AIPHAProcessing

class AIPHAProcessing.AIPHAProcessing(client, processing_folder='processing', worker_instance_type='P2', manager_instance_type='small')

Bases: object

AIPHAProcessing class for data processing with AIPHA API.

create_operator_from_path(path, extension='', is_folder_level='__auto__')

Create an “operator” pointing to a path.

Parameters:

  • path – path to file or folder

  • extension – extension of file

  • is_folder_level – if True, the operator will be executed massive parallel for each file in a folder level, if False, the operator will be executed on file level, if ‘__auto__’, the operator will be executed on folder level if the path is a folder, otherwise on file level

execute()

Execute the call stack.

class fvo(outer_class)

Bases: object

Namespace for fvo functions.

align_top(input_path='__auto__', target_path='__auto__', output_path='__auto__', extension_input_file='.laz', extension_target_file='.laz', extension_output_file='.laz', folder_parallel_processing='__auto__')

Zero centering of XYZ points in a LAZ file

Parameters:

  • input_path – input LAZ folder folder

  • target_path – input LAZ folder folder

  • output_path – output LAZ folder folder

connect_neighbouring_vertices_unassigned(input_path='__auto__', input_vertices='vertices.laz', output_path='__auto__', max_line_distance=0.5, max_line_distance_corner=0.23, min_samples=3, extension_input_file='.laz', extension_output_file='.dxf', folder_parallel_processing='__auto__')

Connect neighbouring vertices in a point cloud

Parameters:

  • input_path – Input folder with all 3D points

  • input_vertices – Input vertices

  • output_path – Output model

  • max_line_distance – Maximum distance between line and points to be considered as inlier

  • max_line_distance_corner – Maximum distance between vertices and points to be considered as inlier

  • min_samples – Minimum number of points to fit a line

estimate_vobject_coordinates(path_source_in='__auto__', path_trafo_out='__auto__', path_source_out='__auto__', extension_file_source_in='.laz', extension_file_trafo_out='.txt', extension_file_source_out='.laz', folder_parallel_processing='__auto__')

estimate vobject coordinates

Parameters:

  • path_source_in – input folder data

  • path_trafo_out – output trafo

  • path_source_out – output folder data

evaluate_model(data_is='is.laz', model_is='is.dxf', model_target='__auto__', vertex_target_distance=0.5, extension_model_target='.dxf', folder_parallel_processing='__auto__')

Compare two 3D models

Parameters:

  • data_is – input file

  • model_is – input file

  • model_target – output folder

  • vertex_target_distance – Distance threshold for corner point suppression.

export_vertices(in_path='__auto__', out_path='__auto__', extension_in_file='.laz', extension_out_file='.dxf', folder_parallel_processing='__auto__')

Export a 3D model from a point cloud

Parameters:

  • in_path – input folder with edges encoded in the intensity field.

  • out_path – output folder

filter_valid_vertices(input_path='__auto__', input_path2='__auto__', input_path_features='__auto__', output_path='__auto__', output_path_features='__auto__', min_distance=0.0, max_distance=100, extension_input_file='.laz', extension_input_file2='.laz', extension_input_file_features='.npy', extension_output_file='.laz', extension_output_file_features='.npy', folder_parallel_processing='__auto__')

Filter valid vertices from a DXF file.

Parameters:

  • input_path – Input laz or txt folder to filter.

  • input_path2 – Input laz or txt folder as reference.

  • input_path_features – Input features folder.

  • output_path – Output laz or txt folder.

  • output_path_features – Output features folder.

  • min_distance – Minimum distance.

  • max_distance – Maximum distance.

import_vertices(in_path='__auto__', layer='-1', out_path='__auto__', extension_in_file='.dxf', extension_out_file='.laz', folder_parallel_processing='__auto__')

Extract visible face3d vertices from a DXF file.

Parameters:

  • in_path – Input DXF folder folder

  • layer – Layer names as comma-separated list

  • out_path – Output folder folder

likelihood(input_path='__auto__', points_path='__auto__', output_path='__auto__', max_distance=0.5, missing_distance=1.5, missing_knn=2, extension_input_file='.laz', extension_points_file='.laz', extension_output_file='.laz', folder_parallel_processing='__auto__')

Compute class conditional probability distribution

Parameters:

  • input_path – input LAZ folder folder

  • points_path – input LAZ folder folder

  • output_path – output LAZ folder folder

  • max_distance – probability max distance

  • missing_distance – interpolate missing points distance

  • missing_knn – interpolate number missing neighbours

model_swap_axis(input_path='__auto__', output_path='__auto__', extension_input_file='.dxf', extension_output_file='.dxf', folder_parallel_processing='__auto__')

model swap axis

Parameters:

  • input_path – input dxf folder folder

  • output_path – output dxf folder folder

optimize_model_graph(in_dxf_path='__auto__', in_point_cloud_path='__auto__', out_dxf_path='__auto__', max_distance=0.35, num_iterations=4, extension_in_dxf_file='.dxf', extension_in_point_cloud_file='.laz', extension_out_dxf_file='.dxf', folder_parallel_processing='__auto__')

Optimize the graph of a 3D model.

Parameters:

  • in_dxf_path – input folder

  • in_point_cloud_path – input folder

  • out_dxf_path – output folder

  • max_distance – Maximum distance between a point and a graph node for the point to be considered a candidate for merging with the graph node.

  • num_iterations – Number of iterations to run the optimization.

simplify_model(in_path='__auto__', out_path='__auto__', layers='1', distance=0.25, extension_in_file='.dxf', extension_out_file='.dxf', folder_parallel_processing='__auto__')

Simplify a 3D model

Parameters:

  • in_path – input folder

  • out_path – output folder

  • layers – layers to be processed

  • distance – Distance threshold for corner point suppression.

zero_centering(input_path='__auto__', output_path='__auto__', extension_input_file='.laz', extension_output_file='.laz', folder_parallel_processing='__auto__')

Zero centering of XYZ points in a LAZ file

Parameters:

  • input_path – input LAZ folder folder

  • output_path – output LAZ folder folder

class image(outer_class)

Bases: object

Namespace for image functions.

assign_georeference(georeferenced_path='__auto__', unreferenced_path='__auto__', output_path='__auto__', extension_georeferenced_file='.tif', extension_unreferenced_file='.tif', extension_output_file='.tif', folder_parallel_processing='__auto__')

Assign georeference from a georeferenced image to an unreferenced image.

Parameters:

  • georeferenced_path – Georeferenced image folder

  • unreferenced_path – Unreferenced image folder

  • output_path – Output georeferenced image folder

canny_edge_detection(input_path='__auto__', output_path='__auto__', sigma=1.0, low_threshold=0.1, high_threshold=0.2, values_subset='', extension_input_file='.tif', extension_output_file='.tif', folder_parallel_processing='__auto__')

Perform Canny edge detection on a georeferenced image and save the detected edges as a raster.

Parameters:

  • input_path – Input image folder

  • output_path – Output edge raster folder

  • sigma – Standard deviation of the Gaussian filter

  • low_threshold – Low threshold for hysteresis

  • high_threshold – High threshold for hysteresis

  • values_subset – Subset of values to extract contours from, all values by default

extract_contours(input_path='__auto__', value=0.5, output_path='__auto__', values_subset='', extension_input_file='.tif', extension_output_file='.shp', folder_parallel_processing='__auto__')

Extract contours from a georeferenced image and save them as a shapefile.

Parameters:

  • input_path – Input georeferenced image folder

  • value – Contour value

  • output_path – Output shapefolder

  • values_subset – Subset of contour values, default is all values

image_metadata(input_path='__auto__', output_path='__auto__', extension_input_file='.tif', extension_output_file='.json', folder_parallel_processing='__auto__')

Obtain metadata of a georeferenced image and save it as a JSON file.

Parameters:

  • input_path – Input georeferenced image folder

  • output_path – Output JSON folder

image_to_matrix(input_path='__auto__', output_path='__auto__', extension_input_file='.tif', extension_output_file='.npy', folder_parallel_processing='__auto__')

Convert an image to a matrix.

Parameters:

  • input_path – Input image folder

  • output_path – Output matrix folder (either .npy or .txt)

matrix_to_image(input_path='__auto__', output_path='__auto__', data_type='uint8', extension_input_file='.npy', extension_output_file='.tif', folder_parallel_processing='__auto__')

Convert a matrix to an image.

Parameters:

  • input_path – Input matrix folder (either .npy or .txt)

  • output_path – Output image folder

  • data_type – Data type of the output image

polygon_to_image(geotiff_path='__auto__', pickle_path='__auto__', output_path='__auto__', extension_geotiff_file='.tif', extension_pickle_file='.pickle', extension_output_file='.tif', folder_parallel_processing='__auto__')

Generate an image of a multipolygon filled inside.

