def run(project_dir, gpu_mon, logger, args):
"""
Runs training of a model in a mpunet project directory.
Args:
project_dir: A path to a mpunet project
gpu_mon: An initialized GPUMonitor object
logger: A mpunet logging object
args: argparse arguments
"""
# Read in hyperparameters from YAML file
from mpunet.hyperparameters import YAMLHParams
hparams = YAMLHParams(project_dir + "/train_hparams.yaml", logger=logger)
validate_hparams(hparams)
# Wait for PID to terminate before continuing?
if args.wait_for:
from mpunet.utils import await_PIDs
await_PIDs(args.wait_for)
# Prepare sequence generators and potential model specific hparam changes
train, val, hparams = get_data_sequences(project_dir=project_dir,
hparams=hparams,
logger=logger,
args=args)
# Set GPU visibility and create model with MirroredStrategy
set_gpu(gpu_mon, args)
import tensorflow as tf
with tf.distribute.MirroredStrategy().scope():
model = get_model(project_dir=project_dir,
train_seq=train,
hparams=hparams,
logger=logger,
args=args)
# Get trainer and compile model
from mpunet.train import Trainer
trainer = Trainer(model, logger=logger)
trainer.compile_model(n_classes=hparams["build"].get("n_classes"),
reduction=tf.keras.losses.Reduction.NONE,
**hparams["fit"])
# Debug mode?
if args.debug:
from tensorflow.python import debug as tfdbg
from tensorflow.keras import backend as K
K.set_session(tfdbg.LocalCLIDebugWrapperSession(K.get_session()))
# Fit the model
_ = trainer.fit(train=train,
val=val,
train_im_per_epoch=args.train_images_per_epoch,
val_im_per_epoch=args.val_images_per_epoch,
hparams=hparams,
no_im=args.no_images,
**hparams["fit"])
save_final_weights(model, project_dir, logger)
def entry_func(args=None):
# Get parser
parser = vars(get_parser().parse_args(args))
# Get parser arguments
cv_dir = os.path.abspath(parser["CV_dir"])
out_dir = os.path.abspath(parser["out_dir"])
create_folders(out_dir)
await_PID = parser["wait_for"]
run_split = parser["run_on_split"]
start_from = parser["start_from"] or 0
num_jobs = parser["num_jobs"] or 1
# GPU settings
num_GPUs = parser["num_GPUs"]
force_GPU = parser["force_GPU"]
ignore_GPU = parser["ignore_GPU"]
monitor_GPUs_every = parser["monitor_GPUs_every"]
# User input assertions
_assert_force_and_ignore_gpus(force_GPU, ignore_GPU)
if run_split:
_assert_run_split(start_from, monitor_GPUs_every, num_jobs)
# Wait for PID?
if await_PID:
from mpunet.utils import await_PIDs
await_PIDs(await_PID)
# Get file paths
script = os.path.abspath(parser["script_prototype"])
hparams = os.path.abspath(parser["hparams_prototype"])
no_hparams = parser["no_hparams"]
# Get list of folders of CV data to run on
cv_folders = get_CV_folders(cv_dir)
if run_split is not None:
if run_split < 0 or run_split >= len(cv_folders):
raise ValueError("--run_on_split should be in range [0-{}], "
"got {}".format(len(cv_folders) - 1, run_split))
cv_folders = [cv_folders[run_split]]
log_appendix = "_split{}".format(run_split)
else:
log_appendix = ""
# Get a logger object
logger = Logger(base_path="./",
active_file="output" + log_appendix,
print_calling_method=False,
overwrite_existing=True)
if force_GPU:
# Only these GPUs fill be chosen from
from mpunet.utils import set_gpu
set_gpu(force_GPU)
if num_GPUs:
# Get GPU sets (up to the number of splits)
gpu_sets = get_free_GPU_sets(num_GPUs, ignore_GPU)[:len(cv_folders)]
elif not num_jobs or num_jobs < 0:
raise ValueError("Should specify a number of jobs to run in parallel "
"with the --num_jobs flag when using 0 GPUs pr. "
"process (--num_GPUs=0 was set).")
