

DataPrep

Documentation | Mail List & Forum

Dataprep is a collection of functions that helps you accomplish tasks before you build a predictive model.

Implementation Status

Currently, you can use dataprep to:

• Collect data from common data sources (through dataprep.data_connector)
• Do your exploratory data analysis (through dataprep.eda)
• ...

Installation

pip install dataprep==0.1.0a2

dataprep is in its alpha stage now, so please manually specific the version number.

Examples & Usages

More detailed examples can be found at the examples folder.

Data Connector

You can download Yelp business search result into a pandas DataFrame, using two lines of code, without taking deep looking into the Yelp documentation!

from dataprep.data_connector import Connector
# Put "yelp" as the first parameter to indicate we want to download some data from Yelp.
# You also need the Yelp access token for programmably access. 
dc = Connector("yelp", auth_params={"access_token":"<Your yelp access token>"})
# Here we want to download data from the "restaurant" endpoint 
# with term "ramen" and location "vancouver"
df = dc.query("businesses", term="ramen", location="vancouver")
# df will be a pandas dataframe.

EDA

There are common tasks during the exploratory data analysis stage, like a quick look at the columnar distribution, or understanding the correlations between columns.

The EDA module categorizes these EDA tasks into functions helping you finish EDA tasks with a single function call.

• Want to understand the distributions for each DataFrame column? Use plot.

from dataprep.eda import plot

df = ...

plot(df)

• Want to understand the correlation between columns? Use plot_correlation.

from dataprep.eda import plot_correlation

df = ...

plot_correlation(df)

• Or, if you want to understand the impact of the missing values for each column, use plot_missing.

from dataprep.eda import plot_missing

df = ...

plot_missing(df)

• You can even drill down to get more information by given plot, plot_correlation and plot_missing a column name.

df = ...

plot_missing(df, x="some_column_name")

Don't forget to checkout the examples folder for detailed demostration!

Contribution

Contribution is always welcome. If you want to contribute to dataprep, be sure to read the contribution guidelines.