Data Science, Python & R

What Happened Python Polars 2.0 enables out-of-core execution by default, targeting 80% of available RAM and using a default 64 GB disk budget. It adds out-of-core sorting and improves streaming group-by, window, join, and approximate-quantile execution. The release also expands SQL support with grouping sets, ROLLUP, CUBE, GROUPING(), and additional window options [1]. Query-planning changes…

What Happened PyTorch is consolidating image, video and audio decoding and encoding in TorchCodec, rather than maintaining those functions across TorchVision and TorchAudio. The recommended path is to decode media into tensors with TorchCodec, apply TorchVision or TorchAudio transforms, and encode with TorchCodec. TorchVision and TorchAudio are now focused on transforms; their models, datasets and…

What Happened Recent PyTorch ecosystem news spans training, model kernels and serving infrastructure—not broad releases across Python and R data-science libraries. The Linux Foundation introduced a PyTorch Certified Associate pathway with four self-paced modules, hands-on labs and an exam covering data handling, model development and optimization. It estimates 15–17 hours of learning and recommends additional…

What Happened Three developments from the Python data-science and ML infrastructure landscape are relevant to platform and engineering leaders: PyTorch conference sessions positioned Ray as the de facto distributed compute fabric for the full AI lifecycle (data curation → multi-node training → production), including large-scale scheduling demos, Ray Direct Transport, and production case…

Findings [1] 2026-09-24 Accelerate Your AI Journey with new Introduction Track at PyTorch Conference NA 2026 and PyTorch Associate Training As deep learning models move rapidly from research prototypes into core enterprise infrastructure, the demand for practical, end-to-end PyTorch expertise has never been higher. Building robust neural networks requires more than just high-level understanding.…

Findings [1] 2026-09-23 From Research Project to Open Source Ecosystem: Bring Your Academic PyTorch Project to PyTorchCon NA Some of the most interesting work being built with PyTorch starts in universities, research labs, student groups, and academic institutions. A new model architecture. A library created to support a paper. A benchmarking framework. A research…

Findings [1] 2026-09-22 Hardware-Agnostic Models in vLLM TL;DR To achieve state-of-the-art performance at the frontier, vLLM is changing its internal implementation in ways that make it incompatible with fullgraph torch.compile. This may have consequences for users who care about out-of-tree accelerators, older GPUs, or more exotic models.… Figure 1: The current state of model…

Findings [1] 2026-09-21 TinyTorch: Don’t Just Import PyTorch. Build It. A framework you write yourself, tensors through transformers TL;DR Every mature systems project eventually needs a teaching version. TinyTorch is a free, open-source curriculum where you build a working ML framework from scratch, tensors through transformers, in pure Python, using… Figure 3: The gradient…

What Happened Polars published a 2.0.0-rc.2 release with breaking changes, new dtypes and APIs, wide-ranging performance optimizations, and numerous stability fixes. Notable items include Map dtype and related operations, Parquet ENUMs now read as strings, deprecation of cut/qcut, removal of a legacy streaming chunk-size constant, and changed behavior for zero-width DataFrame/LazyFrame inputs. The release also…

What Happened Two items in the current ecosystem shift operational and engineering priorities for AI platforms. JAX released v0.11.2 with a mix of new primitives, performance fixes, build/tooling changes and deployment options: added jax.numpy.minmax, jax.lax.log2 and a high-accuracy one_minus_square primitive; symbolic export helpers (jax.export.symbolic_dim_bounds); frozendict pytrees support aligned with PEP 814; widened random.generalized_normal…

What Happened pandas 3.0.6 is a patch release in the 3.0.x series containing regression and bug fixes; the project recommends that all users of the 3.0.x line upgrade. This release is the first pandas build that supports Python 3.15. Installation instructions and distribution channels are the standard PyPI and conda-forge routes: python -m pip install…

What Happened At PyTorch Conference North America 2026 the ecosystem announced a broad set of compiler, kernel-DSL, distributed-training and low-precision initiatives that together change trade-offs for production ML workloads. Highlights include faster torch.compile tracing via a C++ FakeTensor (~30× speedups on aten.mm), new kernel DSLs and autotuning pipelines (Helion/CuteDSL, FlyDSL, Triton improvements), parametrized dynamic-shape CUDA…