Metadata-Version: 2.1
Name: dlrover
Version: 0.3.4
Summary: An Automatic Distributed Deep Learning Framework
Home-page: https://github.com/intelligent-machine-learning/dlrover
Author: Ant Group
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: grpcio-tools (>=1.58.0)
Requires-Dist: psutil
Requires-Dist: pynvml
Requires-Dist: urllib3 (<1.27,>=1.21.1)
Requires-Dist: deprecated
Provides-Extra: k8s
Requires-Dist: kubernetes ; extra == 'k8s'
Provides-Extra: ray
Requires-Dist: ray ; extra == 'ray'
Provides-Extra: tensorflow
Requires-Dist: tensorflow ; extra == 'tensorflow'
Provides-Extra: torch
Requires-Dist: torch ; extra == 'torch'

DLRover helps model developers focus on model algorithm itself, without taking care of any engineering stuff, say, hardware acceleration, distribute running, etc. It provides static and dynamic nodes' configuration automatically,, before and during a model training job running on k8s
