def test_set_density_map_sigma():
model = dpp.HeatmapObjectCountingModel()
assert model._density_sigma == 5
with pytest.raises(TypeError):
model.set_density_map_sigma('4')
model.set_density_map_sigma(2.0)
assert model._density_sigma == 2.0
def test_heatmap_csv_data_load(test_data_dir):
im_dir = os.path.join(test_data_dir, 'test_Ara2013_heatmap')
expected_heatmap_dir = os.path.join(os.path.curdir, 'generated_heatmaps')
if os.path.exists(expected_heatmap_dir):
shutil.rmtree(expected_heatmap_dir)
model = dpp.HeatmapObjectCountingModel()
model.set_image_dimensions(128, 128, 3)
assert model._raw_image_files is None
assert model._raw_labels is None
base_names = [
'ara2013_plant007_rgb', 'ara2013_plant008_rgb', 'ara2013_plant001_rgb',
'ara2013_plant002_rgb', 'ara2013_plant003_rgb', 'ara2013_plant004_rgb',
'ara2013_plant005_rgb', 'ara2013_plant006_rgb'
]
expected_images = [
os.path.join(im_dir, '{}.png'.format(x)) for x in base_names
]
expected_labels = [
os.path.join(expected_heatmap_dir, '{}.npy'.format(x))
for x in base_names
]
# Load data from CSV; generate heatmaps
model.load_heatmap_dataset_with_csv_from_directory(im_dir,
'point_labels.csv',
ext='png')
assert model._raw_image_files == expected_images
assert model._raw_labels == expected_labels
# Load data from CSV and pre-existing heatmaps
model._raw_image_files = None
model._raw_labels = None
model.load_heatmap_dataset_with_csv_from_directory(im_dir,
'point_labels.csv',
ext='png')
assert model._raw_image_files == expected_images
assert model._raw_labels == expected_labels
shutil.rmtree(expected_heatmap_dir)
import deepplantphenomics as dpp
model = dpp.HeatmapObjectCountingModel(debug=True, load_from_saved=False)
# 3 channels for colour, 1 channel for greyscale
channels = 3
# Setup and hyper-parameters
model.set_image_dimensions(128, 128, channels)
model.set_batch_size(32)
model.set_number_of_threads(4)
model.set_learning_rate(0.0001)
model.set_maximum_training_epochs(25)
model.set_test_split(0.75)
model.set_validation_split(0.0)
# Load dataset
model.set_density_map_sigma(4.0)
model.load_heatmap_dataset_with_csv_from_directory('./data', 'point_labels.csv')
# Define a model architecture
model.add_input_layer()
model.add_convolutional_layer(filter_dimension=[3, 3, 3, 16], stride_length=1, activation_function='relu')
model.add_convolutional_layer(filter_dimension=[3, 3, 16, 32], stride_length=1, activation_function='relu')
model.add_convolutional_layer(filter_dimension=[5, 5, 32, 32], stride_length=1, activation_function='relu')
model.add_output_layer()
# Train!
with pytest.raises(ValueError):
model.set_patch_size(-1, 1)
with pytest.raises(TypeError):
model.set_patch_size(1, 1.0)
with pytest.raises(ValueError):
model.set_patch_size(1, -1)
@pytest.mark.parametrize(
"model,bad_loss,good_loss",
[(dpp.ClassificationModel(), 'l2', 'softmax cross entropy'),
(dpp.RegressionModel(), 'softmax cross entropy', 'l2'),
(dpp.SemanticSegmentationModel(), 'l2', 'sigmoid cross entropy'),
(dpp.ObjectDetectionModel(), 'l2', 'yolo'),
(dpp.CountCeptionModel(), 'l2', 'l1'),
(dpp.HeatmapObjectCountingModel(), 'l1', 'sigmoid cross entropy')])
def test_set_loss_function(model, bad_loss, good_loss):
with pytest.raises(TypeError):
model.set_loss_function(0)
with pytest.raises(ValueError):
model.set_loss_function(bad_loss)
model.set_loss_function(good_loss)
def test_set_yolo_parameters():
model = dpp.ObjectDetectionModel()
with pytest.raises(RuntimeError):
model.set_yolo_parameters()
model.set_image_dimensions(448, 448, 3)
model.set_yolo_parameters()