def main():
# dataset_str = 'rgb'
dataset_str = 'smooth'
dataset = MIT67Dataset(dataset_str)
model = load_model('models/vgg16_hybrid_1365_softmax_mit67_' + dataset_str + '.h5')
# first dense layer
layer_idx = 19
# choose image (indexed by unshuffled test set)
img_idx = 0
X,Y = dataset.test_data((224,224), 'rgb', 1)
img = X[img_idx]
# directory to save images to
dir_name = 'mit67_' + dataset_str + '_saliency/'
# save unaltered image for comparison
imsave(dir_name + 'smooth_' + str(img_idx) + '.png', img)
# try seveal filters at this layer
nb_filters = 64
for i in range(nb_filters):
print(i)
saliency = visualize_saliency(model, layer_idx, i, img)
# name format: img_index+layer_index+filter_index
imsave((dir_name + 'sal_' + str(img_idx) +
'_' + str(layer_idx) + '_' + str(i) + '.png'),
saliency
)
def main(): # choose datasets # the first listed will be used for training, # all others will be tested on # dataset_strs = ['line_drawings', 'dR_symmetric', 'dR_asymmetric'] dataset_strs = ['smooth'] # dataset_strs = ['smooth', 'smooth_dR_symmetric', 'smooth_dR_asymmetric'] # dataset_strs = ['rgb'] datasets = [MIT67Dataset(s) for s in dataset_strs] # check for optional preprocessing arg train_and_test(datasets)
def main():
dataset_strs = ['smooth']
datasets = [MIT67Dataset(s) for s in dataset_strs]
# choose which vgg16 model to use
model_str = 'vgg16_hybrid_1365'
# model_str = 'vgg16_hybrid_1365_stride'
# model_str = 'places365_vgg11_runaway_weights.h5'
# changes with which model we're using (number of channels)
# color_mode = 'grayscale'
color_mode = 'rgb'
train_and_test(datasets, model_str, color_mode)
def main():
# CHOOSE DATSETS HERE
# this should be a list of strings of the form
# ['train_dataset', 'test_dataset_1', 'test_dataset_2', ...]
# A model will be trained on the first dataset
# (`train_dataset` here),
# the trained model is then tested on each included `test_dataset`
# the model will by default always be tested at least on the `train_dataset`
# dataset_strs = ['rgb']
dataset_strs = ['smooth']
# convert from strings to Dataset objects
datasets = [MIT67Dataset(s) for s in dataset_strs]
model = train_and_test(datasets)
# save model for later
model.save('models/vgg16_hybrid_1365_mit67_' + dataset_strs[0] + '.h5')
def main():
# CHOOSE 3 datasets
# dataset_strs = ['smooth', 'ribbon', 'taper']
# dataset_strs = ['smooth', 'smooth', 'smooth']
# dataset_strs = ['smooth', 'ribbon', 'ribbon']
# dataset_strs = ['smooth', 'taper', 'taper']
# dataset_strs = ['smooth', 'smooth', 'ribbon']
# dataset_strs = ['ribbon', 'ribbon', 'smooth']
# dataset_strs = ['smooth', 'separation', 'separation']
# dataset_strs = ['smooth', 'ribbon', 'separation']
# dataset_strs = ['smooth', 'taper', 'separation']
dataset_strs = ['ribbon', 'taper', 'separation']
print('using datasets: ')
print(dataset_strs)
datasets = [MIT67Dataset(s) for s in dataset_strs]
train_and_test(datasets)
def main():
# choose 3 datasets
# dataset_strs = ['ribbon', 'ribbon', 'ribbon']
# dataset_strs = ['taper', 'taper', 'taper']
# dataset_strs = ['smooth', 'ribbon', 'ribbon']
# dataset_strs = ['smooth', 'smooth', 'smooth']
# dataset_strs = ['smooth', 'taper', 'taper']
# dataset_strs = ['smooth', 'ribbon', 'ribbon']
dataset_strs = ['smooth', 'ribbon', 'taper']
