def main():
direc_data = '/data/npz_data/cells/unspecified_nuclear_data/nuclear_movie/'
dataset = 'nuclear_movie_same'
training_data = np.load('{}{}.npz'.format(direc_data, dataset))
optimizer = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=0.01, decay=0.99)
in_shape = (14, 14, 1)
model = the_model(input_shape=in_shape) #, n_features=1, reg=1e-5)
train_model_siamese(
model=model,
dataset='nuclear_movie_same',
optimizer=optimizer,
expt='',
it=0,
batch_size=1,
n_epoch=100,
direc_save='/data/models/cells/unspecified_nuclear_data/nuclear_movie',
direc_data=
'/data/npz_data/cells/unspecified_nuclear_data/nuclear_movie/',
lr_sched=lr_sched,
rotation_range=0,
flip=True,
shear=0,
class_weight=None)
# Create output ditrectory, if necessary
pathlib.Path(direc_save).mkdir(parents=True, exist_ok=True)
optimizer = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=0.01, decay=0.99)
file_name = os.path.join(direc_data, dataset + ".npz")
training_data = np.load(file_name)
class_weights = training_data["class_weights"]
for iterate in range(1):
model = the_model(batch_shape=(1, 512, 512, 1),
n_features=3,
reg=1e-5,
softmax=True,
permute=True)
train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
lr_sched=lr_sched,
class_weight=class_weights,
rotation_range=180,
n_epoch = 25
data_format = "channels_first"
dataset = "cytoplasm_61x61"
expt = "bn_feature_net_61x61"
direc_save = "/home/davince/Dropbox (OIST)/deepcell-tf-master/trained_networks/20180330_cytoplasm_raw/"
direc_data = "/home/davince/Dropbox (OIST)/deepcell-tf-master/training_data_npz/20180401_newdata_Raw/"
optimizer = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=0.01, decay=0.99)
class_weights = {0: 1, 1: 1, 2: 1}
for iterate in xrange(3):
model = the_model(n_channels=1, n_features=3, reg=1e-5)
train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
lr_sched=lr_sched,
class_weight=class_weights,
rotation_range=180,
flip=True,
shear=False,
direc_data = "/data/training_data_npz/nuclear_movie/"
# Create output ditrectory, if necessary
pathlib.Path(direc_save).mkdir(parents=True, exist_ok=True)
optimizer = SGD(lr=1e-2, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=1e-2, decay=0.99)
file_name = os.path.join(direc_data, dataset + ".npz")
training_data = np.load(file_name)
for iterate in range(1):
model = the_model(batch_shape=(1, 1, 5, 256, 256),
n_features=3,
reg=1e-5,
location=False,
permute=True,
softmax=False)
trained_model = train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
number_of_frames=5,
lr_sched=lr_sched,
rotation_range=180,
direc_save = "/data/trained_networks/nuclei/"
direc_data = "/data/training_data_npz/nuclei/"
optimizer = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=0.01, decay=0.99)
file_name = os.path.join(direc_data, dataset + ".npz")
training_data = np.load(file_name)
class_weights = training_data["class_weights"]
for iterate in xrange(1):
model = the_model(input_shape=(1, 512, 512),
n_features=3,
reg=1e-5,
softmax=True,
permute=True)
train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
lr_sched=lr_sched,
class_weight=class_weights,
rotation_range=180,
batch_size = 1 n_epoch = 200 dataset = "nuclei_conv_61x61" expt = "retina_net" # direc_save = "/data/trained_networks/nuclei/" direc_data= "/data/training_data_npz/nuclei/" optimizer = Adam(lr=1e-5, clipnorm=0.001) # optimizer = SGD(lr = 0.01, momentum = 0.9, nesterov = True) lr_sched = rate_scheduler(lr = 1e-5, decay = 0.99) # file_name = os.path.join(direc_data, dataset + ".npz") training_data = np.load(file_name) for iterate in xrange(1): model = the_model(num_classes = 1, input_shape = (1,512,512)) trained_model = train_model(model = model, dataset = dataset, optimizer = optimizer, expt = expt, it = iterate, batch_size = batch_size, n_epoch = n_epoch, direc_save = direc_save, direc_data = direc_data, lr_sched = lr_sched, rotation_range = 0, flip = False, shear = False)
direc_save = "/data/trained_networks/nuclei_broad/"
direc_data = "/data/training_data_npz/nuclei_broad/"
optimizer = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=0.01, decay=0.99)
file_name = os.path.join(direc_data, dataset + ".npz")
training_data = np.load(file_name)
class_weights = training_data["class_weights"]
for iterate in xrange(1):
model = the_model(batch_shape=(1, 1, 512, 512),
n_features=16,
reg=1e-5,
softmax=False,
location=True,
permute=True)
train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
lr_sched=lr_sched,
class_weight=class_weights,
rotation_range=180,
# Create output ditrectory, if necessary
pathlib.Path(direc_save).mkdir(parents=True, exist_ok=True)
optimizer = SGD(lr=1e-2, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=1e-2, decay=0.99)
file_name = os.path.join(direc_data, dataset + ".npz")
training_data = np.load(file_name)
class_weights = training_data["class_weights"]
print(class_weights)
for iterate in range(1):
model = the_model(input_shape=(512, 512, 2),
n_features=3,
reg=1e-5,
location=False,
permute=False)
trained_model = train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
lr_sched=lr_sched,
class_weight=class_weights,
rotation_range=180,
direc_save = "/data/trained_networks/HeLa/"
direc_data = "/data/training_data_npz/HeLa/"
optimizer = SGD(lr=1e-2, decay=1e-6, momentum=0.9, nesterov=True)
lr_sched = rate_scheduler(lr=1e-2, decay=0.99)
file_name = os.path.join(direc_data, dataset + ".npz")
training_data = np.load(file_name)
class_weights = training_data["class_weights"]
print(class_weights)
for iterate in xrange(1):
model = the_model(input_shape=(2, 512, 512),
n_features=3,
reg=1e-5,
permute=True)
trained_model = train_model(model=model,
dataset=dataset,
optimizer=optimizer,
expt=expt,
it=iterate,
batch_size=batch_size,
n_epoch=n_epoch,
direc_save=direc_save,
direc_data=direc_data,
lr_sched=lr_sched,
class_weight=class_weights,
rotation_range=180,
flip=True,