<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
    <title>pytorch</title>
    <link rel="self" type="application/atom+xml" href="https://links.biapy.com/guest/tags/628/feed"/>
    <updated>2026-08-02T00:43:59+00:00</updated>
    <id>https://links.biapy.com/guest/tags/628/feed</id>
            <entry>
            <id>https://links.biapy.com/links/12768</id>
            <title type="text"><![CDATA[Lance]]></title>
            <link rel="alternate" href="https://lance.org/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/12768"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[The Open Lakehouse Format for Multimodal AI.

Convert from Parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming.

- [Lance @ GitHub](https://github.com/lance-format/lance).

Related contents:

- [S3 is the perfect place to store data, until you try to search it @ Gordon Murray](https://gordonmurray.ie/data/2026/05/02/s3-is-the-perfect-place-to-store-data-until-you-try-to-search-it.html).]]>
            </summary>
            <updated>2026-05-15T13:58:25+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/12529</id>
            <title type="text"><![CDATA[Aidge]]></title>
            <link rel="alternate" href="https://eclipse.dev/aidge/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/12529"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Aidge is an innovative, open-source framework designed to streamline and accelerate the deployment of Deep Neural Networks onto diverse hardware targets. In today’s rapidly evolving AI landscape, moving from a trained model to a high-performance, production-ready application can be a complex and time-consuming process.

- [Aidge @ Eclipe&amp;#039;s GitLab](https://gitlab.eclipse.org/eclipse/aidge/aidge).

Related contents:

- [AIDGE - Du deep learning sur vos microcontrôleurs @ Korben :fr:](https://korben.info/aidge-framework-ia-embarquee-cea-2.html).]]>
            </summary>
            <updated>2026-04-13T09:22:51+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/12048</id>
            <title type="text"><![CDATA[Ray]]></title>
            <link rel="alternate" href="https://www.ray.io/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/12048"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Scale Machine Learning &amp;amp; AI Computing.

 Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads. 

- [Ray @ GitHub](https://github.com/ray-project/ray).]]>
            </summary>
            <updated>2026-03-09T07:18:36+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/11560</id>
            <title type="text"><![CDATA[Pocket TTS]]></title>
            <link rel="alternate" href="https://github.com/kyutai-labs/pocket-tts" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/11560"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[A TTS that fits in your CPU (and pocket).

A lightweight text-to-speech (TTS) application designed to run efficiently on CPUs. Forget about the hassle of using GPUs and web APIs serving TTS models. With Kyutai&amp;#039;s Pocket TTS, generating audio is just a pip install and a function call away.

Related contents:

- [Pocket TTS @ /kyutai](https://kyutai.org/blog/2026-01-13-pocket-tts).
- [Episode 650: This Old Network @ Linux Unplugged](https://linuxunplugged.com/650).]]>
            </summary>
            <updated>2026-01-22T07:14:25+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/10697</id>
            <title type="text"><![CDATA[Lance]]></title>
            <link rel="alternate" href="https://lancedb.github.io/lance/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/10697"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Modern columnar data format for ML and LLMs implemented in Rust. Convert from parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming.. 

Lance is a modern columnar data format optimized for machine learning and AI applications. It efficiently handles diverse multimodal data types while providing high-performance querying and versioning capabilities.

- [Lance @ GitHub](https://github.com/lancedb/lance).

Related contents:

- [Lance takes aim at Parquet in file format joust @ The Register](https://www.theregister.com/2025/10/14/lance_parquet/).]]>
            </summary>
            <updated>2025-10-17T12:01:52+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/10261</id>
            <title type="text"><![CDATA[S3GD]]></title>
            <link rel="alternate" href="https://github.com/WhyPhyLabs/s3gd" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/10261"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[S3GD is a highly optimized, PyTorch-compatible Triton implementation of the Smoothed SignSGD optimizer, meant for reinforcement learning post-training.

