Inline completion provider using tranformers.js for JupyterLab.
Project description
jupyterlab-transformers-completer
Inline completion provider using transformers.js
for JupyterLab
This extension is aimed for developers of JupyterLab extensions (and advanced JupyterLab users) to explore the integration of the inline completions API added in JupyterLab 4.1.
All models linked from this demonstration run exclusively in your browser, and are:
- order of magnitudes smaller than the state-of-the-art models,
- producing correspondingly lower accuracy of suggestions/answers.
These models are not vetted for accuracy nor propriety and should not be deployed without further validation.
Usage
- Go to Settings → Inline Completer → choose the models for code (in code cells and scripts) and text (in markdown cells and plain files) generation.
- The models will be downloaded, compiled, and cached in your browser as indicated by pop-up notifications in bottom right corner.
- Start typing a few words in the code cell or Markdown cell and observe the suggestions; hover over to see shortcuts.
- Adjust model configuration in settings as needed; in particular increasing the repetition penalty, adjusting temperature and top k is recommended.
Known issues
- Sometimes it is required to go to settings after installation and modify settings to trigger model download and compilation
- Sometimes the browser will cache a faulty (e.g not fully downloaded) file resulting in Syntax Error when parsing; you can try in an incognito/private mode to confirm that this is transient and clear browser cache to remove such file.
Requirements
- JupyterLab >= 4.1.0 or Jupyter Notebook >= 7.1.0
- A browser supporting:
SharedArrayBuffer
- Web Workers
- Dynamic import for workers (behind
dom.workers.modules.enabled
in Firefox) - (optional, for faster inference) WebGPU (behind
dom.webgpu.enabled
in Firefox)
jupyter-server
to enable additional headers (jupyverse
andjupyterlite
not tested yet)
When this extension is enabled, the server will return additional headers, which will prevent fetching external resources, for example the extension logos from GitHub will no longer load in the extension panel.
The additional headers are used to enable synchronous communication with WebWorker via SharedArrayBuffer
:
Cross-Origin-Opener-Policy: same-origin,
Cross-Origin-Embedder-Policy: require-corp
Install
To install the extension, execute:
pip install -U 'jupyterlab>=4.1.0' jupyterlab-transformers-completer
Uninstall
To remove the extension, execute:
pip uninstall jupyterlab-transformers-completer
Contributing
Development install
Note: You will need NodeJS to build the extension package.
The jlpm
command is JupyterLab's pinned version of
yarn that is installed with JupyterLab. You may use
yarn
or npm
in lieu of jlpm
below.
# Clone the repo to your local environment
# Change directory to the jupyterlab-transformers-completer directory
# Install package in development mode
pip install -e "."
# Link your development version of the extension with JupyterLab
jupyter labextension develop . --overwrite
# Rebuild extension Typescript source after making changes
jlpm build
You can watch the source directory and run JupyterLab at the same time in different terminals to watch for changes in the extension's source and automatically rebuild the extension.
# Watch the source directory in one terminal, automatically rebuilding when needed
jlpm watch
# Run JupyterLab in another terminal
jupyter lab
With the watch command running, every saved change will immediately be built locally and available in your running JupyterLab. Refresh JupyterLab to load the change in your browser (you may need to wait several seconds for the extension to be rebuilt).
By default, the jlpm build
command generates the source maps for this extension to make it easier to debug using the browser dev tools. To also generate source maps for the JupyterLab core extensions, you can run the following command:
jupyter lab build --minimize=False
Development uninstall
pip uninstall jupyterlab-transformers-completer
In development mode, you will also need to remove the symlink created by jupyter labextension develop
command. To find its location, you can run jupyter labextension list
to figure out where the labextensions
folder is located. Then you can remove the symlink named @jupyterlab/transformers-completer
within that folder.
Packaging the extension
See RELEASE
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file jupyterlab_transformers_completer-0.3.0.tar.gz
.
File metadata
- Download URL: jupyterlab_transformers_completer-0.3.0.tar.gz
- Upload date:
- Size: 325.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.8.6
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9bdb4352ccbb95b1222d8a7c36c0fea778f191fdb0958ff0691164da57437fbb |
|
MD5 | dda7024376f0d41c04eb3b4a7d26325f |
|
BLAKE2b-256 | 8a0b48844fd53bc5132e73d04f47e0fd55943ae83975a16a3015cc2909b113ae |
Provenance
File details
Details for the file jupyterlab_transformers_completer-0.3.0-py3-none-any.whl
.
File metadata
- Download URL: jupyterlab_transformers_completer-0.3.0-py3-none-any.whl
- Upload date:
- Size: 232.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.8.6
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 00c1c73b1c49506c83809c30501de19b50066b4dc38e2320d59512aecbb192cf |
|
MD5 | df475462ab66ad928420386bca996075 |
|
BLAKE2b-256 | 9725ebce5ec8c42fef4745c019d9b64f54bdd1b2e52f297d371fe3e96fca25e6 |