把 NumPy 风格的 Python 代码自动微分、向量化并 JIT 编译到 GPU/TPU,写起来像 NumPy,跑起来像高性能计算。
面向人类友好的深度学习 API,用 Python 编写,可跑在 TensorFlow、JAX、PyTorch 上,帮开发者快速搭建和训练神经网络。
把 PyTorch、TensorFlow、JAX、NumPy 等框架的机器学习代码互相转换,省去为不同框架重写模型的麻烦。
每周更新的机器学习 Python 库排行榜,按类别和评分整理优质库,帮你快速找到 AutoML、深度学习、数据可视化等方向的趁手工具。
从公开研究文献出发,用第一性原理复现 Claude Mythos 架构的理论实现,供研究者拆解其注意力机制与模型设计思路。
面向 AI 开发者的实验追踪与模型管理平台,记录训练指标、超参数和数据集版本,支持从实验到生产的全流程协作。
专为 GraalVM 和 HotSpot 打造的 Kubernetes 原生 Java 框架,启动快、内存占用低,适合云原生微服务和容器化部署。
用 PyTorch 逐行复现 60 多篇深度学习经典论文,代码旁配注释讲解,涵盖 Transformer、优化器、GAN、强化学习等,适合边读论文边对照实现学习。
Bug fixes Built C++-generated intrinsic LLVM IR on Windows with clang-cl instead of emitting empty bitcode, enabling vectorized CPU intrinsics on Windows (https
New features Added jax.scipy.special.erfcinv for the inverse complementary error function. Added jax.lax.polynomial for polynomial evaluation. The polynomial primitive is
New features Added jax.numpy.minmax (and jnp.minmax ), which returns both the minimum and maximum of an array, matching NumPy 2.3+ ( 40089). Added jax.lax.log2 and primit
…es Added an error check for trying to deserialize JAX exports that are older than the backwards compatibility window. Without this check the deserialization of expired ar
…g custom derivative rules with the experimental hijax API ( hijax-custom-derivatives ), along with jax.experimental.hijax helpers for deriving VJPHiPrimitive autodiff rul
… variants, support for google's new ML framework [JAX](https://github.com/google/jax), hyperparameter tuning utilities, and more. CleanRL has come a long way making high-