0.21.2

What's Changed

• fix #172 -- missing steel_attenion.metal by @davidkoski in #173

Full Changelog: 0.21.1...0.21.2

0.21.1

What's Changed

• MLXArray (Data, Dtype) constructor , DType size. by @louen in #158

Full Changelog: 0.21.0...0.21.1

0.21.0

What's Changed

• Add a shell script to replace the PrepareMetalHeaders plugin by @louen in #163
• make update method open by @davidkoski in #167
• fix: Fix wrong Wh shape in GRU and LSTM by @minghuaw in #168
• fix update of module tuple (seen for Qwen2VL) by @davidkoski in #164
• fix pow with scalar input by @davidkoski in #170
• fix: Fix wrong layer norm used in TransformerDecoderLayer by @minghuaw in #171
• adopt new mlx-c api, move to mlx v0.21.0 by @davidkoski in #150

New Contributors

• @louen made their first contribution in #163

Full Changelog: 0.18.1...0.21.0

0.18.1

Fix for #139

What's Changed

• Support using class probabilities as targets in func crossEntropy() by @minghuaw in #142
• Change default reduction type of binaryCrossEntropy to mean by @minghuaw in #144
• Add the groups parameter to MLXNN.Conv1d by @lucasnewman in #154
• switch to v0.18.1 mlx tag by @davidkoski in #155

New Contributors

• @lucasnewman made their first contribution in #154

Full Changelog: 0.18.0...0.18.1

v0.18.0 release

• move to v0.18.0 of mlx

• https://github.com/ml-explore/mlx-c v0.0.10
• ml-explore/mlx@v0.16.0...v0.18.0

• turn on additional swift 6 concurrency checks and fix issues

0.16.2

mlx v0.16.0 + #130

0.16.1

mlx v0.16.0 + #127 to help with asyncEval

v0.16.0

Incorporate mlx v0.16.0

ml-explore/mlx@v0.15.2...v0.16.0
https://github.com/ml-explore/mlx-c/pull/31/files

v0.15.2

mlx v0.15.2 and mlx-c aligned with v0.14.0

v0.12.1

mlx-swift with mlx v0.12.1