zabis13.r-universe.dev

[zabis13] sd2R 0.2.1

lbsbmsu@mail.ru (Yuri Baramykov) — Tue, 02 Jun 2026 20:39:41 GMT

Provides Stable Diffusion image generation using the 'ggmlR' library, with no 'Python' or external API dependencies. Supports text-to-image and image-to-image generation for SD 1.x, SD 2.x, 'SDXL', and Flux. A single sd_generate() function handles the entire pipeline, including sampling and high-resolution output. Features multi-GPU support, a 'Shiny' GUI, and runs on CPU or 'Vulkan' GPU across Linux, macOS, and Windows.

[zabis13] cgvR 0.1.2

lbsbmsu@mail.ru (Yuri Baramykov) — Mon, 01 Jun 2026 17:32:30 GMT

Provides interactive 3D visualization for large-scale Cayley graphs. Specifically designed for analyzing state spaces of the 'TopSpin' puzzle. Leverages the 'Datoviz' library and Vulkan-based GPU rendering for smooth real-time exploration of large graphs and complex state transitions. Implements efficient coordinate mapping for high-dimensional permutation groups, allowing users to visualize the connectivity and structural properties of the puzzle's state space. The rendering engine provides high-performance visuals and interactive camera controls, making it suitable for mathematical analysis of group-theoretic puzzles within the R environment.

[zabis13] AV1R 0.1.3

lbsbmsu@mail.ru (Yuri Baramykov) — Mon, 01 Jun 2026 17:31:45 GMT

Converts legacy microscopy video formats (H.264/H.265, AVI/MJPEG, TIFF stacks) to the modern 'AV1' codec with minimal quality loss. Typical use cases include compressing large TIFF stacks from confocal microscopy and time-lapse experiments from hundreds of gigabytes to manageable sizes, re-encoding MP4 files exported from 'CellProfiler', 'ImageJ'/'Fiji', and microscope software with approximately 2x better compression at the same visual quality, and converting legacy AVI (MJPEG) and H.265 recordings to a single patent-free format suited for long-term archival. Automatically selects the best available backend: GPU hardware acceleration via 'Vulkan' 'VK_KHR_VIDEO_ENCODE_AV1' or 'VAAPI' (tested on AMD RDNA4; bundled headers, builds with any 'Vulkan' SDK >= 1.3.275), with automatic fallback to CPU encoding through 'FFmpeg' and 'SVT-AV1'. User controls quality via a single CRF parameter; each backend adapts automatically (CPU and Vulkan use CRF directly, VAAPI targets 55 percent of input bitrate). TIFF stacks use near-lossless CRF 5 by default, with optional proportional scaling via tiff_scale (multiplier or bounding box, aspect ratio always preserved). Small frames are automatically scaled up to meet hardware encoder minimums. Audio tracks are preserved automatically. Provides a simple R API for batch conversion of entire experiment folders.

[zabis13] llamaR 0.2.4

lbsbmsu@mail.ru (Yuri Baramykov) — Mon, 01 Jun 2026 17:31:00 GMT

Provides 'R' bindings to 'llama.cpp' for running Large Language Models ('LLMs') locally with optional 'Vulkan' GPU acceleration via 'ggmlR'. Supports model loading, text generation, 'tokenization', token-to-piece conversion, 'embeddings' (single and batch), encoder-decoder inference, low-level batch management, chat templates, 'LoRA' adapters, explicit backend/device selection, multi-GPU split, and 'NUMA' optimization. Includes a high-level 'ragnar'-compatible embedding provider ('embed_llamar'). Built on top of 'ggmlR' for efficient tensor operations.

[zabis13] ggmlR 0.7.8

lbsbmsu@mail.ru (Yuri Baramykov) — Mon, 01 Jun 2026 16:48:33 GMT

Provides 'R' bindings to the 'GGML' tensor library for machine learning, optimized for 'Vulkan' GPU acceleration with a transparent CPU fallback. The package features a 'Keras'-like sequential API and a 'PyTorch'-style 'autograd' engine for building, training, and deploying neural networks. Key capabilities include high-performance 5D tensor operations, 'f16' precision, and efficient quantization. It supports native 'ONNX' model import (50+ operators) and 'GGUF' weight loading from the 'llama.cpp' and 'Hugging Face' ecosystems. Designed for zero-overhead inference via dedicated weight buffering, it integrates seamlessly as a 'parsnip' engine for 'tidymodels' and provides first-class learners for the 'mlr3' framework. See for more information about the underlying library.

[zabis13] drogonR 0.1.7

lbsbmsu@mail.ru (Yuri Baramykov) — Sun, 24 May 2026 14:29:50 GMT

Provides an 'R' interface to the 'Drogon' high-performance 'C++' 'HTTP' server framework (). Offers a 'plumber'-style application programming interface for building 'REST' services from 'R' with substantially higher throughput.

[cran] drogonR 0.1.6

lbsbmsu@mail.ru (Yuri Baramykov) — Wed, 13 May 2026 10:47:17 GMT

[cran] cgvR 0.1.2

lbsbmsu@mail.ru (Yuri Baramykov) — Mon, 11 May 2026 21:45:38 GMT

[zabis13] cayleyR 0.2.3

lbsbmsu@mail.ru (Yuri Baramykov) — Thu, 09 Apr 2026 20:51:58 GMT

Implements algorithms for analyzing Cayley graphs of permutation groups for the TopSpin puzzle. Provides methods for cycle detection, state space exploration, bidirectional BFS pathfinding, and finding optimal operation sequences in permutation groups generated by shift and reverse operations. Includes C++ implementations of core operations via 'Rcpp' for performance, including a depth-limited BFS path shortener implemented entirely in C++.