Private Libraries

Zepl can leverage any Python, R, or Scala / Spark library whether it is stored in a public repository or private. Teams that build their own custom tools that are leveraged by their data teams will have custom libraries that they need to access in their notebooks. In Zepl, you can upload any private library and install that library as part of a Zepl Image. This gives you full control over your shared Data Science workflow in Zepl.
What are Private Libraries
A Private Library is a packaged file specific to your programming language, which contains custom functions or packages. Please ensure that private libraries are in the following standard formats for each respective programming language:
Python: .tar, .gz, and .whl 
Scala / Spark: .jar 
R: .tar 
Using Private Libraries
Create a packaged library
Please follow standard practices for bundling your library for each programming language before uploading to Zepl. This will ensure Zepl can install your library to an Image. The following links should provide information on how to properly bundle your packages:
Python: https://packaging.python.org/en/latest/tutorials/packaging-projects/​
R: https://cran.r-project.org/doc/manuals/r-release/R-admin.html​
Scala / Spark: https://docs.scala-lang.org/tutorials/scala-with-maven.html#creating-a-jar​
How Zepl installs your Private Libraries to an Image
It might be helpful to know how Zepl installs your Private Library to an Image. When your library name is added to a new image, the image builder will attempt to install your library using a language specific package manger. Python uses pip, R uses a built in package manager, and Scala / Spark uses Maven. Your library will be installed using the following package manager install statement:
Python: pip install <private library name>==<version>
R: install.packages("<private library name>")
Scala / Spark: .jar is added to the container $SPARK_HOME location
Upload a library
1.Navigate to Resources > Private Libraries
2.Select Add Library
3.Enter a Name (no spaces)
The name you enter here will be used when you build this library into an Image
4.Description (optional)
5.Use the dropdown to specify the Language corresponding to this package
The language you select here will correspond to the language used to import this library in an Image
6.Upload your packaged file
Install in an new Image
1.Navigate to Resources > Images
2.Select Create New
3.Add a Name and Description (optional)
4.Select the language corresponding the type of package uploaded in the previous step
Import in a Notebook
After your image completes the build process you will be able start using it in any notebook!
1.Create a new notebook or Open an existing notebook
2.Select Settings
3.Select the new Image that was just built with your Private Library
4.Create a paragraph and specify the interpreter corresponding to the language of your Private Library
5.In the code, import your library

Last modified 1yr ago