{"name": "QLearn", "package_name": "QLearn_1", "description": "QLearn preforms automatic training of a neural network model on raster data", "about": "QLearn is a plugin that allows you to train a UNet segmentation neural network architecture to segment and classify raster data. It can also use pretrained models to make predictions on raster data\r\n\r\nRequired Dependencies: torch and torchvision\r\nInstall with OSGeo4W Shell (pip3 install torch torchvision)\r\n\r\nQGIS Integration: Train and use machine learning models without leaving QGIS\r\nAutomatic Data Preprocessing: Automatic alignment, rescaling, reclassifying, and normalization of rasters\r\nTesting and Validation: Automatically splits datasets into training and evaluation sets and provides model evaluation metrics\r\nConfidence Filtering: Optional filtering of predictions based on confidence levels\r\nMultiband Support: Works with any number of input bands and output classes", "homepage": "https://qlearn.readthedocs.io/en/latest/index.html", "repository": "https://github.com/0graph/QLearn", "tracker": "https://github.com/0graph/QLearn/issues", "author": "Adam B", "tags": ["raster", "analysis", "classification", "processing", "training", "machine learning", "segmentation", "neural network", "image recognition"], "downloads": 1155, "latest_version": "1.0", "versions": [{"version": "1.0", "experimental": false, "qgis_min": "3.26.2", "qgis_max": "3.99.0", "downloads": 1168, "uploaded_by": "abialecki", "upload_datetime": "2025-04-06T23:56:13.990656"}]}