LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Ecosystems
Launchpads

Search

Python

apex

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

PythonEmerging

Stars

10

Forks

—

Contributors

8

Last push

22mo ago

Recent commits

Latest commits.

Fix illegal memory access with multi_tensor_apply size above INT_MAX
b3bd26aGreg Brockman22mo ago
Allow Configurable Cache Directory (#1821)
59b80eeLei Mao23mo ago
Do not monkey-patch class methods when registering distopt pre-forward hooks (#1820)
f9f19d5Tim Moon23mo ago
release gil (#1816)
23c1f86Masaki Kozuki24mo ago
deprecate uses of torch.cuda.amp (#1813)
f8e60c4Patrick Wang24mo ago

Use torch.testing.all_close instead of get_max_diff in test_lamb.py (#1806)

7b73b12Patrick Wang25mo ago

Fix reduce_blocks_into_lanes race condition (#1798)

a7de60ePatrick Wang26mo ago

NCCL userbuffer for DP RS in DistOpt (#1797)

f3f0492Qiyu Wan26mo ago

Top contributors

Builders behind this project.

definitelynotmcarilli