Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Classification algorithms now ignore pixels with nodata in the input image. Image classification algorithms are now much faster. Improved show_cmap(), added _show_distance_metrics() and subclass methods.Fixed bug related to initialization value of euclidian distances.

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Image classifiers MinDist, SAM and SID can now label pixels exceeding a given distance metric as unclassified.

Signed-off-by: Daniel Scheffler <danschef@gfz-potsdam.de>

Feature/add random forest regression

See merge request !5

Added random_state to KMeansRSImage.get_random_spectra_from_each_cluster(). Added kwargs to classify_image() to allow passing init args to classifiers. Fixed CFG.spechomo_n_clusters, CFG.spechomo_classif_alg and CFG.spechomo_kNN_n_neighbors ignored in spectral homogenization. Added test_predict_by_machine_learner__RFR_L8_S2().

Fixed issue #95 (DEM_Creator: passing only 2 UTM coordinates causes No-Data triangles at left and right side of DEM).

Revised nodata value handling of RSImage_ClusterPredictor.predict() and added possibility to statically set the nodata value of the predicted image.

Revised image classification algorithms to speed them up in multiprocessing. Added multiprocessing tests for classification algorithms.

Restricted tree depth of Random Forest Regressors to 10 to avoid overfitting and to drastically reduce file size of classifiers.