skills.rest
cwinkelmann
cwinkelmannCOMMUNITY
Explore all skills by this author →
GitHub

local-vision-trainer

Train and fine-tune computer vision models locally on CUDA GPUs.

2|Updated Mar 6, 2026
Search Tags:#transformers#computer-vision#yolo#gpu-training#local-training#mega-detector
One-click install
npx skills add https://github.com/cwinkelmann/usde-innovations-applications-forest-it --skill local-vision-trainer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: local-vision-trainer
Source: https://github.com/cwinkelmann/usde-innovations-applications-forest-it/tree/main/.claude/skills/training
Command: npx skills add https://github.com/cwinkelmann/usde-innovations-applications-forest-it --skill local-vision-trainer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Train and fine-tune computer vision models locally on CUDA GPUs, removing cloud training costs and data transfer bottlenecks.

Core Features & Use Cases

  • Local training of object detection and image classification on a single workstation with a CUDA GPU.
  • MegaDetector fine-tuning support (MD1000-larch, MD1000-sorrel, MDV6-rtdetr-c) via train_megadetector.py.
  • Support for diverse architectures: YOLOv8 detection, RTDETRv2, YOLOS, DETR for detection; ViT, DINOv2, MobileViT, ResNet for classification; Mask2Former and SegFormer for segmentation.
  • Experiment logging to WandB and TensorBoard, with deterministic training options and GPU-friendly configurations.
  • Practical workflow: dataset YAML format (nc, names) and a robust three-phase training strategy to preserve MegaDetector knowledge while fine-tuning.

Quick Start

Run the training with a dataset YAML and pretrained weights to start local vision model fine-tuning on your CUDA GPU.

Frequently Asked Questions about local-vision-trainer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fine-tune MegaDetector locally on a CUDA GPU?▼

Fine-tune MegaDetector locally on a CUDA GPU by running train_megadetector.py with your dataset YAML and pretrained weights. It applies a three-phase training strategy to preserve detection knowledge while adapting to new classes.

What dataset format do I need to train YOLOv8 or RTDETRv2 models locally?▼

Training YOLOv8 or RTDETRv2 models locally requires a dataset YAML file specifying the number of classes (nc) and class names. This format standardizes input for the local training workflow.

Do I need Python packages like torch and ultralytics installed for local vision training?▼

Yes, local vision training requires Python packages such as torch and ultralytics installed on your workstation. These dependencies enable CUDA GPU acceleration for object detection and image classification tasks.

Can I train Transformers-based detection models like YOLOS and DETR on my local hardware?▼

Yes, you can train Transformers-based detection models like YOLOS and DETR on local hardware. The training supports diverse architectures including ViT, DINOv2, and MobileViT for classification tasks as well.

Does local GPU training support experiment logging to TensorBoard and WandB?▼

Yes, local GPU training supports experiment logging to TensorBoard and WandB. It also provides deterministic training options and GPU-friendly configurations to monitor and control your computer vision model training.

What is the best way to perform image segmentation locally without cloud costs?▼

Train image segmentation models like Mask2Former and SegFormer locally on a CUDA GPU to remove cloud costs and data transfer bottlenecks. This approach enables efficient fine-tuning directly on your workstation.

Related Skills

Explore More Data & Analytics Skills→
raoxuan98-hashraoxuan98-hash
2

ctr-training

Configure and run CTR prediction model training with FuxiCTR.

Community
Advanced
ovachieverovachiever
52

weights-and-biases

Automate ML experiment tracking and model registry management via the wandb API.

Community
Advanced
SologaSologa
1

pytorch-lightning

Organize PyTorch code into LightningModules and LightningDataModules for scalable neural network training with PyTorch Lightning's Trainer, including multi-GPU/TPU orchestration, distributed strategies (DDP, FSDP, DeepSpeed), mixed precision, logging (WandB, TensorBoard) and callbacks.

Community
Advanced
svngokusvngoku
8

unsloth-hf-jobs

Fine-tune LLMs and VLMs on HF Jobs with Unsloth for GPU training.

Community
Advanced