Search Results for "trains"
Sort By:
Showing 88 open source projects for "trains"
View related business solutions-
$300 Free Credits to Build on Google CloudPut your $300 in credit toward real workloads, then keep building with free monthly usage for 20+ products. No commitment and no charge until you upgrade.
-
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
1
JuMP
Modeling language for Mathematical Optimization
...JuMP makes it easy to formulate and solve a range of problem classes, including linear programs, integer programs, conic programs, semidefinite programs, and constrained nonlinear programs. JuMP is used to solve large-scale inventory routing problems at Renault, schedule trains at Thales Inc., plan power grid expansion at PSR, and route school buses. -
2
Microduck RL
RL training environments for Microduck (mjlab)
Microduck RL provides reinforcement-learning environments for training behaviors for Pollen Robotics' small bipedal Microduck robot. It is built on mjlab with MuJoCo Warp and trains PPO policies at a 50 Hz control rate. The repository includes tasks for walking, standing up, sitting, kicking, rolling, skating, crouching, and other physical behaviors. Its sim-to-real setup models actuator physics, backlash, domain randomization, and reward design. Trained policies can be exported to ONNX and deployed through the separate Microduck runtime. ... -
3
Halo
Framework built by White Circle for training large language
Halo is an open-source training framework for large language and multimodal models, designed to scale from a single GPU to multi-node clusters. It supports pretraining, supervised fine-tuning, preference optimization, distillation, and fully asynchronous multi-turn reinforcement learning. The framework trains Hugging Face models directly, so checkpoints remain compatible with standard from_pretrained workflows. Distributed execution supports expert, context, tensor, and expert-tensor parallelism alongside FSDP2. Performance features include FlashAttention, DeepEP, Liger kernels, Grouped GEMM, BF16 optimizers, LoRA, and QLoRA. RL rollouts can run through vLLM or SGLang while overlapping with training. ... -
4
OpenTTD
Open source simulation game based on Transport Tycoon Deluxe
OpenTTD is an open source transport simulation game based on the popular game Transport Tycoon Deluxe by Chris Sawyer, with several additions. OpenTTD is as close to the original as you can get, but has been extended with plenty of great features. These include bigger maps (up to 64 times bigger than the original), stable multiplayer mode that allows for up to 255 players in 15 companies, in-game downloading of heightmaps, AIs, scenarios, and so much more. With its many additions and... -
Veeam Data Platform v13.1 - Get Your Free TrialTry Veeam Data Platform today. Experience the unified platform that's secure by design, portable by default, and proven to recover clean, fast, and anywhere.
-
5
ML Intern
ML engineer that reads papers, trains models, and ships ML models
ML Intern is a repository by Hugging Face that provides educational content and projects aimed at helping learners gain practical experience in machine learning and AI development. It is designed to simulate the experience of working as a machine learning intern, offering tasks and exercises that mirror real-world workflows. The project includes tutorials, datasets, and example implementations that guide users through different aspects of ML development. It emphasizes hands-on learning,... -
6
PilottAI
Python framework for building scalable multi-agent systems
pilottai is an AI-based autonomous drone navigation system utilizing reinforcement learning for real-time decision-making. It is designed for simulating and training drones to fly safely through dynamic environments using AI-based controllers. -
7
Spark NLP
State of the Art Natural Language Processing
...Spark ML provides a set of machine learning applications that can be built using two main components, estimators and transformers. The estimators have a method that secures and trains a piece of data to such an application. The transformer is generally the result of a fitting process and applies changes to the target dataset. These components have been embedded to be applicable to Spark NLP. Pipelines are a mechanism for combining multiple estimators and transformers in a single workflow. They allow multiple chained transformations along a machine-learning task. -
8
DeepSpec
A full-stack codebase for training and evaluating speculative decoding
...It provides the components needed to prepare data, train draft models, and measure acceptance behavior against target models. The workflow starts with data preparation, including prompt download, target answer regeneration, and target cache construction. It then trains a draft model using configuration files for different algorithms and target model setups. The evaluation pipeline measures speculative decoding performance across benchmark tasks such as math, coding, instruction-following, and chat-style datasets. Overall, it is useful for researchers and engineers studying faster language model inference through speculative decoding methods. -
9
SkillOpt
Text-space optimizer that trains reusable natural-language skills
SkillOpt is a Microsoft research project for improving frozen LLM agents by optimizing reusable natural-language skill documents. Instead of changing model weights, it treats a compact skill file as the trainable state of the agent. The system learns from agent rollouts, reflection, bounded edits, and validation gates to produce better instructions over time. Its output is a deployable best_skill.md artifact that can be reused across agent tasks. The project is focused on making agents more... -
Host LLMs in Production With On-Demand GPUsDeploy your model, get an endpoint, pay only for compute time. No GPU provisioning or infrastructure management required.