Parameters:

  • geotiff_path – Geotiff folder with size and resolution information

  • pickle_path – Pickle folder containing the Shapely polygon

  • output_path – Output georeferenced TIFF image

resize_image(input_path='__auto__', output_path='__auto__', new_grid_size=1.0, compression='None', extension_input_file='.tif', extension_output_file='.tif', folder_parallel_processing='__auto__')

resize image

Parameters:

  • input_path – Input georeferenced image folder

  • output_path – Output georeferenced image folder

  • new_grid_size – New grid size in meters

  • compression – Compression method (e.g., deflate, lzw)

retile_images(path_reference='.', path_to_retile='.', output_path='out1', extension_ref='.tif', extension_ret='.tif', folder_parallel_processing='__auto__')

retile images

Parameters:

  • path_reference – Reference folder with image dimensions and geolocations that should be used for retiling

  • path_to_retile – Folder with images that should be retiled to match reference

  • output_path – Folder with retiled images

  • extension_ref – file extension

  • extension_ret – file extension

class ml3d(outer_class)

Bases: object

Namespace for ml3d functions.

evaluate_semantic_segmentation(prediction_path='__auto__', ground_truth_path='__auto__', class_names='1,2,3,4', invalid_label=0, extension_prediction_path='.labels', extension_ground_truth_path='.labels', folder_parallel_processing='__auto__')

Evaluate semantic segmentation

Parameters:

  • prediction_path – Path to prediction folder or folder

  • ground_truth_path – Path to ground truth folder or folder

  • class_names – class names

  • invalid_label – Invalid label

knn_classification(in_path_to_points='__auto__', in_path_from_points='__auto__', out_path_labels='__auto__', out_path_probs='__auto__', k=3, max_distance=1.0, to_points_names='X,Y,Z', from_point_names='X,Y,Z', from_class_name='classification', extension_in_path_to_points='.laz', extension_in_path_from_points='.laz', extension_out_path_labels='.labels', extension_out_path_probs='.npy', folder_parallel_processing='__auto__')

knn classification

Parameters:

  • in_path_to_points – input point cloud to be labeled

  • in_path_from_points – input reference point cloud

  • out_path_labels – out class labels

  • out_path_probs – out class probabilities

  • k – number of neighbors

  • max_distance – maximum distance

  • to_points_names – names of points to be labeled

  • from_point_names – names of reference points

  • from_class_name – name of reference classification

knn_probabilistic_classification(in_path_to_points='__auto__', in_path_from_points='__auto__', in_path_from_probs='__auto__', out_path_labels='__auto__', out_path_probs='__auto__', k=3, max_distance=1.0, to_points_names='X,Y,Z', from_point_names='X,Y,Z', from_class_name='classification', extension_in_path_to_points='.laz', extension_in_path_from_points='.laz', extension_in_path_from_probs='.npy', extension_out_path_labels='.labels', extension_out_path_probs='.npy', folder_parallel_processing='__auto__')

knn probabilistic classification

Parameters:

  • in_path_to_points – input point cloud to be labeled

  • in_path_from_points – input reference point cloud

  • in_path_from_probs – input reference class probabilities

  • out_path_labels – out class labels

  • out_path_probs – out class probabilities

  • k – number of neighbors

  • max_distance – maximum distance

  • to_points_names – names of points to be labeled

  • from_point_names – names of reference points

  • from_class_name – name of reference classification

regression_inference(in_path_points='data_train/points', out_path_predictions='data_train/predictions', in_model_path='parameters_model', folder_parallel_processing='__auto__')

regression inference

Parameters:

  • in_path_points – input directory with training data

  • out_path_predictions – output directory with predictions

  • in_model_path – model path

regression_training(in_path_points='data_train/points', in_path_gt='data_train/gt', out_model_path='parameters_model', voxel_size=0.02, zero_centering='True', point_names='X,Y,Z', feature_names='ones', learning_rate=3e-06, weight_decay=1e-09, num_epochs=500, batch_size=2, save_after_epochs=100, network_type='FCNN128', criterion_type='L1Mean', num_classes=1, folder_parallel_processing='__auto__')

regression training

Parameters:

  • in_path_points – input directory with training data

  • in_path_gt – input directory with training data

  • out_model_path – model path

  • voxel_size – voxel size

  • zero_centering – zero centering

  • point_names – point names

  • feature_names – feature names

  • learning_rate – learning rate

  • weight_decay – regularization decay

  • num_epochs – number of epochs

  • batch_size – batch size for training

  • save_after_epochs – save after epochs

  • network_type – model type of backbone network

  • criterion_type – model type of criterion

  • num_classes – number of classes

semantic_inference_pt_v2m2(data_in_path='__auto__', in_model_parameters_path='trained_model/model_ptv2m2', out_label_path='__auto__', out_probability_path='__auto__', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='X,Y,Z', label_name='classification', resolution=0.05, number_of_votes=5, encoding_channels='60,120,240,256', encoding_groups='12,24,48,64', decoding_channels='30,60,120,144', decoding_groups='6,12,24,48', extension_data_in_path='.laz', extension_out_label_path='.labels', extension_out_probability_path='.npy', folder_parallel_processing='__auto__')

PT v2m2 Inference

Parameters:

  • data_in_path – folder that contains the test data

  • in_model_parameters_path – path to model

  • out_label_path – folder that contains the results

  • out_probability_path – folder that contains the results

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • number_of_votes – number of votes

  • encoding_channels – comma separated list of layer sizes

  • encoding_groups – comma separated list of layer sizes

  • decoding_channels – comma separated list of layer sizes

  • decoding_groups – comma separated list of layer sizes

semantic_inference_pt_v3m1(data_in_path='__auto__', in_model_parameters_path='trained_model/model_ptv2m2', out_label_path='__auto__', out_probability_path='__auto__', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='X,Y,Z', label_name='classification', resolution=0.05, number_of_votes=5, extension_data_in_path='.laz', extension_out_label_path='.labels', extension_out_probability_path='.npy', folder_parallel_processing='__auto__')

PT v3m1 Inference

Parameters:

  • data_in_path – folder that contains the test data

  • in_model_parameters_path – path to model

  • out_label_path – folder that contains the results

  • out_probability_path – folder that contains the results

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • number_of_votes – number of votes

semantic_inference_rfcr(data_in_path='__auto__', results_labels_path='__auto__', results_probabilities_path='__auto__', in_model_parameters_path='results/Log_2022-11-10_11-42-05', number_of_votes=5, feature_names='red,green,blue', point_names='x,y,z', extension_data_in_path='.laz', extension_results_labels_path='.labels', extension_results_probabilities_path='.npy', folder_parallel_processing='__auto__')

semantic inference rfcr

Parameters:

  • data_in_path – folder to data

  • results_labels_path – folder to labels

  • results_probabilities_path – folder to probabilities

  • in_model_parameters_path – path to model

  • number_of_votes – number of votes to vote for a class

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

semantic_inference_scf(data_in_path='__auto__', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='x,y,z', label_name='classification', feature_dimensions='12,48,96,192,384', batch_size=2, results_labels_path='__auto__', in_model_parameters_path='results/Log_2022-11-10_11-42-05', results_probabilities_path='__auto__', number_of_votes=5, extension_data_in_path='.laz', extension_results_labels_path='.labels', extension_results_probabilities_path='.npy', folder_parallel_processing='__auto__')

semantic inference scf

Parameters:

  • data_in_path – folder to folder that contains the training data

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • feature_dimensions – feature dimensions

  • batch_size – batch_size

  • results_labels_path – folder to labels

  • in_model_parameters_path – path to model

  • results_probabilities_path – folder to probabilities

  • number_of_votes – number of votes to vote for a class

semantic_inference_spunet(data_in_path='__auto__', in_model_parameters_path='trained_model/model_1', out_label_path='__auto__', out_probability_path='__auto__', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='X,Y,Z', label_name='classification', resolution=0.05, channels='32,64,128,256,256,128,96,96', layers='2,3,4,6,2,2,2,2', number_of_votes=5, extension_data_in_path='.laz', extension_out_label_path='.labels', extension_out_probability_path='.npy', folder_parallel_processing='__auto__')

Spunet Inference

Parameters:

  • data_in_path – folder that contains the test data

  • in_model_parameters_path – path to model

  • out_label_path – folder that contains the results

  • out_probability_path – folder that contains the results

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • channels – comma separated list of channels

  • layers – comma separated list of layers

  • number_of_votes – number of votes

semantic_training_pt_v2m2(data_in_path='__auto__', out_model_parameters_path='trained_model/model_ptv2m2', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='X,Y,Z', label_name='classification', resolution=0.05, max_epochs=500, learning_rate=0.01, batch_size=10, final_div_factor=100, div_factor=10, weight_decay=0.005, encoding_channels='60,120,240,256', encoding_groups='12,24,48,64', decoding_channels='30,60,120,144', decoding_groups='6,12,24,48', extension_data_in_path='', folder_parallel_processing='__auto__')

Pt v2m2 Training

Parameters:

  • data_in_path – folder to folder that contains the training data

  • out_model_parameters_path – path to model

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • max_epochs – maximum number of epochs

