{"name": "dzetsaka : AI-Powered Remote Sensing Classification", "package_name": "dzetsaka", "version": "3.4.6", "experimental": false, "qgis_min": "3.0.0", "qgis_max": "3.99.0", "downloads": 1001, "uploaded_by": "lennepkade", "upload_datetime": "2019-05-28T03:08:51.020069", "changelog": "3.4.6\n* Minor fixes and remove SLOO training due to error in code\n3.4.5\n* Fix bug when predicting a raster with a previous model and no vector loaded in Qgis.\n3.4.4\n* Remove install of sklearn with python pip (causes bugs).\n* Force n_jobs=1 (1 cpu) while learning.\n3.4.2\n* Fix bug when trying to install sklearn at launch.\n3.4.1\n* Automatically install sklearn (if pip is installed)\n* Precise in the confusion matrix that lines are references and columns prediction.\n3.4\n* Add welcome message if first installation, with, I hope, good tips for new users\n3.3.1\n* Store settings with QSettings (keep settings when updating plugin now!)\n* Correct bug when no nodata value was defined in raster source (default value now : -9999)\n3.3\n* Correct error when chaining with dzetsaka in Processing Toolbox\n3.2\n* Add Domain Adaptation in Processing Toolbox (thanks to POT library)\n* In the settings box, you can choose to have experimental function in the Processing Toolbox\n* Minor fixes\n3.1.1\n* Select providers type (Standard or Experimental), to use latests code in the processing toolbox (but no guarantee at all).\n3.1\n* Replace scipy with numpy when possible\n* Correct bug in predicting an already trained model\n* Specify a way on Windows to install scikit-learn and use SVM/RF/KNN\n3.0.3\n* Add confirmation box if two different projections\n* Correct bug when loading model\n3.0.2\n* Add progress bar for GUI\n3.0.1\n* Minor fixes (with icons)\n3.0.0\n* First version of dzetsaka for Qgis 3.\n* TODO : Progress bar when using UI.\n* TODO : Historical Map algorithms", "external_deps": null, "download_url": "https://plugins.qgis.org/plugins/dzetsaka/version/3.4.6/download/"}