📄 中文摘要
mlx-snn 是首个基于苹果 MLX 框架原生构建的脉冲神经网络（SNN）库。随着 SNN 研究的快速发展，现有的主要库如 snnTorch、Norse、SpikingJelly 和 Lava 都针对 PyTorch 或自定义后端，导致苹果硅用户缺乏原生选择。mlx-snn 提供六种神经元模型（LIF、IF、Izhikevich、Adaptive LIF、Synaptic、Alpha）、四种代理梯度函数、四种脉冲编码方法（包括 EEG 特定编码器），以及完整的时间反向传播训练管道。该库利用 MLX 的统一内存架构、惰性评估和可组合函数变换（mx.grad、mx.compile），实现了在苹果硅硬件上高效的 SNN 研究。
📄 English Summary
mlx-snn: Spiking Neural Networks on Apple Silicon via MLX
mlx-snn is the first spiking neural network (SNN) library built natively on Apple's MLX framework. As SNN research continues to grow rapidly, existing major libraries such as snnTorch, Norse, SpikingJelly, and Lava primarily target PyTorch or custom backends, leaving users of Apple Silicon without a native option. mlx-snn offers six neuron models (LIF, IF, Izhikevich, Adaptive LIF, Synaptic, Alpha), four surrogate gradient functions, four spike encoding methods (including an EEG-specific encoder), and a complete backpropagation-through-time training pipeline. The library leverages MLX's unified memory architecture, lazy evaluation, and composable function transforms (mx.grad, mx.compile) to enable efficient SNN research on Apple Silicon hardware.