Related contents:

- [S3GD Optimizer Algorithm @ WhyPhyLabs](https://whyphy.ai/blog/Aug%202025/08-22-2025-s3gd-blog.md).]]>
            </summary>
            <updated>2025-09-18T06:02:21+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/10094</id>
            <title type="text"><![CDATA[PyTorch]]></title>
            <link rel="alternate" href="https://pytorch.org/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/10094"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Tensors and Dynamic neural networks in Python with strong GPU acceleration.

PyTorch is a Python package that provides two high-level features:

- Tensor computation (like NumPy) with strong GPU acceleration
- Deep neural networks built on a tape-based autograd system

You can reuse your favorite Python packages such as NumPy, SciPy, and Cython to extend PyTorch when needed.

- [PyTorch @ GitHub](https://github.com/pytorch/pytorch).

Related contents:

- [Intro to PyTorch. Easy to follow, visual introduction. @ 0byte.io](https://0byte.io/articles/pytorch_introduction.html).
- [Faire tourner un LLM localement sur votre ordinateur @ Quoi de neuf les devs ? :fr:](https://happytodev.substack.com/p/brent-roose-est-linvite-du-n147-de?open=false#%C2%A7faire-tourner-un-llm-localement-sur-votre-ordinateur).
- [the bug that taught me more about PyTorch than years of using it  @ matmols](https://elanapearl.github.io/blog/2025/the-bug-that-taught-me-pytorch/).
- [The annotated PyTorch training loop @ idlemachines](https://idlemachines.co.uk/essays/pytorch-training-loop).]]>
            </summary>
            <updated>2026-06-26T11:36:58+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/483</id>
            <title type="text"><![CDATA[Monarch]]></title>
            <link rel="alternate" href="https://github.com/pytorch-labs/monarch" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/483"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[PyTorch Single Controller.

Monarch is a distributed execution engine for PyTorch. Our overall goal is to deliver the high-quality user experience that people get from single-GPU PyTorch, but at cluster scale.]]>
            </summary>
            <updated>2025-08-28T17:18:10+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/1783</id>
            <title type="text"><![CDATA[🤗 Transformers]]></title>
            <link rel="alternate" href="https://huggingface.co/docs/transformers/index" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/1783"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[State-of-the-art Machine Learning for PyTorch, TensorFlow, and JAX.

🤗 Transformers provides APIs and tools to easily download and train state-of-the-art pretrained models. Using pretrained models can reduce your compute costs, carbon footprint, and save you the time and resources required to train a model from scratch. 

- [🤗 Transformers @ GitHub](https://github.com/huggingface/transformers).

Related contents:

- [Running inference in web extensions @ dist://ed](https://blog.mozilla.org/en/mozilla/ai/ai-tech/running-inference-in-web-extensions/).]]>
            </summary>
            <updated>2025-08-28T20:53:50+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/1784</id>
            <title type="text"><![CDATA[ONNX Runtime]]></title>
            <link rel="alternate" href="https://onnxruntime.ai/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/1784"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator.

ONNX Runtime inference can enable faster customer experiences and lower costs, supporting models from deep learning frameworks such as PyTorch and TensorFlow/Keras as well as classical machine learning libraries such as scikit-learn, LightGBM, XGBoost, etc. 

- [ONNX Runtime @ GitHub](https://github.com/microsoft/onnxruntime).

Related contents:

- [Running inference in web extensions @ dist://ed](https://blog.mozilla.org/en/mozilla/ai/ai-tech/running-inference-in-web-extensions/).]]>
            </summary>
            <updated>2025-08-28T20:53:50+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/1981</id>
            <title type="text"><![CDATA[JAX]]></title>
            <link rel="alternate" href="https://jax.readthedocs.io/en/latest/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/1981"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[High performance array computing.

 Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more 

- [JAX @ GitHub](https://github.com/jax-ml/jax).

Related contents:

- [The PyTorch developer&amp;#039;s guide to JAX fundamentals @ Google Cloud Blog](https://cloud.google.com/blog/products/ai-machine-learning/guide-to-jax-for-pytorch-developers).]]>
            </summary>
            <updated>2025-08-28T21:26:10+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/2270</id>
            <title type="text"><![CDATA[OpenVINO]]></title>
            <link rel="alternate" href="https://docs.openvino.ai/2024/index.html" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/2270"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference.

