def __init__(self, datasetName):
super(GroceryParameters,self).__init__(datasetName)
self.classes = ('__background__', # always index 0
'avocado', 'orange', 'butter', 'champagne', 'eggBox', 'gerkin', 'joghurt', 'ketchup',
'orangeJuice', 'onion', 'pepper', 'tomato', 'water', 'milk', 'tabasco', 'mustard')
# roi generation
self.roi_minDimRel = 0.04
self.roi_maxDimRel = 0.4
self.roi_minNrPixelsRel = 2 * self.roi_minDimRel * self.roi_minDimRel
self.roi_maxNrPixelsRel = 0.33 * self.roi_maxDimRel * self.roi_maxDimRel
# model training / scoring
self.classifier = 'nn'
self.cntk_num_train_images = 25
self.cntk_num_test_images = 5
self.cntk_mb_size = 5
self.cntk_max_epochs = 20
self.cntk_momentum_per_sample = 0.8187307530779818
# postprocessing
self.nmsThreshold = 0.01
# database
self.imdbs = dict() # database provider of images and image annotations
for image_set in ["train", "test"]:
self.imdbs[image_set] = imdb_data(image_set, self.classes, self.cntk_nrRois, self.imgDir, self.roiDir, self.cntkFilesDir, boAddGroundTruthRois=(image_set!='test'))
def __init__(self, datasetName):
super(GroceryParameters,self).__init__(datasetName)
self.classes = ('__background__', # always index 0
'avocado', 'orange', 'butter', 'champagne', 'eggBox', 'gerkin', 'joghurt', 'ketchup',
'orangeJuice', 'onion', 'pepper', 'tomato', 'water', 'milk', 'tabasco', 'mustard')
# roi generation
self.roi_minDimRel = 0.04
self.roi_maxDimRel = 0.4
self.roi_minNrPixelsRel = 2 * self.roi_minDimRel * self.roi_minDimRel
self.roi_maxNrPixelsRel = 0.33 * self.roi_maxDimRel * self.roi_maxDimRel
# model training / scoring
self.classifier = 'nn'
self.cntk_num_train_images = 25
self.cntk_num_test_images = 5
self.cntk_mb_size = 5
self.cntk_max_epochs = 20
self.cntk_momentum_time_constant = 10
# postprocessing
self.nmsThreshold = 0.01
# database
self.imdbs = dict() # database provider of images and image annotations
for image_set in ["train", "test"]:
self.imdbs[image_set] = imdb_data(image_set, self.classes, self.cntk_nrRois, self.imgDir, self.roiDir, self.cntkFilesDir, boAddGroundTruthRois=(image_set!='test'))
saveCompressed=True,
skipCheck=True)
classes = (
'__background__', # always index 0
'drone',
'dummy')
datasetName = "Drones"
imgDir = "/home/slapbot/my_side_projects/drone-detection/DataSets/Drones/"
roiDir = "/home/slapbot/my_side_projects/drone-detection/Detection/FastRCNN/proc/Drones_500/rois/"
imdbs = dict() # database provider of images and image annotations
for image_set in ["train", "test"]:
imdbs[image_set] = imdb_data(image_set,
classes,
cntk_nrRois,
imgDir,
roiDir,
cntkFilesDir,
boAddGroundTruthRois=(image_set != 'test'))
imdb = imdbs[image_set]
net = DummyNet(4096, imdb.num_classes, outParsedDir)
evalTempDir = None
classifier = "nn"
nmsThreshold = 0.01
test_net(net,
imdb,
evalTempDir,
None,
# model training / scoring
classifier = 'nn'
cntk_num_train_images = 25
cntk_num_test_images = 5
cntk_mb_size = 5
cntk_max_epochs = 20
cntk_momentum_time_constant = 10
# postprocessing
nmsThreshold = 0.01
# database
imdbs = dict() # database provider of images and image annotations
for image_set in ["train", "test"]:
imdbs[image_set] = imdb_data(image_set, classes, cntk_nrRois, imgDir, roiDir, cntkFilesDir, boAddGroundTruthRois = (image_set!='test'))
elif datasetName.startswith("pascalVoc"):
imgDir = pascalDataDir
if datasetName.startswith("pascalVoc_aeroplanesOnly"):
classes = ('__background__', 'aeroplane')
lutImageSet = {"train": "trainval.aeroplaneOnly", "test": "test.aeroplaneOnly"}
else:
classes = ('__background__', # always index 0
'aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable',
'dog', 'horse', 'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor')
lutImageSet = {"train": "trainval", "test": "test"}
# use cntk_nrRois = 4000. more than 99% of the test images have less than 4000 rois, but 50% more than 2000
# model training / scoring