  • learning_rate – learning rate

  • batch_size – batch size

  • final_div_factor – final div factor for learning rate

  • div_factor – div factor for learning rate

  • weight_decay – weight decay

  • encoding_channels – comma separated list of layer sizes

  • encoding_groups – comma separated list of layer sizes

  • decoding_channels – comma separated list of layer sizes

  • decoding_groups – comma separated list of layer sizes

semantic_training_pt_v3m1(data_in_path='__auto__', out_model_parameters_path='trained_model/model_ptv2m2', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='X,Y,Z', label_name='classification', resolution=0.05, max_epochs=500, learning_rate=0.01, batch_size=10, final_div_factor=100, div_factor=10, weight_decay=0.005, extension_data_in_path='', folder_parallel_processing='__auto__')

Pt v3m1 Training

Parameters:

  • data_in_path – folder to folder that contains the training data

  • out_model_parameters_path – path to model

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • max_epochs – maximum number of epochs

  • learning_rate – learning rate

  • batch_size – batch size

  • final_div_factor – final div factor for learning rate

  • div_factor – div factor for learning rate

  • weight_decay – weight decay

semantic_training_rfcr(data_in_path='__auto__', out_model_parameters_path='trained_model/model_1', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='x,y,z', label_name='classification', resolution=0.06, max_epochs=500, learning_rate=0.01, batch_size=10, learning_rate_decay=0.1, learning_momentum=0.98, learning_gradient_clip_norm=100, first_features_dim=128, extension_data_in_path='', folder_parallel_processing='__auto__')

semantic training rfcr

Parameters:

  • data_in_path – folder to folder that contains the training data

  • out_model_parameters_path – path to model

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • max_epochs – maximum number of epochs

  • learning_rate – learning rate

  • batch_size – batch size

  • learning_rate_decay – learning rate decay

  • learning_momentum – learning momentum

  • learning_gradient_clip_norm – learning gradient clip threshold

  • first_features_dim – first features dimension

semantic_training_scf(data_in_path='__auto__', out_model_parameters_path='trained_model/model_1', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='x,y,z', label_name='classification', max_epochs=500, learning_rate=0.01, learning_rate_decay=0.95, feature_dimensions='16,64,128,256,512', batch_size=2, extension_data_in_path='', folder_parallel_processing='__auto__')

semantic training scf

Parameters:

  • data_in_path – folder to folder that contains the training data

  • out_model_parameters_path – path to model

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • max_epochs – maximum number of epochs

  • learning_rate – learning rate

  • learning_rate_decay – learning rate decay

  • feature_dimensions – feature dimensions

  • batch_size – batch_size

semantic_training_spunet(data_in_path='__auto__', out_model_parameters_path='trained_model/model_1', class_names='1,2,3,4,5,6,7,8', feature_names='red,green,blue', point_names='X,Y,Z', label_name='classification', resolution=0.05, max_epochs=500, learning_rate=0.01, batch_size=10, final_div_factor=100, div_factor=10, weight_decay=0.005, channels='32,64,128,256,256,128,96,96', layers='2,3,4,6,2,2,2,2', extension_data_in_path='', folder_parallel_processing='__auto__')

Spunet Training

Parameters:

  • data_in_path – folder to folder that contains the training data

  • out_model_parameters_path – path to model

  • class_names – comma separated list of class names. Class 0 is always given and is used to denote unlabeled points.

  • feature_names – comma separated list of features that are provided

  • point_names – comma separated list of point identifiers in (las/laz)

  • label_name – label name for (las/laz)

  • resolution – resolution of the subsampled point cloud

  • max_epochs – maximum number of epochs

  • learning_rate – learning rate

  • batch_size – batch size

  • final_div_factor – final div factor for learning rate

  • div_factor – div factor for learning rate

  • weight_decay – weight decay

  • channels – comma separated list of channels

  • layers – comma separated list of layers

universal_inference(in_paths='__auto__', out_paths='__auto__', in_model_path='parameters_model', extension_in_files='', extension_out_files='', folder_parallel_processing='__auto__')

universal inference

Parameters:

  • in_paths – input folders with training data

  • out_paths – output folders with training data

  • in_model_path – model path

universal_training(in_path='data_train', out_model_path='parameters_model_test', voxel_size=0.02, zero_centering='True', point_names='X,Y,Z', feature_names='', label_names='classification', num_classes=1, label_scales='0.01', learning_rate=3e-06, learning_decay=0.9999, num_epochs=200000, regularization_decay=1e-09, batch_size=2, save_after_epochs=100, backbone_type='MinkUNet14A', head_type='HeadPointwise', criterion_type='L1Sum', probabilistic='True', hidden_layers=8, store_in_memory='True', folder_parallel_processing='__auto__')

universal training

Parameters:

  • in_path – input directory with training data

  • out_model_path – model path

  • voxel_size – voxel size

  • zero_centering – zero centering

  • point_names – point names

  • feature_names – feature names

  • label_names – label names

  • num_classes – number of classes

  • label_scales – label scales

  • learning_rate – learning rate

  • learning_decay – learning rate decay

  • num_epochs – number of epochs

  • regularization_decay – regularization decay

  • batch_size – batch size for training

  • save_after_epochs – save after epochs

  • backbone_type – model type of backbone network

  • head_type – model type of head network

  • criterion_type – model type of criterion

  • probabilistic – estimate probabilities: labels in [0,1]

  • hidden_layers – number of hidden layers

  • store_in_memory – store training data in memory

vertices_estimation_inference(in_paths='__auto__', out_paths='__auto__', in_model_path='parameters_model_test', batch_size=1, extension_in_files='', extension_out_files='', folder_parallel_processing='__auto__')

vertices estimation inference

Parameters:

  • in_paths – input folders or directory with training data

  • out_paths – output folders containing the vertices

  • in_model_path – model path

  • batch_size – batch size for training

vertices_estimation_training(in_path='data_train', in_vertices_path='data_train_vertices', out_model_path='parameters_model_test', voxel_size=0.02, zero_centering='True', point_names='X,Y,Z', feature_names='', label_names='classification', num_classes=1, label_scales='0.01', learning_rate=1e-05, learning_decay=0.999, num_epochs=2000000, regularization_decay=1e-09, batch_size=5, save_after_epochs=100, backbone_type='MinkUNet14A', head_type_prob='HeadPointwise', criterion_type_prob='BCEMean', hidden_layers=8, max_interpolation_distance=0.75, dist_threshold=0.35, score_threshold=0.4, point_estimation_layers=3, point_estimation_channels=8, criterion_type_point='L1Mean', weight_pred=1.0, weight_prob=2.0, weight_reconstruction=4.0, probabilistic='True', folder_parallel_processing='__auto__')

vertices estimation training

Parameters:

  • in_path – input directory with training data

  • in_vertices_path – input directory with corresponding vertex data

  • out_model_path – model path

  • voxel_size – voxel size

  • zero_centering – zero centering

  • point_names – point names

  • feature_names – feature names

  • label_names – label names

  • num_classes – number of classes

  • label_scales – label scales

  • learning_rate – learning rate

  • learning_decay – learning rate decay

  • num_epochs – number of epochs

  • regularization_decay – regularization decay

  • batch_size – batch size for training

  • save_after_epochs – save after epochs

  • backbone_type – model type of backbone network

  • head_type_prob – model type of head network

  • criterion_type_prob – model type of criterion

  • hidden_layers – number of hidden layers

  • max_interpolation_distance – maximum distance to interpolate occluded points

  • dist_threshold – distance threshold for non-maximum suppression

  • score_threshold – score threshold for non-maximum suppression

  • point_estimation_layers – number of hidden layers for point estimation

  • point_estimation_channels – number of channels for point estimation

  • criterion_type_point – model type of criterion for point estimation

  • weight_pred – weight for point estimation

  • weight_prob – weight for probability estimation

  • weight_reconstruction – weight for reconstruction estimation

  • probabilistic – probabilistic

wireframe_estimation_inference(in_paths='__auto__', out_result_paths='__auto__', in_model_path='parameters_wireframe', knn_line=15, mode_wireframe_estimation='knn unassigned', num_votes=10, rotation_axis='z', extension_in_files='', extension_out_result_files='', folder_parallel_processing='__auto__')

[hidden] Wireframe estimation inference

Parameters:

  • in_paths – input folders or directory with training data

  • out_result_paths – output folders containing the wireframes

  • in_model_path – model path

  • knn_line – knn line

  • mode_wireframe_estimation – mode for wireframe estimation

  • num_votes – number of votes for wireframe estimation

  • rotation_axis – rotation axis

wireframe_estimation_training(in_path='data_train', in_wireframe_path='data_train_wireframe', out_model_path='parameters_wireframe_14A_bce_interpolation', voxel_size=0.02, zero_centering='True', point_names='X,Y,Z', feature_names='', label_names='classification', num_classes=1, label_scales='0.01', learning_rate=5e-06, learning_decay=0.999, num_epochs=2000000, regularization_decay=1e-10, batch_size=5, save_after_epochs=1, backbone_type='MinkUNet14A', head_type_prob='HeadPointwise', criterion_type_prob='BCEMean', hidden_layers=8, max_interpolation_distance=0.75, dist_threshold=0.35, score_threshold=0.5, point_estimation_layers=3, point_estimation_channels=32, criterion_type_point='L1Mean', wireframe_criterion_type='BCEMean', wireframe_estimation_layers=3, wireframe_estimation_channels=32, weight_pred=2, weight_prob=6.5, weight_reconstruction=4.5, weight_wireframe=9, knn_line=10, distance_line=0.3, probabilistic='True', store_in_memory='True', mode_wireframe_estimation='knn', maximum_wireframe_samples=2500, wireframe_subsampling=5, wireframe_extrapolation_sampling=2, only_train_wireframe='False', folder_parallel_processing='__auto__')