OpenVINO is an open-source toolkit for optimizing and deploying deep learning models from cloud to edge. It accelerates deep learning inference across various use cases, such as generative AI, video, audio, and language with models from popular frameworks like PyTorch, TensorFlow, ONNX, and more. Convert and optimize models, and deploy across a mix of Intel® hardware and environments, on-premises and on-device, in the browser or in the cloud.

- [OpenVINO @ GitHub](https://github.com/openvinotoolkit/openvino).
- [OpenVINO @ Hugging Face](https://huggingface.co/OpenVINO).]]>
            </summary>
            <updated>2026-01-23T10:38:15+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/2450</id>
            <title type="text"><![CDATA[picklescan]]></title>
            <link rel="alternate" href="https://github.com/mmaitre314/picklescan" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/2450"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Security scanner detecting Python Pickle files performing suspicious actions]]>
            </summary>
            <updated>2025-08-28T22:45:03+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/2519</id>
            <title type="text"><![CDATA[TorchGeo]]></title>
            <link rel="alternate" href="http://blank" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/2519"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[TorchGeo: datasets, samplers, transforms, and pre-trained models for geospatial data.

TorchGeo is a PyTorch domain library, similar to torchvision, providing datasets, samplers, transforms, and pre-trained models specific to geospatial data.

- [TorchGeo @ GitHub](https://github.com/microsoft/torchgeo).]]>
            </summary>
            <updated>2025-08-28T22:57:01+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/3083</id>
            <title type="text"><![CDATA[ExecuTorch]]></title>
            <link rel="alternate" href="https://pytorch.org/executorch/stable/index.html" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/3083"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[On-device AI across mobile, embedded and edge for PyTorch 

ExecuTorch is an end-to-end solution for enabling on-device inference capabilities across mobile and edge devices including wearables, embedded devices and microcontrollers. It is part of the PyTorch Edge ecosystem and enables efficient deployment of PyTorch models to edge devices.

- [ExecuTorch @ GitHub](https://github.com/pytorch/executorch).]]>
            </summary>
            <updated>2025-08-29T00:30:24+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/3164</id>
            <title type="text"><![CDATA[Meta Lingua]]></title>
            <link rel="alternate" href="https://github.com/facebookresearch/lingua" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/3164"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs. 

Meta Lingua is a minimal and fast LLM training and inference library designed for research. Meta Lingua uses easy-to-modify PyTorch components in order to try new architectures, losses, data, etc. We aim for this code to enable end to end training, inference and evaluation as well as provide tools to better understand speed and stability. While Meta Lingua is currently under development, we provide you with multiple apps to showcase how to use this codebase.]]>
            </summary>
            <updated>2025-08-29T00:44:40+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/3391</id>
            <title type="text"><![CDATA[PerCo]]></title>
            <link rel="alternate" href="https://github.com/Nikolai10/PerCo" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/3391"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[PyTorch implementation of PerCo (Towards Image Compression with Perfect Realism at Ultra-Low Bitrates, ICLR 2024)]]>
            </summary>
            <updated>2025-08-29T01:21:57+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/4371</id>
            <title type="text"><![CDATA[MLX]]></title>
            <link rel="alternate" href="https://github.com/ml-explore/mlx" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/4371"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[An array framework for Apple silicon.

MLX is an array framework for machine learning research on Apple silicon, brought to you by Apple machine learning research.

- [s4e13 - UNDERSCORE_ : $50 000 pour hacker l&amp;#039;IA de Google ! @ Micode&amp;#039;s Twitch :fr:](https://www.twitch.tv/videos/2103081056).]]>
            </summary>
            <updated>2025-08-29T04:06:24+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/4784</id>
            <title type="text"><![CDATA[MLflow]]></title>
            <link rel="alternate" href="https://www.mlflow.org/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/4784"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[A platform for the machine learning lifecycle.

