Offline inference and metric calculation library
Project description
crossfit
Multi Node Multi GPU Offline Inference and metric calculation library
Installation
Install this library using pip:
pip install crossfit
Installation from source (for cuda 12.x)
git clone https://github.com/rapidsai/crossfit.git
cd crossfit
pip install --extra-index-url https://pypi.nvidia.com ".[cuda12x]"
Usage
Usage instructions go here.
Development
To contribute to this library, first create a conda environment with the necessary dependencies:
cd crossfit
mamba env create -f conda/environments/cuda_dev.yaml
conda activate crossfit_dev
Now install the dependencies and test dependencies:
pip install -e '.[test]'
To run the tests:
pytest
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
crossfit-0.0.6.tar.gz
(96.1 kB
view details)
Built Distribution
crossfit-0.0.6-py3-none-any.whl
(177.3 kB
view details)
File details
Details for the file crossfit-0.0.6.tar.gz.
File metadata
- Download URL: crossfit-0.0.6.tar.gz
- Upload date:
- Size: 96.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 6a83502810e3417b5f857f0cabcc1552fa1dc0c1d61f1a47ebf7df6d963c5f66 |
|
| MD5 | fe55cd25ad2baa0ae506aa51b8e3af75 |
|
| BLAKE2b-256 | 0180f63f6386f9be9caf0388af36d7dc7a9e56c8d72b16b0abe301762b2a0c04 |
File details
Details for the file crossfit-0.0.6-py3-none-any.whl.
File metadata
- Download URL: crossfit-0.0.6-py3-none-any.whl
- Upload date:
- Size: 177.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 807ebb2c28de50039213338052159bd8596937fc6405e6b11ea3150e8eadf76f |
|
| MD5 | 47ac57992eb2abfc5b1a704dae3487d8 |
|
| BLAKE2b-256 | 7b0c99c7c7080bf4020fa704d722c38144bff7388bd8d592d68bd2890835a3f5 |
