A aibrary and CLI for performing experiments with LLMs.
Project description
MIMiC - Modeulau Iaith Mawr i'r Cymraeg (Large Language Models for Welsh)
This package provides a utilities for performing training and inference with LLMs, and a demo gradio application showcasing the inference on trained models.
Each of the components is exposed as library functions but also have command line interface (CLI) counterpart.
For CLI usage, see the inline help provided by the CLI:
python -m techiaith.mimic --help
techiaith-mimic is developed and maintained by the Uned Technolegau Iaith (UTI) https://techiaith.cymru/ team, backed by Bangor University.
https://bangor.ac.uk/.
Techiaith UTI is a self-funded research unit that develops technologies for the Welsh language.
To learn more about who specifically contributed to this codebase, see our contributors page.
This code is made available under the MIT License.
Install
Install this package from PyPI with pip using a venv or mamba/conda environment:
pip install techiaith-mimic
or directly from github:
pip install git@github.com:techiaith/mimic.git
Development
Configure your development environment to use the tools refered to in the pyproject.toml file (mypy, ruff, black).
Releases - Publishing to PyPI
Change the version as appropriate in pyproject.toml, then build, check and upload:
pip install build twine
python -m build
python -m twine check dist/*
python -m twine upload dist/*
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
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file techiaith_mimic-2025.1.28.dev2.tar.gz.
File metadata
- Download URL: techiaith_mimic-2025.1.28.dev2.tar.gz
- Upload date:
- Size: 10.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.0.1 CPython/3.10.16
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
bcc3b8e0b265952bb1e3c723f6c57b4ef8b93b8e79f0e54ebfea57657ca9c2b5
|
|
| MD5 |
bfe33d2d0f3967dacb46e962eebe968d
|
|
| BLAKE2b-256 |
1ce5e89d1300a24ad289e76d8a6864b0c2579bfc2998575281556fec5eb9cd48
|
File details
Details for the file techiaith_mimic-2025.1.28.dev2-py3-none-any.whl.
File metadata
- Download URL: techiaith_mimic-2025.1.28.dev2-py3-none-any.whl
- Upload date:
- Size: 9.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.0.1 CPython/3.10.16
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
4bbe83d11f712b018124a54e4e56fe6bff5f4cbabc4a2e9ba0394cb263b33bde
|
|
| MD5 |
c6ef1c2a8251924ca78d7a5fefba97ed
|
|
| BLAKE2b-256 |
0a5fa056b4b36a39beedfb61412b74030b40f3c60bdc1dd062cdb523c3db7619
|