else:
gpu_sets = ["''"] * parser["num_jobs"]
# Get process pool, lock and GPU queue objects
lock = Lock()
gpu_queue = Queue()
for gpu in gpu_sets:
gpu_queue.put(gpu)
procs = []
if monitor_GPUs_every is not None and monitor_GPUs_every:
logger("\nOBS: Monitoring GPU pool every %i seconds\n" %
monitor_GPUs_every)
# Start a process monitoring new GPU availability over time
stop_event = Event()
t = Process(target=monitor_GPUs,
args=(monitor_GPUs_every, gpu_queue, num_GPUs, ignore_GPU,
gpu_sets, stop_event))
t.start()
procs.append(t)
else:
stop_event = None
try:
for cv_folder in cv_folders[start_from:]:
gpus = gpu_queue.get()
t = Process(target=run_sub_experiment,
args=(cv_folder, out_dir, script, hparams, no_hparams,
gpus, gpu_queue, lock, logger))
t.start()
procs.append(t)
for t in procs:
if not t.is_alive():
t.join()
except KeyboardInterrupt:
for t in procs:
t.terminate()
if stop_event is not None:
stop_event.set()
for t in procs:
t.join()
def entry_func(args=None):
# Project base path
args = vars(get_argparser().parse_args(args))
basedir = os.path.abspath(args["project_dir"])
overwrite = args["overwrite"]
continue_training = args["continue_training"]
eval_prob = args["eval_prob"]
await_PID = args["wait_for"]
dice_weight = args["dice_weight"]
print("Fitting fusion model for project-folder: %s" % basedir)
# Minimum images in validation set before also using training images
min_val_images = 15
# Fusion model training params
epochs = args['epochs']
fm_batch_size = args["batch_size"]
# Early stopping params
early_stopping = args["early_stopping"]
# Wait for PID?
if await_PID:
from mpunet.utils import await_PIDs
await_PIDs(await_PID)
# Fetch GPU(s)
num_GPUs = args["num_GPUs"]
force_gpu = args["force_GPU"]
# Wait for free GPU
if not force_gpu:
await_and_set_free_gpu(N=num_GPUs, sleep_seconds=120)
else:
set_gpu(force_gpu)
# Get logger
logger = Logger(base_path=basedir,
active_file="train_fusion",
overwrite_existing=overwrite)
# Get YAML hyperparameters
hparams = YAMLHParams(os.path.join(basedir, "train_hparams.yaml"))
# Get some key settings
n_classes = hparams["build"]["n_classes"]
if hparams["build"]["out_activation"] == "linear":
# Trained with logit targets?
hparams["build"][
"out_activation"] = "softmax" if n_classes > 1 else "sigmoid"
# Get views
views = np.load("%s/views.npz" % basedir)["arr_0"]
del hparams["fit"]["views"]
# Get weights and set fusion (output) path
weights = get_best_model("%s/model" % basedir)
weights_name = os.path.splitext(os.path.split(weights)[-1])[0]
fusion_weights = "%s/model/fusion_weights/" \
"%s_fusion_weights.h5" % (basedir, weights_name)
create_folders(os.path.split(fusion_weights)[0])
# Log a few things
log(logger, hparams, views, weights, fusion_weights)
# Check if exists already...
if not overwrite and os.path.exists(fusion_weights):
from sys import exit
print("\n[*] A fusion weights file already exists at '%s'."
