bird_cloud_gnn.radar_dataset

Module for reading the files and passing as datasets to DGL

Module Contents

Classes

RadarDataset

Dataset for DGL created from CSVs in a folder.

class bird_cloud_gnn.radar_dataset.RadarDataset(data, features, target, name='Radar', num_nodes=100, max_edge_distance=50.0, max_poi_per_label=200, points_of_interest=None, use_missing_indicator_columns=False, add_edges_to_poi=False, skip_cache=False)

Bases: dgl.data.DGLDataset

Dataset for DGL created from CSVs in a folder.

For every labeled point in the point cloud, the number of neighbours in a specific radius is checked. If the number of neighbours is big enough, the point is selected. A data point is a graph created by taking a selected point and a number of its neighbours.

data_folder

Folder with the CSV files.

Type:

str

features

List of features expected to be present at every CSV file.

Type:

array of str

target

Target column. 0, 1 or missing expected.

Type:

str

num_nodes

If a point has less than this amount of neighbours, it is ignored.

Type:

int

max_edge_distance

Creates a edge between two nodes if their distance is less than this value.

Type:

float

max_poi_per_label

Select at most this amount of POIs. If there are more POIs, they are chosen randomly.

Type:

int

missing_indicator_skip_columns = ['range', 'azimuth', 'elevation', 'x', 'y', 'z', 'centered_x', 'centered_y']

oneline_description()

Description of the dataset to uniquely identify it in logs

_read_one_file(data_path)

Reads a file and creates the graphs and labels for it.

_process_data(data, origin='')

process()

Internal function for the DGLDataset. Process the folder to create the graphs.

save()

Overwite to realize your own logic of saving the processed dataset into files.

It is recommended to use dgl.data.utils.save_graphs to save dgl graph into files and use dgl.data.utils.save_info to save extra information into files.

load()

Overwite to realize your own logic of loading the saved dataset from files.

It is recommended to use dgl.data.utils.load_graphs to load dgl graph from files and use dgl.data.utils.load_info to load extra information into python dict object.

cache_dir()

has_cache()

Overwrite to realize your own logic of deciding whether there exists a cached dataset.

By default False.

__getitem__(i)

Gets the data object at index.

__len__()

The number of examples in the dataset.