Accelerate was created for PyTorch users who like to write the training loop of PyTorch models but are reluctant to write and maintain the boilerplate code needed to use multi-GPUs/TPU/fp16. Accelerate abstracts exactly and only the boilerplate code related to multi-GPUs/TPU/fp16 and leaves the rest of your code unchanged.
| Uploaded | Mon Mar 31 21:55:21 2025 |
| md5 checksum | edbc2f9cb2064296be9c1a8011b683e4 |
| arch | x86_64 |
| build | py312h06a4308_0 |
| depends | huggingface_hub >=0.21.0, numpy >=1.17,<2.0.0, packaging >=20.0, psutil, python >=3.12,<3.13.0a0, pytorch >=1.10.0, pyyaml, safetensors >=0.3.1, setuptools |
| license | Apache-2.0 |
| license_family | Apache |
| md5 | edbc2f9cb2064296be9c1a8011b683e4 |
| name | huggingface_accelerate |
| platform | linux |
| sha1 | d594d7acf1643906ed70aba6ed80b6ccbacd8cdf |
| sha256 | 0e9103f60b98cdb89ec8bcf77a1ce528578c6cacb4c63753655faf615063bfab |
| size | 577234 |
| subdir | linux-64 |
| timestamp | 1725023433695 |
| version | 0.33.0 |