-
10
MiniMind-V
"Big Model" trains a visual multimodal VLM with 26M parameters
MiniMind-V is an experimental open-source project that aims to train a very small multimodal vision–language model (VLM) from scratch with extremely low compute and cost, making research and experimentation accessible to more people. The repository showcases training workflows and code designed to produce a 26-million parameter model—including both image and text capabilities—using minimal resources in very little time, reflecting a trend toward democratizing AI research. MiniMind-V combines... -
11
CutLER
Code release for Cut and Learn for Unsupervised Object Detection
CutLER is an approach for unsupervised object detection and instance segmentation that trains detectors without human-annotated labels, and the repo also includes VideoCutLER for unsupervised video instance segmentation. The method follows a “Cut-and-LEaRn” recipe: bootstrap object proposals, refine them iteratively, and train detection/segmentation heads to discover objects across diverse datasets. The codebase provides training and inference scripts, model configs, and references to benchmarking results that report large gains over prior unsupervised baselines. ... -
12
Diffusion for World Modeling
Learning agent trained in a diffusion world model
Diffusion for World Modeling is an experimental reinforcement learning system that trains intelligent agents inside a simulated environment generated by a diffusion-based world model. The project introduces the idea of using diffusion models, commonly used for image generation, to simulate the dynamics of an environment and predict future states based on previous observations and actions. Instead of interacting directly with a real environment, the reinforcement learning agent learns within a generative model that produces frames representing the environment. ... -
13
HunyuanDiT
Diffusion Transformer with Fine-Grained Chinese Understanding
HunyuanDiT is a high-capability text-to-image diffusion transformer with bilingual (Chinese/English) understanding and multi-turn dialogue capability. It trains a diffusion model in latent space using a transformer backbone and integrates a Multimodal Large Language Model (MLLM) to refine captions and support conversational image generation. It supports adapters like ControlNet, IP-Adapter, LoRA, and can run under constrained VRAM via distillation versions. LoRA, ControlNet (pose, depth, canny), IP-adapter to extend control over generation. ... -
14
OpenVINO Training Extensions
Trainable models and NN optimization tools
...When ote_cli is installed in the virtual environment, you can use the ote command line interface to perform various actions for templates related to the chosen task type, such as running, training, evaluating, exporting, etc. ote train trains a model (a particular model template) on a dataset and saves results in two files. ote optimize optimizes a pre-trained model using NNCF or POT depending on the model format. NNCF optimization used for trained snapshots in a framework-specific format. POT optimization used for models exported in the OpenVINO IR format. -
15
MedicalGPT
MedicalGPT: Training Your Own Medical GPT Model with ChatGPT Training
MedicalGPT training medical GPT model with ChatGPT training pipeline, implementation of Pretraining, Supervised Finetuning, Reward Modeling and Reinforcement Learning. MedicalGPT trains large medical models, including secondary pre-training, supervised fine-tuning, reward modeling, and reinforcement learning training. -
16
This project is a collection of software and embedded system designs that are developed for controling trains on a modular model railroad as used by FREMO.