print('using datasets: ')
print(dataset_strs)
datasets = [MIT67Dataset(s) for s in dataset_strs]
# choose which vgg16 model to use
model_str = 'vgg16_hybrid_1365'
# model_str = 'vgg16_hybrid_1365_stride'
train_and_test(datasets, model_str)
def main():
dataset_str = 'rgb'
dataset = MIT67Dataset(dataset_str)
img_size = (224, 224)
color_mode = 'rgb'
x_train, y_train = dataset.train_data(img_size, color_mode, 1)
x_test, y_test = dataset.test_data(img_size, color_mode, 1)
# First Trial -- Import UNMODIFIED VGG16
model = vgg16_hybrid_1365(1)
train_and_test(model, x_train, y_train, x_test, y_test)
# clear memory
K.clear_session()
# Second Trial -- Replace Pooling layers with larger stride
model = vgg16_hybrid_1365_stride(1)
train_and_test(model, x_train, y_train, x_test, y_test)
def bottleneck_features():
model = vgg16_hybrid_1365(1)
img_size = (224, 224)
color_mode = 'rgb'
# Load RGB dataset
rgb_dataset = MIT67Dataset('rgb')
x_rgb_train, y_rgb_train = rgb_dataset.train_data(img_size, color_mode, 1)
x_rgb_test, y_rgb_test = rgb_dataset.test_data(img_size, color_mode, 1)
rgb_train_labels = np.asarray([np.argmax(y) for y in y_rgb_train])
rgb_test_labels = np.asarray([np.argmax(y) for y in y_rgb_test])
color_mode = 'grayscale'
# load weighted line drawings
intact_dataset = MIT67Dataset('smooth')
x1_train, y1_train = intact_dataset.train_data(img_size, color_mode, 1)
x1_test, y1_test = intact_dataset.test_data(img_size, color_mode, 1)
dR_dataset = MIT67Dataset('dR_weighted')
x2_train, y2_train = dR_dataset.train_data(img_size, color_mode, 1)
x2_test, y2_test = dR_dataset.test_data(img_size, color_mode, 1)
d2R_dataset = MIT67Dataset('d2R_weighted')
x3_train, y3_train = d2R_dataset.train_data(img_size, color_mode, 1)
x3_test, y3_test = d2R_dataset.test_data(img_size, color_mode, 1)
x_train = np.ndarray(shape=(x1_train.shape[0:3] + (3, )))
x_test = np.ndarray(shape=(x1_test.shape[0:3] + (3, )))
# put each dataset in respective channels
x_train[:, :, :, 0] = x1_train.squeeze()
x_train[:, :, :, 1] = x2_train.squeeze()
x_train[:, :, :, 2] = x3_train.squeeze()
x_test[:, :, :, 0] = x1_test.squeeze()
x_test[:, :, :, 1] = x2_test.squeeze()
x_test[:, :, :, 2] = x3_test.squeeze()
x_train_labels = np.asarray([np.argmax(y) for y in y1_train])
x_test_labels = np.asarray([np.argmax(y) for y in y1_test])
# get rgb bottleneck features
rgb_bneck_train = model.predict(x_rgb_train)
rgb_bneck_test = model.predict(x_rgb_test)
# other bneck features
x_bneck_train = model.predict(x_train)
x_bneck_test = model.predict(x_test)
# concatenate features
bneck_train = np.concatenate((rgb_bneck_train, x_bneck_train), axis=0)
bneck_test = np.concatenate((rgb_bneck_test, x_bneck_test), axis=0)
train_labels = np.concatenate((rgb_train_labels, x_train_labels), axis=0)
test_labels = np.concatenate((rgb_test_labels, x_test_labels), axis=0)
np.save('bneck_train', bneck_train)
np.save('bneck_test', bneck_test)
np.save('train_labels', train_labels)
np.save('test_labels', test_labels)
print('done')
def main():
dataset_str = 'smooth'
datasets = [MIT67Dataset(dataset_str)]
train_and_test(datasets)
def main():
dataset_str = 'smooth'
dataset = MIT67Dataset(dataset_str)
train_and_visualize(dataset)