LDSI
Project description
LDSI (Laiss Declarative Self Improve) is a powerful framework for programming and optimizing language models. It empowers developers to rapidly build, iterate, and enhance modular AI systems with the following key features:
-
Systematic Optimization: LDSI offers a range of optimizers to refine your AI programs, from generating improved instructions to fine-tuning model weights.
-
Modular Architecture: Build complex AI systems using pre-defined, interchangeable modules, simplifying the development process and reducing reliance on intricate prompting techniques.
-
Cross-LM Compatibility: Seamlessly integrate and enhance performance across various language models, from state-of-the-art GPT models to local options like T5-base or Llama2.
-
Declarative Approach: Define your AI system's behavior and goals declaratively, allowing LDSI to handle the underlying optimization and execution details.
-
Rapid Prototyping: Accelerate your development cycle with LDSI's intuitive API and built-in tools for testing and evaluation.
LDSI is designed to streamline the process of creating sophisticated AI applications, making it an invaluable tool for both researchers and practitioners in the field of natural language processing and AI development.
Installation
pip install ldsi
To install the very latest from main:
pip install git+https://github.com/laiss-labs/laiss-dsi.git
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
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 ldsi-1.0.3.tar.gz.
File metadata
- Download URL: ldsi-1.0.3.tar.gz
- Upload date:
- Size: 187.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.0.1 CPython/3.12.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8b7951335780fe7c3d3d17d1e79febfb53a9dd8399a4797165927df4c662e20b
|
|
| MD5 |
b8a67800af461c36824869d151f429f9
|
|
| BLAKE2b-256 |
ced03ff421022b0940e03fb84eed82b071505de34df532175de833187e315d27
|
File details
Details for the file ldsi-1.0.3-py3-none-any.whl.
File metadata
- Download URL: ldsi-1.0.3-py3-none-any.whl
- Upload date:
- Size: 243.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.0.1 CPython/3.12.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
94cad77df083da53e102647698892259c6e1f68c54e5245ecdee2128e3424573
|
|
| MD5 |
534aef9797de7120e118567bd9f106e0
|
|
| BLAKE2b-256 |
1638f83e47f775ec904a4e121b1d74513998354c9ca8c7f542d8e3b8d3b3c4ad
|