MLflow is a platform to streamline machine learning development, including tracking experiments, packaging code into reproducible runs, and sharing and deploying models. MLflow offers a set of lightweight APIs that can be used with any existing machine learning application or library (TensorFlow, PyTorch, XGBoost, etc), wherever you currently run ML code (e.g. in notebooks, standalone applications or the cloud). MLflow&amp;#039;s current components are:

- [MLflow @ GitHub](https://github.com/mlflow/mlflow/).
- [Setting up a Development Machine with MLFlow and MinIO @ MinIO Blog](https://blog.min.io/setting-up-a-development-machine-with-mlflow-and-minio/).]]>
            </summary>
            <updated>2025-08-29T05:15:00+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/5440</id>
            <title type="text"><![CDATA[ImageBind]]></title>
            <link rel="alternate" href="https://github.com/facebookresearch/ImageBind" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/5440"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[ImageBind One Embedding Space to Bind Them All.

PyTorch implementation and pretrained models for ImageBind. For details, see the paper: ImageBind: One Embedding Space To Bind Them All.

ImageBind learns a joint embedding across six different modalities - images, text, audio, depth, thermal, and IMU data. It enables novel emergent applications ‘out-of-the-box’ including cross-modal retrieval, composing modalities with arithmetic, cross-modal detection and generation.]]>
            </summary>
            <updated>2025-08-29T07:03:57+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/5897</id>
            <title type="text"><![CDATA[fastai]]></title>
            <link rel="alternate" href="https://github.com/fastai/fastai" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/5897"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[fastai is a deep learning library which provides practitioners with high-level components that can quickly and easily provide state-of-the-art results in standard deep learning domains, and provides researchers with low-level components that can be mixed and matched to build new approaches. It aims to do both things without substantial compromises in ease of use, flexibility, or performance. This is possible thanks to a carefully layered architecture, which expresses common underlying patterns of many deep learning and data processing techniques in terms of decoupled abstractions. These abstractions can be expressed concisely and clearly by leveraging the dynamism of the underlying Python language and the flexibility of the PyTorch library. fastai includes:]]>
            </summary>
            <updated>2025-08-29T08:20:36+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/6179</id>
            <title type="text"><![CDATA[YOLOv5]]></title>
            <link rel="alternate" href="https://github.com/ultralytics/yolov5" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/6179"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[YOLOv5 in PyTorch &amp;gt; ONNX &amp;gt; CoreML &amp;gt; TFLite.
YOLOv5 is the world&amp;#039;s most loved vision AI, representing Ultralytics open-source research into future vision AI methods, incorporating lessons learned and best practices evolved over thousands of hours of research and development.]]>
            </summary>
            <updated>2025-08-29T09:06:58+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/6360</id>
            <title type="text"><![CDATA[VToonify]]></title>
            <link rel="alternate" href="https://github.com/williamyang1991/VToonify" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/6360"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[VToonify: Controllable High-Resolution Portrait Video Style Transfer.
[SIGGRAPH Asia 2022] VToonify: Controllable High-Resolution Portrait Video Style Transfer.
This repository provides the official PyTorch implementation for the following paper:]]>
            </summary>
            <updated>2025-08-29T09:37:18+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/6533</id>
            <title type="text"><![CDATA[MLflow]]></title>
            <link rel="alternate" href="https://mlflow.org/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/6533"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[An open source platform for the machine learning lifecycle.

MLflow is a platform to streamline machine learning development, including tracking experiments, packaging code into reproducible runs, and sharing and deploying models. MLflow offers a set of lightweight APIs that can be used with any existing machine learning application or library (TensorFlow, PyTorch, XGBoost, etc), wherever you currently run ML code (e.g. in notebooks, standalone applications or the cloud).

- [MLflow @ GitHub](https://github.com/mlflow/mlflow).]]>
            </summary>
            <updated>2025-08-29T10:05:36+00:00</updated>
        </entry>
    </feed>
