# eVision

> eVision 26.10.0-beta.2.dev documentation


2026, Euresys

## Pages in this subsection

- [Deep Learning Inspection](index.html.md): Overview of eVision's deep learning inspection tools, EasyClassify, EasySegment, and EasyLocate, with links to Deep Learning Studio and hardware support.
- [Deep Learning Studio](deep_learning_studio.html.md): Introduces Deep Learning Studio, the GUI application for creating datasets, training, and evaluating eVision deep learning tools.
- [Workflow](easydeeplearning/index.html.md): Overview of the Deep Learning Studio workflow, from choosing a tool and annotating a dataset to training, inference, and benchmarking.
- [Choose Your Tool](easydeeplearning/choose_your_tool.html.md): Compares EasyClassify, EasySegment, and EasyLocate to help you choose the right deep learning tool for your application.
- [Managing the Dataset and the Annotations](easydeeplearning/managing_the_dataset_and_the_annotations.html.md): Explains how to add images to a dataset and annotate them with labels, segmentation, or objects for eVision deep learning tools.
- [Managing the Dataset Splits](easydeeplearning/managing_the_dataset_splits.html.md): Explains why and how to split a dataset into training, validation, and test sets to reliably evaluate a deep learning tool.
- [Using Data Augmentation](easydeeplearning/using_data_augmentation.html.md): Describes the geometric, color, and noise transformations available for data augmentation when training a deep learning tool.
- [Training a Deep Learning Tool](easydeeplearning/training_a_deep_learning_tool.html.md): Explains how to train a deep learning tool, covering iterations, batch size, deterministic training, and asynchronous execution.
- [Using a Deep Learning Tool](easydeeplearning/using_a_deep_learning_tool.html.md): Explains how to apply a trained deep learning tool to new images in Deep Learning Studio and through the API.
- [Benchmarking a Deep Learning Tool](easydeeplearning/benchmarking_a_deep_learning_tool.html.md): Explains how to benchmark a trained deep learning tool in Deep Learning Studio to measure throughput and latency under different execution settings.
- [Execution Settings and Hardware Support (CPU/GPU)](engines_and_hardware_support_cpu-gpu.html.md): Describes the CPU and GPU execution engines for eVision deep learning tools, including OpenVINO, TensorRT, multi-GPU use, and memory settings.
- [EasyClassify - Multi-label Image Classification](easyclassify/index.html.md): Guide to EasyClassify, the deep learning image classifier, covering capacity, image format, pretrained models, heatmaps, training, and inference.
- [EasySegment - Detecting and Segmenting Defects](easysegment/index.html.md): Overview of EasySegment's supervised and unsupervised defect segmentation modes and when to use each.
- [EasySegment Unsupervised](easysegment/easysegment_unsupervised.html.md): Guide to EasySegment Unsupervised, which trains only on good images to detect unexpected defects, covering properties and metrics.
- [EasySegment Supervised](easysegment/easysegment_supervised.html.md): Guide to EasySegment Supervised, which learns from ground truth segmentation maps to detect and segment defects, covering properties and metrics.
- [EasyLocate - Locating Objects and Defects](easylocate/index.html.md): Guide to EasyLocate, the deep learning tool for locating objects with bounding boxes or interest points, covering properties, training, and validation.

## Optional

- [Top-level llms.txt](../../llms.txt): Complete documentation index.