[hidden] wireframe estimation training

Parameters:

  • in_path – input directory with training data

  • in_wireframe_path – input directory with corresponding wireframe data

  • out_model_path – model path

  • voxel_size – voxel size

  • zero_centering – zero centering

  • point_names – point names

  • feature_names – feature names

  • label_names – label names

  • num_classes – number of classes

  • label_scales – label scales

  • learning_rate – learning rate

  • learning_decay – learning rate decay

  • num_epochs – number of epochs

  • regularization_decay – regularization decay

  • batch_size – batch size for training

  • save_after_epochs – save after epochs

  • backbone_type – model type of backbone network

  • head_type_prob – model type of head network

  • criterion_type_prob – model type of criterion

  • hidden_layers – number of hidden layers

  • max_interpolation_distance – maximum distance to interpolate occluded points

  • dist_threshold – distance threshold for non-maxima suppression

  • score_threshold – score threshold for non-maxima suppression

  • point_estimation_layers – number of hidden layers for point estimation

  • point_estimation_channels – number of channels for point estimation

  • criterion_type_point – model type of criterion for point estimation

  • wireframe_criterion_type – model type of criterion for wireframe estimation

  • wireframe_estimation_layers – number of hidden layers for wireframe estimation

  • wireframe_estimation_channels – number of channels for wireframe estimation

  • weight_pred – weight for point estimation

  • weight_prob – weight for probability estimation

  • weight_reconstruction – weight for reconstruction

  • weight_wireframe – weight for wireframe estimation

  • knn_line – number of nearest neighbours for line estimation

  • distance_line – distance threshold for line estimation

  • probabilistic – probabilistic

  • store_in_memory – store in memory

  • mode_wireframe_estimation – wireframe mode

  • maximum_wireframe_samples – maximum number of wireframe samples

  • wireframe_subsampling – wireframe subsampling factor

  • wireframe_extrapolation_sampling – wireframe extrapolation sampling factor

  • only_train_wireframe – only train wireframe

class ops3d(outer_class)

Bases: object

Namespace for ops3d functions.

align_points(path_source_in='segmented_object', path_transformation_in='transformations', path_source_out='aligned_points', folder_parallel_processing='__auto__')

align points

Parameters:

  • path_source_in – input folder data

  • path_transformation_in – input folder transformation

  • path_source_out – output folder

assign_point_labels(path_source_in='__auto__', path_labels_in='__auto__', path_source_out='__auto__', dtype='classification', all_type='', extension_file_source_in='.laz', extension_file_labels_in='.npy', extension_file_source_out='.laz', folder_parallel_processing='__auto__')

assign point labels

Parameters:

  • path_source_in – input folder data

  • path_labels_in – input folder labels

  • path_source_out – output folder

  • dtype – value

  • all_type – values to load

crop_and_merge_polygons(point_cloud_paths='__auto__', polygon_path='__auto__', output_path='__auto__', extension_point_cloud_files='.laz', extension_polygon_file='.pickle', extension_output_file='.laz', folder_parallel_processing='__auto__')

crop and merge polygons

Parameters:

  • point_cloud_paths – Input folder folder for the point clouds

  • polygon_path – Input folder folder for the polygon (pickle)

  • output_path – Output folder folder for the cropped point cloud

crop_circle(in_path='__auto__', out_path='__auto__', latitude=1, longitude=1, lat_lon_path='__auto__', radius=75, cols='', max_num_processes=0, extension_in_file='.laz', extension_out_file='.laz', extension_lat_lon_file='.laz', folder_parallel_processing='__auto__')

crop circle

Parameters:

  • in_path – input folder

  • out_path – output folder

  • latitude – latitude

  • longitude – longitude

  • lat_lon_path – (optional) folder with lat lon coordinates

  • radius – radius for cropping

  • cols – columns to be used, leave empty for all

  • max_num_processes – maximum number of processes

crop_points_to_polygon(in_points_path='__auto__', in_polygon_path='__auto__', out_path='__auto__', cols_in='', extension_in_points_file='.laz', extension_in_polygon_file='.pickle', extension_out_file='.laz', folder_parallel_processing='__auto__')

crop points to polygon

Parameters:

  • in_points_path – Input folder folder for the point cloud

  • in_polygon_path – Input folder folder for the polygon (pickle)

  • out_path – Output folder folder for the cropped point cloud

  • cols_in – columns to load

crop_to_equal_value_range(path1_in='segmented_object1', path2_in='segmented_object2', path1_out='crop_relative_height1', path2_out='crop_relative_height2', reference='max', axis=2, max_num_processes=0, folder_parallel_processing='__auto__')

crop to equal value range

Parameters:

  • path1_in – input folder

  • path2_in – input folder

  • path1_out – output folder

  • path2_out – output folder

  • reference – [max, min, same]: same value range relative to maximum point [max], relative to minimum point [min] or absolute coordinates [same]

  • axis – axis to crop values

  • max_num_processes – Number of parallel processes

density_based_clustering(pathname='__auto__', cluster_id_pathname='__auto__', cluster_centers_pathname='__auto__', wireframe_pathname='__auto__', epsilon=0.25, min_samples=0, dim=3, wireframe='False', extension_filename='.laz', extension_cluster_id_filename='.npy', extension_cluster_centers_filename='.laz', extension_wireframe_filename='.npy', folder_parallel_processing='__auto__')

Density-based Point Cloud Clustering

Parameters:

  • pathname – Input .laz folder folder

  • cluster_id_pathname – Output cluster IDs folder folder

  • cluster_centers_pathname – Output cluster centers .laz folder folder

  • wireframe_pathname – Output wireframe connections folder folder

  • epsilon – DBSCAN epsilon

  • min_samples – DBSCAN min_samples

  • dim – Point dimension

  • wireframe – Whether to compute wireframe connections

filter_label_disagreement_knn(path_points_in='__auto__', path_labels_in='__auto__', path_label_disagrement_in='__auto__', path_label_disagrement_out='__auto__', distance=2, classes_to_compare='2', comparison_type='2', class_to_filter=1, dim_data=3, knn=2, comparison_axis=-1, invalid_label=0, extension_file_points_in='.laz', extension_file_labels_in='.npy', extension_file_label_disagrement_in='.npy', extension_file_label_disagrement_out='.npy', folder_parallel_processing='__auto__')

filter label disagreement knn

Parameters:

  • path_points_in – input folder [.laz or .las]

  • path_labels_in – input folder [.txt or .npy]

  • path_label_disagrement_in – input folder[.txt or .npy]

  • path_label_disagrement_out – output folder [.txt or .npy]

  • distance – distance threshold

  • classes_to_compare – classes to compare, comma separated

  • comparison_type – [ge: greater equal, le: less equal]

  • class_to_filter – class to filter

  • dim_data – Dimensions to use: 3: x,y,z; 2: x, y

  • knn – k-nearest-neighbours

  • comparison_axis – axis to compare: -1: eucledian distance; 0, 1 or 2: distance along x, y or z axis

  • invalid_label – invalid label

filter_label_noise(path_in_data='__auto__', path_in_labels='__auto__', path_out='__auto__', k_nearest_neighbours=5, sigma=10.0, dim=3, invalid_label=0, extension_file_in_data='.laz', extension_file_in_labels='.labels', extension_file_out='.laz', folder_parallel_processing='__auto__')

filter label noise

Parameters:

  • path_in_data – input folder data

  • path_in_labels – input folder labels

  • path_out – output folder

  • k_nearest_neighbours – k nearest neighbours

  • sigma – sigma

  • dim – dim

  • invalid_label – invalid class label

fit_line_model(path_in='segmented_object', path_out='fit_line_model', residual_threshold=30.05, min_samples=2, max_trials=1, max_dim=3, max_num_processes=0, folder_parallel_processing='__auto__')

fit line model

Parameters:

  • path_in – input folder

  • path_out – output folder

  • residual_threshold – maximum quantile

  • min_samples – minimum quantile

  • max_trials – maximum number of trials

  • max_dim – max_dim 0: x, 1: y, 3: z

  • max_num_processes – Number of parallel processes

get_bounding_box(in_path='__auto__', dimension=3, out_path='__auto__', extension_in_file='.laz', extension_out_file='.npy', folder_parallel_processing='__auto__')

Get bounding box from las or laz file

Parameters:

  • in_path – Input .laz folder folder

  • dimension – Dimension of the point cloud

  • out_path – Output bounding box folder folder

get_meta_data(in_path='__auto__', out_path='__auto__', extension_in_file='.laz', extension_out_file='.json', folder_parallel_processing='__auto__')

Get meta data from las or laz file

Parameters:

  • in_path – Input .laz folder folder

  • out_path – Output meta data folder folder

get_point_values(path_source_in='__auto__', path_labels_out='__auto__', dtype='classification', decomposed_labels='True', extension_file_source_in='.laz', extension_file_labels_out='.txt', folder_parallel_processing='__auto__')

get point values

Parameters:

  • path_source_in – input folder [.laz or .las]

  • path_labels_out – output folder [.txt or .npy]

  • dtype – type

  • decomposed_labels – type

iterative_closest_point(path_source_in='__auto__', path_target_in='__auto__', path_source_out='__auto__', path_trafo_out='__auto__', metric='point2point', threshold=0.2, max_correspondences=5, extension_file_source_in='.laz', extension_file_target_in='.laz', extension_file_source_out='.laz', extension_file_trafo_out='.txt', folder_parallel_processing='__auto__')

iterative closest point

Parameters:

  • path_source_in – input source folder

  • path_target_in – input target folder

  • path_source_out – output folder

  • path_trafo_out – output transformation

  • metric – [max, min, same]: same value range relative to maximum point [max], relative to minimum point [min] or absolute coordinates [same]

  • threshold – threshold to crop values

  • max_correspondences – threshold max nearest neighbours

iterative_outlier_removal(path_in='segmented_object', path_out='iterative_outlier_removal', decay_factor=0.98, iteration_count=10, max_num_processes=0, folder_parallel_processing='__auto__')

iterative outlier removal

Parameters:

  • path_in – input folder

  • path_out – output folder

  • decay_factor – maximum quantile

  • iteration_count – minimum quantile

  • max_num_processes – Number of parallel processes

make_laz_from_values(path_values_in='__auto__', path_points_out='__auto__', dtype='X,Y,Z', scale='0.01,0.01,0.01', point_format=7, extension_file_values_in='.npy', extension_file_points_out='.laz', folder_parallel_processing='__auto__')

make laz from values

Parameters:

  • path_values_in – input folder data

  • path_points_out – output folder

  • dtype – data channels

  • scale – scale value

  • point_format – point format

make_line_model_from_points(path_in='segmented_object', path_out='vobject_coordinates3D', dim=3, max_num_processes=0, folder_parallel_processing='__auto__')

make line model from points

Parameters:

  • path_in – input folder data

  • path_out – output folder

  • dim – dimension

  • max_num_processes – maximum number of processes

point_cloud_to_dsm(path_points_in='__auto__', path_dsm_out='__auto__', path_dtm_out='__auto__', path_chm_out='__auto__', grid_size=0.5, dtype='float32', extension_file_points_in='.laz', extension_file_dsm_out='.tif', extension_file_dtm_out='.tif', extension_file_chm_out='.tif', folder_parallel_processing='__auto__')

point cloud to dsm

Parameters:

  • path_points_in – input points

  • path_dsm_out – dsm folder

  • path_dtm_out – dtm folder

  • path_chm_out – chm folder

  • grid_size – grid size

  • dtype – data type

quantile_filter(path_in='segmented_object', path_out='quantile_filterd', max_quantile=0.995, min_quantile=0.3, axis=2, max_num_processes=0, folder_parallel_processing='__auto__')

quantile filter

Parameters:

  • path_in – input folder

  • path_out – output folder

  • max_quantile – maximum quantile

  • min_quantile – minimum quantile

  • axis – axis 0: x, 1: y, 2: z

  • max_num_processes – Number of parallel processes

retile_generate_grid_globally(in_paths='__auto__', dimension=3, grid_size='20,20,50', offset_factor=0.0, reference_point='', out_path_tiles='__auto__', out_path_mapping_slice_point_cloud='slices', out_path_mapping_point_cloud_to_tiles='mapping_point_cloud_to_tiles', out_path_mapping_tiles_to_point_cloud='mapping_tiles_to_point_cloud', extension_in_paths='.laz', extension_out_path_tiles='', folder_parallel_processing='__auto__')

Create grid for retileing point clouds over multiple georeferenced point clouds

Parameters:

  • in_paths – folder to laz folders to be retiled

  • dimension – Dimension to be retiled (x,y) or (x,y,z)

  • grid_size – Grid size for retileing

  • offset_factor – Offset factor for grid generation

  • reference_point – Reference point for grid generation, empty for default (min_x, min_y, min_z)

  • out_path_tiles – Output bounding box / tiles folder

  • out_path_mapping_slice_point_cloud – Output path for mapping that contains the point clouds (including neighbouring point clouds) that are used to generate slices from point cloud x

  • out_path_mapping_point_cloud_to_tiles – Output path for mapping that contains the tiles that are generated from point cloud x

  • out_path_mapping_tiles_to_point_cloud – Output path for mapping that contains the point clouds that are used to generate tile x

retile_generate_grid_locally(in_path='__auto__', dimension=3, grid_size='20,20,50', offset_factor=0.0, reference_point='', out_path_tiles='__auto__', prefix_tiles='grid_', out_path_mapping_slice_point_cloud='__auto__', out_path_mapping_point_cloud_to_tiles='__auto__', out_path_mapping_tiles_to_point_cloud='__auto__', extension_in_path='.laz', extension_out_path_tiles='', extension_out_path_mapping_slice_point_cloud='.txt', extension_out_path_mapping_point_cloud_to_tiles='.txt', extension_out_path_mapping_tiles_to_point_cloud='.txt', folder_parallel_processing='__auto__')

Create grid for retileing individual point clouds

Parameters:

  • in_path – folder to laz folders to be retiled

  • dimension – Dimension to be retiled (x,y) or (x,y,z)

  • grid_size – Grid size for retileing

  • offset_factor – Offset factor for grid generation

  • reference_point – Reference point for grid generation, empty for default (min_x, min_y, min_z)

  • out_path_tiles – Output bounding box / tiles folder

  • prefix_tiles – Prefix for the tiles

  • out_path_mapping_slice_point_cloud – Output folder for mapping that contains the point clouds (including neighbouring point clouds) that are used to generate slices from point cloud x

  • out_path_mapping_point_cloud_to_tiles – Output folder for mapping that contains the tiles that are generated from point cloud x

  • out_path_mapping_tiles_to_point_cloud – Output folder for mapping that contains the point clouds that are used to generate tile x

retile_grid_to_point_cloud(in_path_grid='point_cloud_grid', in_path_mapping='__auto__', out_path_points='__auto__', cols='', extension_in_path_mapping='.txt', extension_out_path_points='.laz', folder_parallel_processing='__auto__')

retile point clouds to grid

Parameters:

  • in_path_grid – folder that contains the retiled point clouds

  • in_path_mapping – Mapping that specifies, which point clouds of the grid intersect with the original point cloud

  • out_path_points – Output folder for the merged point cloud

  • cols – colums used from point cloud, default is all columns

retile_grid_to_point_probabilities(in_path_point_grid='None', in_path_probabilities='point_cloud_probabilities', in_path_mapping='__auto__', out_path_probabilities='__auto__', class_names='1,2,3,4,5,6,7,8,9,10', extension_in_path_mapping='.txt', extension_out_path_probabilities='.npy', folder_parallel_processing='__auto__')

retile point clouds to grid

Parameters:

  • in_path_point_grid – folder that contains the retiled point clouds

  • in_path_probabilities – folder that contains the probabilities of the retiled point clouds

  • in_path_mapping – Mapping that specifies, which point clouds of the grid intersect with the original point cloud

  • out_path_probabilities – Output folder for the merged probabilities

  • class_names – class names

retile_point_cloud_to_grid(in_path_points='__auto__', in_path_grids='grid1.npy,grid2.npy,grid3.npy', out_path_points='out.laz,out2.laz', cols='', extension_in_path_points='.laz', folder_parallel_processing='__auto__')

retile point clouds to grid

Parameters:

  • in_path_points – Output folder for mapping that contains the point clouds (including neighbouring point clouds) that are used to generate slices from point cloud x

  • in_path_grids – Output path for mapping that contains the tiles that are generated from point cloud x

  • out_path_points – Output path to retiled point clouds

  • cols – colums used from point cloud, default is all columns

select_center_object(in_directory='laz_files', out_path='__auto__', latitude=1, longitude=1, extension_out_file='.laz', folder_parallel_processing='__auto__')

select center object

Parameters:

  • in_directory – input directory

  • out_path – output folder

  • latitude – latitude

  • longitude – longitude

select_points_by_value(path_source_in='__auto__', min_value=1, max_value=1, attribute='classification', path_source_out='__auto__', keep_empty='True', extension_file_source_in='', extension_file_source_out='', folder_parallel_processing='__auto__')

Selects points by value of attribute

Parameters:

  • path_source_in – input folder data

  • min_value – minimum value

  • max_value – maximum value

  • attribute – feature for selection

  • path_source_out – output folder

  • keep_empty – save empty files

uniform_down_sampling(input_path='__auto__', cols='', output_path='__auto__', every_k_points=2, extension_input_file='.laz', extension_output_file='.laz', folder_parallel_processing='__auto__')

Uniform down sampling of point cloud

Parameters:

  • input_path – Input point cloud folder

  • cols – Columns to read from input file, default is all columns

  • output_path – Output point cloud folder

  • every_k_points – Keep every k points

uniform_down_sampling_voxel(input_path='__auto__', cols='', output_path='__auto__', voxel_size=0.05, extension_input_file='.laz', extension_output_file='.laz', folder_parallel_processing='__auto__')

Uniform down sampling of point cloud using voxel grids

Parameters:

  • input_path – Input point cloud folder

  • cols – Columns to read from input file, default is all columns

  • output_path – Output point cloud folder

  • voxel_size – voxel size

uniform_downsampling(path_in='__auto__', path_out='__auto__', k=3, dtype='', extension_file_in='.laz', extension_file_out='.laz', folder_parallel_processing='__auto__')

uniform downsampling

Parameters:

  • path_in – input folder data

  • path_out – output folder

  • k – k

  • dtype – values from point cloud, e.g. X,Y,Z

voxel_downsampling(path_in='__auto__', path_out='__auto__', voxel_size=0.1, dtype='', extension_file_in='.laz', extension_file_out='.laz', folder_parallel_processing='__auto__')

deprecated, please use unfiorm_down_sampling_voxel instead!

Parameters:

  • path_in – input folder data

  • path_out – output folder

  • voxel_size – voxel size

  • dtype – values from point cloud, e.g. X,Y,Z

class qc(outer_class)

Bases: object

Namespace for qc functions.

report_image_completeness(in_path='__auto__', in_meta_data_path='__auto__', out_path='__auto__', grid_size=0.5, populated_class=1, small_holes_class=100, large_holes_class=255, keep_error_free='True', extension_in_file='.txt', extension_in_meta_data_file='.json', extension_out_file='.txt', folder_parallel_processing='__auto__')

report image completeness

Parameters:

  • in_path – folder with count of classes

  • in_meta_data_path – folder with metadata

  • out_path – output report folder

  • grid_size – grid size

  • populated_class – populated class

  • small_holes_class – small holes class

  • large_holes_class – large holes class

  • keep_error_free – Save empty files?

report_lidar_completeness(in_path='__auto__', out_path='__auto__', grid_size=0.5, populated_class=1, small_holes_class=100, large_holes_class=255, keep_error_free='True', extension_in_file='.txt', extension_out_file='.txt', folder_parallel_processing='__auto__')

report lidar completeness

Parameters:

  • in_path – folder with erroneous points

  • out_path – output report folder

  • grid_size – grid size

  • populated_class – populated class

  • small_holes_class – small holes class

  • large_holes_class – large holes class

  • keep_error_free – Save empty files?

report_qc_classification(in_path='__auto__', out_path='__auto__', error_classes='148,149', error_names='Line,Tower', keep_error_free='True', extension_in_file='.laz', extension_out_file='.txt', folder_parallel_processing='__auto__')

report qc classification

Parameters:

  • in_path – folder with erroneous points

  • out_path – output report folder

  • error_classes – error classes

  • error_names – error names

  • keep_error_free – Save empty files?

report_vegetation_occurance(in_path='__auto__', out_path='__auto__', ground_classes_old='2,3,6,7,15', ground_classes_new='1,3,9,11,15', vegetation_old='6,7,15', vegetation_new='9,11,15', keep_error_free='True', extension_in_file='.txt', extension_out_file='.txt', folder_parallel_processing='__auto__')

report vegetation occurance

Parameters:

  • in_path – folder with erroneous points

  • out_path – output report folder

  • ground_classes_old – ground classes

  • ground_classes_new – ground classes

  • vegetation_old – vegetation old classes

  • vegetation_new – vegetation new classes

  • keep_error_free – Save empty files?

class shp(outer_class)

Bases: object

Namespace for shp functions.

extract_multipolygons_from_shp(shp_path='__auto__', out_polygon_path='polygons/', out_attributes_path='attributes/', shape_id=-1, name_id=0, extension_shp_file='', folder_parallel_processing='__auto__')

extract multipolygons from shp

Parameters:

  • shp_path – input shp folder

  • out_polygon_path – folder with polygons from shape file

  • out_attributes_path – folder with records from shape file

  • shape_id – id of polygon: [-1 parses all polygons]

  • name_id – id of polygon: [-1 ignores name]

intersecting_polygons(input_path='__auto__', comparison_path='polygons', output_path='__auto__', extension_input_file='.pickle', extension_output_file='.txt', folder_parallel_processing='__auto__')

intersecting polygons

Parameters:

  • input_path – Input folder for the polygon

  • comparison_path – Input folder containing polygons for comparison

  • output_path – Output folder for the list of intersecting polygon foldernames

make_polygon_from_json(input_path='__auto__', output_path='__auto__', point_identifiers='min_x,min_y;min_x,max_y;max_x,max_y;max_x,min_y', extension_input_file='.json', extension_output_file='.pickle', folder_parallel_processing='__auto__')

make polygon from json

Parameters:

  • input_path – Input folder for the json folder

  • output_path – Output folder for the polygon folder

  • point_identifiers – Point identifiers for the polygon

wireframe_to_dxf(input_path='__auto__', edges_path='__auto__', output_path='__auto__', extension_input_file='.laz', extension_edges_file='.npy', extension_output_file='.dxf', folder_parallel_processing='__auto__')

wireframe to dxf

Parameters:

  • input_path – Input folder for the vertices

  • edges_path – Input folder for the edges

  • output_path – Output folder for the dxf model

class sys(outer_class)

Bases: object

Namespace for sys functions.

copy_file_in_cloud(target='__auto__', destination='__auto__', extension_target='', extension_destination='', folder_parallel_processing='__auto__')

copy file in cloud

Parameters:

  • target – Target to be moved

  • destination – Destination

create_directory_in_cloud(destination='__auto__', extension_destination='', folder_parallel_processing='__auto__')

create directory in cloud

Parameters:

destination – Destionation location on host. default folder: ./data

download_data_to_cloud(url='__auto__', destination='__auto__', protocol='', download_type=0, username='', password='', port=21, extension_url='', extension_destination='', folder_parallel_processing='__auto__')

download data to cloud

Parameters:

  • url – URL to data

  • destination – Destionation location on host. default folder: ./data

  • protocol – protocol: : automatically try to infer protocol, ftp: ftp, sftp: sftp

  • download_type – download type: 0: all files from folder, 1: individual file

  • username – Username

  • password – Password

  • port – port

download_from_host_to_aipha(url='__auto__', port='22', username='ubuntu', identity_path='', location='file.laz', destination='__auto__', extension_url='.1', extension_destination='.laz', folder_parallel_processing='__auto__')

Download a path from a host via ssh

Parameters:

  • url – Url to host

  • port – Port to host

  • username – Username to host

  • identity_path – Path to identity file on aipha

  • location – Path to download from host

  • destination – Location to upload to aipha

download_from_s3_to_aipha(access_key_id='YOUR_KEY_ID', secret_access_key='YOUR_SECRET_KEY', aws_region='eu-central-1', location='file.laz', destination='__auto__', bucket_name='Your S3 bucket', extension_destination='.laz', folder_parallel_processing='__auto__')

Download a path from a S3 bucket

Parameters:

  • access_key_id – AWS access key ID

  • secret_access_key – AWS secret access key

  • aws_region – AWS region

  • location – Path to download from s3

  • destination – Location to upload to aipha

  • bucket_name – S3 bucket name

find_file_paths(input_paths='__auto__', output_paths='__auto__', search_path='/search_folder', replace_in='', replace_out='', substrings='', extension_input_files='.txt', extension_output_files='.txt', folder_parallel_processing='__auto__')

find file paths

Parameters:

  • input_paths – File containing the list of foldernames

  • output_paths – Path to save the modified folderlist

  • search_path – Folder to traverse for finding files

  • replace_in – The part to replace in the filenames

  • replace_out – The new part to replace with

  • substrings – a list of substrings that need to occure in the file paths to be vallid

list_files_in_cloud(target='__auto__', path_out='__auto__', extension_target='', extension_file_out='.txt', folder_parallel_processing='__auto__')

list files in cloud

Parameters:

  • target – Target to be listet

  • path_out – output_folder

move_file_in_cloud(target='__auto__', destination='__auto__', extension_target='', extension_destination='', folder_parallel_processing='__auto__')

move file in cloud

Parameters:

  • target – Target to be moved

  • destination – Destination

recursive_list(target='__auto__', destination='__auto__', extension_target='', extension_destination='.txt', folder_parallel_processing='__auto__')

recursive list

Parameters:

  • target – Target folder to be listed recursively

  • destination – Output folder

remove_files_from_cloud(target='__auto__', extension_target='', folder_parallel_processing='__auto__')

remove files from cloud

Parameters:

target – Target to be deleted

rename_file_in_cloud(target='__auto__', prefix='', suffix='', replace_from='', replace_to='', replace_count=0, extension_target='', folder_parallel_processing='__auto__')

rename file in cloud

Parameters:

  • target – Target to be renamed

  • prefix – add prefix

  • suffix – add suffix

  • replace_from – replace string in filename

  • replace_to – replace string in filename

  • replace_count – replace string in filename

select_by_identifier(original_path='original_folder', original_identifier_path='__auto__', output_path='output_folder', extension_original_identifier_file='.txt', folder_parallel_processing='__auto__')

select by identifier

Parameters:

  • original_path – original folder

  • original_identifier_path – original identifiers

  • output_path – output folder

select_corresponding_path(original_path='__auto__', original_identifier_path='__auto__', corresponding_path='__auto__', output_path='__auto__', selection_criteria='oldest', default_value='__original__', extension_original_file='.txt', extension_original_identifier_file='.txt', extension_corresponding_file='.txt', extension_output_file='.txt', folder_parallel_processing='__auto__')

select corresponding path

Parameters:

  • original_path – original folders

  • original_identifier_path – original identifiers

  • corresponding_path – corresponding folders

  • output_path – output folder

  • selection_criteria – selection criteria: [oldest, newest, shortest, longest]

  • default_value – default value if no corresponding path is found

split_path(in_path='__auto__', out_path='__auto__', split_type='filename', extension_in_path='', extension_out_path='', folder_parallel_processing='__auto__')

split path

Parameters:

  • in_path – input folder

  • out_path – output folder

  • split_type – split type: [filename, dirname, basename, ext]

touch_file_in_cloud(target='__auto__', extension_target='.txt', folder_parallel_processing='__auto__')

touch file in cloud

Parameters:

target – File to be touched

upload_data_from_cloud(url='__auto__', target='__auto__', protocol='', username='', password='', port=21, extension_url='', extension_target='', folder_parallel_processing='__auto__')

upload data from cloud

Parameters:

  • url – destination URL

  • target – Target location on host for upload. default folder: ./data

  • protocol – protocol: : automatically try to infer protocol, ftp: ftp, sftp: sftp

  • username – Username

  • password – Password

  • port – port

upload_from_aipha_to_host(url='__auto__', port='22', username='ubuntu', identity_path='', target='__auto__', location='file.laz', extension_url='.1', extension_target='.laz', folder_parallel_processing='__auto__')

Upload a path to a host via ssh

Parameters:

  • url – Url to host

  • port – Port to host

  • username – Username to host

  • identity_path – Path to identity file on aipha

  • target – Path to upload from aipha

  • location – Location of file to upload on host

upload_from_aipha_to_s3(access_key_id='YOUR_KEY_ID', secret_access_key='YOUR_SECRET_KEY', aws_region='eu-central-1', target='__auto__', location='file.laz', bucket_name='Your S3 bucket', extension_target='.laz', folder_parallel_processing='__auto__')

Upload a path to a S3 bucket

Parameters:

  • access_key_id – AWS access key ID

  • secret_access_key – AWS secret access key

  • aws_region – AWS region

  • target – Path to upload from aipha

  • location – Location of file to upload on s3

  • bucket_name – S3 bucket name

class tdp(outer_class)

Bases: object

Namespace for tdp functions.

convert_laz_point_formats(path_in='__auto__', path_out='__auto__', format=7, extension_file_in='.laz', extension_file_out='.labels', folder_parallel_processing='__auto__')

convert laz point formats

Parameters:

  • path_in – input folder

  • path_out – results folder

  • format – format

merge_and_split_results_csv(new_tower_path='new_paths.txt', last_tower_path='last_paths.txt', reference_tower_path='reference_paths.txt', results_path_csv='results.csv', results_plots_path='results_plots', merged_results_path_csv='results/Reports_2023', resturctured_plots_path='results/10-Plots-Tragwerke', input_path_structure_path='input_file_structure.txt', year='2023', folder_parallel_processing='__auto__')

[atr] Merge results csv

Parameters:

  • new_tower_path – input new path data

  • last_tower_path – input last path

  • reference_tower_path – input reference path

  • results_path_csv – input results.csv path

  • results_plots_path – input results_plots path

  • merged_results_path_csv – output path

  • resturctured_plots_path – output path

  • input_path_structure_path – input file structure path

  • year – year

point_cloud_classification_inference(path_in='__auto__', path_out='__auto__', model_path='network_parameters', cols_data='X,Y,Z', cols_labels='classification', extension_file_in='.laz', extension_file_out='.labels', folder_parallel_processing='__auto__')

point cloud classification inference

Parameters:

  • path_in – input folder

  • path_out – results folder

  • model_path – path to model

  • cols_data – attributes used

  • cols_labels – label name

point_cloud_filter_label_noise(path_in_data='__auto__', path_in_labels='__auto__', path_out='__auto__', k_nearest_neighbours=5, sigma=10.0, dim=3, invalid_label=0, extension_file_in_data='.laz', extension_file_in_labels='.labels', extension_file_out='.laz', folder_parallel_processing='__auto__')

point cloud filter label noise

Parameters:

  • path_in_data – input folder data

  • path_in_labels – input folder labels

  • path_out – output folder

  • k_nearest_neighbours – k nearest neighbours

  • sigma – sigma

  • dim – dim

  • invalid_label – invalid class label

segment_objects(in_points_path='__auto__', in_labels_path='__auto__', out_directory='segmented_object', out_prefix='object', label_col='classification', object_class=68, max_distance=2, min_points=100, extension_in_points_file='', extension_in_labels_file='', folder_parallel_processing='__auto__')

segment objects

Parameters:

  • in_points_path – input folder points

  • in_labels_path – input folder labels

  • out_directory – output directory

  • out_prefix – output filename prefix

  • label_col – label column id

  • object_class – obejct class

  • max_distance – maximum distance for segmentation

  • min_points – minimum number of points

tower_displacement(laz_in_path_new='__auto__', laz_in_path_old='__auto__', laz_in_path_ref='__auto__', tower_name='', year_new='2022', year_old='2020', year_ref='2018', results_out_path='__auto__', plots_out_path='plots/', extension_laz_in_file_new='.laz', extension_laz_in_file_old='.laz', extension_laz_in_file_ref='.laz', extension_results_out_file='.txt', folder_parallel_processing='__auto__')

tower displacement

Parameters:

  • laz_in_path_new – laz input folder new data

  • laz_in_path_old – laz input folder last data

  • laz_in_path_ref – laz input folder first data

  • tower_name – tower name

  • year_new – year of new data

  • year_old – year of old data

  • year_ref – year of reference data

  • results_out_path – result folder folder

  • plots_out_path – result folder path

class val(outer_class)

Bases: object

Namespace for val functions.

add_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=0.0, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Add a constant value to a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Constant value to add (default: 0.0)

argmax(inpath='__auto__', outpath='__auto__', dtype='float', axis=-1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Argmax of a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • axis – Axis to find the argmax (default: None)

argmin(inpath='__auto__', outpath='__auto__', dtype='float', axis=-1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Argmin of a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • axis – Axis to find the argmax (default: None)

connected_components_labeling(pathname_in='__auto__', pathname_out='__auto__', dtype='float', no_type=0.0, value=1.0, extension_filename_in='.npy', extension_filename_out='.npy', folder_parallel_processing='__auto__')

Perform connected components labeling on a matrix.

Parameters:

  • pathname_in – Input folder folder for the matrix

  • pathname_out – Output folder folder for the labeled matrix

  • dtype – Data type of the matrix (default: float)

  • no_type – Value representing no_type in the matrix (default: 0.0)

  • value – Value representing value in the matrix (default: 1.0)

count_unique_values(pathname_in='__auto__', pathname_out='__auto__', dtype='float', ignore='nan', extension_filename_in='.npy', extension_filename_out='.npy', folder_parallel_processing='__auto__')

Count unique occurrences of values in a matrix.