-
17
SwitchWithIt Ver 1.7.10.52
Model Railroad Shipment Based Operations Program
Version 1.7.10.52 UPDATED 02/07/26 Model Railroad shipment based operations program. Generates switchlists for freight, unit, mixed, and passenger trains. Trains can begin or end in Towns, Yards (Normal and Staging), or Industries. Also included is an exportable inventory function with image field . The program is portable. It can be loaded on a thumb drive or other memory device and run from any Windows based computer (not phone). Works on Windows XP, Vista, 7, 8, 10, and 11. -
18
XTrkCAD Model RR Track Planner
Model Railway CAD program
XTrackCAD is a open-source, free, CAD program for designing model railroad layouts. You can easily create layout of any scale or size. Libraries for many brands, scales and gauges of track and turnouts/points and model structures are included. Adding new components is easy with the built-in editor. It runs on Windows, Mac and Linux, -
19
simutrans
A free transportation simulator game
Simutrans is a cross-platform simulation game where players try to successfully manage transportation systems between places by land, air, and water for passengers, mail, and goods. Planes, ships, trains, trams, trucks, buses, or monorails are at your disposal, but factories operate based on contracts and passengers can only travel to their set destinations. -
20
dtcltiny - SRCP Model Train Controller
SRCP model train controller
dtcltiny is a small SRCP client to control digital model trains on SRCP systems. dtcltiny needs a SRCP server (e.g. erddcd or srcpd) as hardware link. In cooperation with locking table spdrs60 (http://spdrs60.sourceforge.net/) dtcltiny provides support for train automation. -
21
TCNOpen
Components for IEC61375 standardised communication
...TCN (Train Communication Network) is a series of international standards (IEC61375) developed by Working Group 43 of the IEC (International Electrotechnical Commission), specifying a communication system for the data communication within and between vehicles of a train. It is currently in use on many thousands of trains in the world in order to allow electronic devices to exchange information while operating aboard the same train. TCNOpen follows the Open Source scheme, as the software is jointly developed by participating companies, according to their role, so as to achieve cheaper, quicker and better quality results. -
22
Label Sleuth
Open source no-code system for text annotation and building of text
An open-source no-code system for text annotation and building text classifiers. No AI knowledge needed. From task definition to working model in just a few hours! While domain experts label their data, Label Sleuth automatically trains in the background-appropriate machine learning models. To avoid wasted labeling effort, Label Sleuth employs active learning techniques to guide the user in what they should be labeled next. Domain experts can quickly start labeling their data through an intuitive user interface. Developed by researchers across industry and academia, Label Sleuth incorporates the latest research from human-computer interaction, natural language processing, and artificial intelligence. ... -
23
PKU Beaver
Constrained Value Alignment via Safe Reinforcement Learning
...These annotations include categories such as harmful language, unethical behavior, privacy violations, and other sensitive topics. By incorporating constraint-based optimization methods, Safe-RLHF trains models that balance reward objectives with safety requirements, ensuring that harmful outputs are penalized during training. -
24
PyTorch Implementation of SDE Solvers
Differentiable SDE solvers with GPU support and efficient sensitivity
...The program outputs figures to the path specified by <TRAIN_DIR>. Training should stabilize after 500 iterations with the default hyperparameters. examples/sde_gan.py learns an SDE as a GAN, as in [2], [3]. The example trains an SDE as the generator of a GAN, whilst using a neural CDE [4] as the discriminator. -
25
iJEPA
Official codebase for I-JEPA
...This objective sidesteps generative pixel losses and avoids heavy negative sampling, producing features that transfer strongly with linear probes and minimal fine-tuning. The design scales naturally with Vision Transformer backbones and flexible masking strategies, and it trains stably at large batch sizes. i-JEPA’s predictions are made in embedding space, which is computationally efficient and better aligned with downstream discrimination tasks. The repository provides training recipes, data pipelines, and evaluation code that clarify which masking patterns and architectural choices matter most.
