Skip to main content

Automatically reload modified Python modules in notebooks and scripts.

Project description

PyPI Version Test Coverage Read the Docs Status View on GitHub

Overview

LiveImport reliably and automatically reloads Python modules in Jupyter notebooks. Unlike IPython's autoreload extension, LiveImport reloads are well-defined, deterministic operations that follow the semantics of a developer's import statements exactly. It maintains consistency between modules by automatically reloading dependent modules as needed, ensuring up-to-date references across a notebook and its modules.

LiveImport is designed for developers who interactively build Jupyter notebooks together with external Python code, and who want predictable reloading with minimal mystery.

Given a cell like

%%liveimport
from common import *
from nets import ConvNet, ResidualNet as ResNet
import hyperparam as hp

LiveImport will execute the import statements, then automatically reload common, nets, or hyperparam whenever their source files change. When LiveImport reloads, it will rebind names in the notebook as described by the import statements. If nets imports from hyperparam, then when hyperparam is modified, LiveImport will automatically reload nets after hyperparam.

To make the cell above transparent to IDEs like Visual Studio Code, you can hide the cell magic:

#_%%liveimport
from common import *
from nets import ConvNet, ResidualNet as ResNet
import hyperparam as hp

Hidden cell magic is a user experience feature tailored for modern notebook development. Others include tracking indirectly imported modules, protection against reloading in the middle of multi-cell runs, and optional reload notification.

If you currently use autoreload, you might consider comparing LiveImport to autoreload.

Documentation

See liveimport.readthedocs.io for a user guide and an API reference.

Installation

You can install LiveImport from PyPI with

$ pip install liveimport

You can also clone the repository and install it directly.

$ git clone https://github.com/escreven/liveimport.git
$ cd liveimport
$ pip install .

LiveImport requires Python 3.10 or greater and IPython 7.23.1 or greater.

Reliability

LiveImport includes automated tests with 100% code coverage, which are run on MacOS, Linux, and Windows with Python 3.10, 3.11, 3.12, 3.13, and 3.14 using both the oldest supported and latest versions of IPython. Notebook integration features are tested using notebook 5.7.0 and notebook latest. See the GitHub workflow for details.

If you have a copy of the source, you can run the tests yourself with

$ python test/main.py

Questions or Issues

If you have any questions or encounter any issues, please submit them on GitHub.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

liveimport-1.2.4.tar.gz (18.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

liveimport-1.2.4-py3-none-any.whl (18.5 kB view details)

Uploaded Python 3

File details

Details for the file liveimport-1.2.4.tar.gz.

File metadata

  • Download URL: liveimport-1.2.4.tar.gz
  • Upload date:
  • Size: 18.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.10

File hashes

Hashes for liveimport-1.2.4.tar.gz
Algorithm Hash digest
SHA256 e575349e0cd929625fb09a4fe4bdf4d45e190c07d6e51fba9d1fe22af064600b
MD5 be5ea27c5806d2bef4337332f331e021
BLAKE2b-256 7688bf0bea6aecf8f33d0e2189b1c763724651733d1729ba2c5df1b18d45aec4

See more details on using hashes here.

File details

Details for the file liveimport-1.2.4-py3-none-any.whl.

File metadata

  • Download URL: liveimport-1.2.4-py3-none-any.whl
  • Upload date:
  • Size: 18.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.10

File hashes

Hashes for liveimport-1.2.4-py3-none-any.whl
Algorithm Hash digest
SHA256 49585a07a7a8375d8209d687cb2557a84defe6ab88269621314f4d2c2b0727d1
MD5 d1bd45945b365ce0553c3a6fd907f07b
BLAKE2b-256 d29e6e90222bac15d1fff8dd72324645c0e9c87c4cfcf8b8bb38ea737a61604b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page