Parameters:

  • pathname_in – Input folder folder for the matrix

  • pathname_out – Output folder folder for the unique counts matrix

  • dtype – Data type of the matrix (default: float)

  • ignore – Data value to ignore

divide_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1.0, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Divide a constant value from a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Constant value to divide (default: 1.0)

equal_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Equal operator on a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Value to compare (default: 1)

greater_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Greater operator on a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Value to compare (default: 1)

greater_equal_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Greater equal operator on a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Value to compare (default: 1)

hstack(path_values_in='__auto__', path_values_out='__auto__', dtype='str', extension_file_values_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')
Parameters:

  • path_values_in – input folder [.npy, .labels or .txt]

  • path_values_out – output folder [.txt or .npy]

  • dtype – data type

less_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Less operator on a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Value to compare (default: 1)

less_equal_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Less equal operator on a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Value to compare (default: 1)

mask_subset(path_values1_in='__auto__', path_mask_in='__auto__', path_values_out='__auto__', extension_file_values1_in='.npy', extension_file_mask_in='.npy', extension_file_values_out='.txt', folder_parallel_processing='__auto__')

mask subset

Parameters:

  • path_values1_in – input folder [.txt or .npy]

  • path_mask_in – input folder that contains [0,1] values

  • path_values_out – output folder [.txt or .npy]

masked_assign_constant(path_values_in='__auto__', constant=0.0, path_mask_in='__auto__', path_values_out='__auto__', extension_file_values_in='.npy', extension_file_mask_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

masked assign constant

Parameters:

  • path_values_in – input folder [.txt or .npy]

  • constant – constant value to assign

  • path_mask_in – input folder that contains [0,1] values

  • path_values_out – output folder [.txt or .npy]

max(inpath='__auto__', outpath='__auto__', dtype='float', extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Maximum of a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

min(inpath='__auto__', outpath='__auto__', dtype='float', extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Minimum of a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

multiply_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1.0, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Multiply a constant value from a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Constant value to multiply (default: 1.0)

not_equal_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Not equal operator on a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Value to compare (default: 1)

remap_values(path_values_in='__auto__', path_values_out='__auto__', map_in='1,2,3,4', map_out='3,1,2,2', dtype_in='int32', dtype_out='int32', unmapped='0', extension_file_values_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

remap values

Parameters:

  • path_values_in – input folder [.txt, .labels or .npy]

  • path_values_out – output folder [.txt, .labels or .npy]

  • map_in – map in

  • map_out – map out

  • dtype_in – data type input

  • dtype_out – data type output

  • unmapped – default value for values where no mapping exists

replace_in_string_array(path_in='__auto__', path_out='__auto__', tokens_to_replace='', replacement_tokens='', extension_file_in='.txt', extension_file_out='.txt', folder_parallel_processing='__auto__')

Replace tokens in a string array

Parameters:

  • path_in – Input foldername

  • path_out – Output foldername

  • tokens_to_replace – Tokens to replace, comma separated

  • replacement_tokens – Replacement tokens, comma separated

replace_strings(path_in='__auto__', path_out='__auto__', replace_from='', replace_to='', extension_file_in='.txt', extension_file_out='.txt', folder_parallel_processing='__auto__')

Substrings replacement in an ASCII file

Parameters:

  • path_in – Path to the input folder

  • path_out – Path to the output folder

  • replace_from – Comma-separated list of substrings to replace

  • replace_to – Comma-separated list of replacement substrings

resize_slice_matrix(pathname_in='__auto__', pathname_out='__auto__', dtype='float', indices=':,124:,:3', default_value=0.0, extension_filename_in='.npy', extension_filename_out='.npy', folder_parallel_processing='__auto__')

Resize and slice a matrix based on indices.

Parameters:

  • pathname_in – Input folder folder for the matrix

  • pathname_out – Output folder folder for the resized and sliced matrix

  • dtype – Data type of the matrix (default: float)

  • indices – Indices to slice the matrix (in NumPy slicing convention)

  • default_value – Default value to fill when resizing (default: 0.0)

slice_string_array(path_in='__auto__', path_out='__auto__', slices='', extension_file_in='.txt', extension_file_out='.txt', folder_parallel_processing='__auto__')

Slice a string array

Parameters:

  • path_in – Input foldername

  • path_out – Output foldername

  • slices – Slices to take

sliced_assign_constant(path_in='__auto__', path_out='__auto__', indices=':', constant=0.0, extension_file_in='.npy', extension_file_out='.txt', folder_parallel_processing='__auto__')

sliced assign constant

Parameters:

  • path_in – input folder [.txt or .npy]

  • path_out – output folder [.txt or .npy]

  • indices – indices to slice

  • constant – constant value to assign

subtract_constant(inpath='__auto__', outpath='__auto__', dtype='float', constant=0.0, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Subtract a constant value from a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • constant – Constant value to subtract (default: 0.0)

sum(inpath='__auto__', outpath='__auto__', dtype='float', axis=-1, extension_infile='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Sum all values of a matrix.

Parameters:

  • inpath – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

  • axis – axis to sum [default -1: no axis is used]

values_add(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values add

Parameters:

  • path_values1_in – input folder [.npy, .labels or .txt]

  • path_values2_in – input folder [.npy, .labels or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

values_assign(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values assign

Parameters:

  • path_values1_in – input folder [.npy or .txt]

  • path_values2_in – input folder [.npy or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

values_distance(pathname_is='__auto__', pathname_should='__auto__', output_path='__auto__', dtype='float', no_type=0.0, value=1.0, gridsize=1.0, extension_filename_is='.npy', extension_filename_should='.npy', extension_output_file='.npy', folder_parallel_processing='__auto__')

Compute Euclidean distance from is matrix to should matrix.

Parameters:

  • pathname_is – Input folder folder for is matrix

  • pathname_should – Input folder folder for should matrix

  • output_path – Output folder folder for distances matrix

  • dtype – Data type of the matrices (default: float)

  • no_type – Value representing no_type in the matrices (default: 0.0)

  • value – Value representing value in the matrices (default: 1.0)

  • gridsize – Resolution of the spatial grid in meters (default: 1.0)

values_divide(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values divide

Parameters:

  • path_values1_in – input folder [.npy or .txt]

  • path_values2_in – input folder [.npy or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

values_equal(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values equal

Parameters:

  • path_values1_in – input folder [.npy or .txt]

  • path_values2_in – input folder [.npy or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

values_greater(inpath1='__auto__', inpath2='__auto__', outpath='__auto__', dtype='float', extension_infile1='.npy', extension_infile2='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Elementwiese greater operator on a matrix.

Parameters:

  • inpath1 – Input folder folder

  • inpath2 – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

values_greater_equal(inpath1='__auto__', inpath2='__auto__', outpath='__auto__', dtype='float', extension_infile1='.npy', extension_infile2='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Elementwiese greater equal operator on a matrix.

Parameters:

  • inpath1 – Input folder folder

  • inpath2 – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

values_hstack(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', dtype='str', extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values hstack

Parameters:

  • path_values1_in – input folder [.npy, .labels or .txt]

  • path_values2_in – input folder [.npy, .labels or .txt]

  • path_values_out – output folder [.txt or .npy]

  • dtype – data type

values_less(inpath1='__auto__', inpath2='__auto__', outpath='__auto__', dtype='float', extension_infile1='.npy', extension_infile2='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Elementwiese less operator on a matrix.

Parameters:

  • inpath1 – Input folder folder

  • inpath2 – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

values_less_equal(inpath1='__auto__', inpath2='__auto__', outpath='__auto__', dtype='float', extension_infile1='.npy', extension_infile2='.npy', extension_outfile='.npy', folder_parallel_processing='__auto__')

Elementwiese less equal operator on a matrix.

Parameters:

  • inpath1 – Input folder folder

  • inpath2 – Input folder folder

  • outpath – Output folder folder

  • dtype – Data type of the matrix (default: float)

values_masked_assign(path_values1_in='__auto__', path_values2_in='__auto__', path_mask_in='__auto__', path_values_out='__auto__', extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_mask_in='.npy', extension_file_values_out='.txt', folder_parallel_processing='__auto__')

values masked assign

Parameters:

  • path_values1_in – input folder [.txt or .npy]

  • path_values2_in – input folder [.txt or .npy]

  • path_mask_in – input folder that contains [0,1] values

  • path_values_out – output folder [.txt or .npy]

values_multiply(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values multiply

Parameters:

  • path_values1_in – input folder [.npy or .txt]

  • path_values2_in – input folder [.npy or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

values_not_equal(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values not equal

Parameters:

  • path_values1_in – input folder [.npy or .txt]

  • path_values2_in – input folder [.npy or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

values_sliced_assign(path_in='__auto__', path_out='__auto__', indices=':', path_values_in='__auto__', default_value=0.0, extension_file_in='.npy', extension_file_out='.txt', extension_file_values_in='.npy', folder_parallel_processing='__auto__')

values sliced assign

Parameters:

  • path_in – input folder [.txt or .npy]

  • path_out – output folder [.txt or .npy]

  • indices – indices to slice

  • path_values_in – input folder [.txt or .npy]

  • default_value – default value to assign

values_subtract(path_values1_in='__auto__', path_values2_in='__auto__', path_values_out='__auto__', ignore_label=nan, value_subset1=nan, extension_file_values1_in='.npy', extension_file_values2_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')

values subtract

Parameters:

  • path_values1_in – input folder [.npy or .txt]

  • path_values2_in – input folder [.npy or .txt]

  • path_values_out – output folder [.txt or .npy]

  • ignore_label – ignore value default: nan

  • value_subset1 – ignore value default: nan

vstack(path_values_in='__auto__', path_values_out='__auto__', dtype='str', extension_file_values_in='.npy', extension_file_values_out='.npy', folder_parallel_processing='__auto__')
Parameters:

  • path_values_in – input folder [.npy, .labels or .txt]

  • path_values_out – output folder [.txt or .npy]

  • dtype – data type