"\n Use the --overwrite flag to overwrite." % fusion_weights)
exit(0)
# Load validation data
images = ImagePairLoader(**hparams["val_data"], logger=logger)
is_validation = {m.identifier: True for m in images}
# Define random sets of images to train on simul. (cant be all due
# to memory constraints)
image_IDs = [m.identifier for m in images]
if len(images) < min_val_images:
# Pick N random training images
diff = min_val_images - len(images)
logger("Adding %i training images to set" % diff)
# Load the training data and pick diff images
train = ImagePairLoader(**hparams["train_data"], logger=logger)
indx = np.random.choice(np.arange(len(train)),
diff,
replace=diff > len(train))
# Add the images to the image set set
train_add = [train[i] for i in indx]
for m in train_add:
is_validation[m.identifier] = False
image_IDs.append(m.identifier)
images.add_images(train_add)
# Append to length % sub_size == 0
sub_size = args["images_per_round"]
rest = int(sub_size * np.ceil(len(image_IDs) / sub_size)) - len(image_IDs)
if rest:
image_IDs += list(np.random.choice(image_IDs, rest, replace=False))
# Shuffle and split
random.shuffle(image_IDs)
sets = [
set(s) for s in np.array_split(image_IDs,
len(image_IDs) / sub_size)
]
assert (contains_all_images(sets, image_IDs))
# Define fusion model (named 'org' to store reference to orgiginal model if
# multi gpu model is created below)
fusion_model = FusionModel(n_inputs=len(views),
n_classes=n_classes,
weight=dice_weight,
logger=logger,
verbose=False)
if continue_training:
fusion_model.load_weights(fusion_weights)
print("\n[OBS] CONTINUED TRAINING FROM:\n", fusion_weights)
import tensorflow as tf
with tf.distribute.MirroredStrategy().scope():
# Define model
unet = init_model(hparams["build"], logger)
print("\n[*] Loading weights: %s\n" % weights)
unet.load_weights(weights, by_name=True)
# Compile the model
logger("Compiling...")
metrics = [
"sparse_categorical_accuracy", sparse_fg_precision, sparse_fg_recall
]
fusion_model.compile(optimizer=Adam(lr=1e-3),
loss=fusion_model.loss,
metrics=metrics)
fusion_model._log()
try:
_run_fusion_training(sets, logger, hparams, min_val_images,
is_validation, views, n_classes, unet,
fusion_model, early_stopping, fm_batch_size,
epochs, eval_prob, fusion_weights)
except KeyboardInterrupt:
pass
finally:
if not os.path.exists(os.path.split(fusion_weights)[0]):
os.mkdir(os.path.split(fusion_weights)[0])
# Save fusion model weights
# OBS: Must be original model if multi-gpu is performed!
fusion_model.save_weights(fusion_weights)
def run(args):
cv_dir = os.path.abspath(args.CV_dir)
# Get list of folders of CV data to run on
cv_folders = get_CV_folders(cv_dir)
assert_args(args, n_splits=len(cv_folders))
out_dir = os.path.abspath(args.out_dir)
hparams_dir = os.path.abspath(args.hparams_prototype_dir)
prepare_hparams_dir(hparams_dir)
create_folders(out_dir)
# Wait for PID?
if args.wait_for:
from mpunet.utils import await_PIDs
await_PIDs(args.wait_for)
if args.run_on_split is not None:
cv_folders = [cv_folders[args.run_on_split]]
log_appendix = "_split{}".format(args.run_on_split)
else:
log_appendix = ""
# Get a logger object
logger = Logger(base_path="./",
active_file="output" + log_appendix,
print_calling_method=False,
overwrite_existing=True)
if args.force_GPU:
# Only these GPUs fill be chosen from
from mpunet.utils import set_gpu
set_gpu(args.force_GPU)
if args.num_GPUs:
# Get GPU sets (up to the number of splits)
gpu_sets = get_free_GPU_sets(args.num_GPUs,
args.ignore_GPU)[:len(cv_folders)]
elif not args.num_jobs or args.num_jobs < 0:
raise ValueError("Should specify a number of jobs to run in parallel "
"with the --num_jobs flag when using 0 GPUs pr. "
"process (--num_GPUs=0 was set).")
else:
gpu_sets = ["''"] * args.num_jobs
# Get process pool, lock and GPU queue objects
lock = Lock()
gpu_queue = Queue()
for gpu in gpu_sets:
gpu_queue.put(gpu)
# Get file paths
script = os.path.abspath(args.script_prototype)
# Get GPU monitor process
running_processes, stop_event = start_gpu_monitor_process(
args, gpu_queue, gpu_sets, logger)
try:
for cv_folder in cv_folders[args.start_from:]:
gpus = gpu_queue.get()
t = Process(target=run_sub_experiment,
args=(cv_folder, out_dir, script, hparams_dir,
args.no_hparams, gpus, gpu_queue, lock, logger))
t.start()
running_processes.append(t)
for t in running_processes:
if not t.is_alive():
t.join()
except KeyboardInterrupt:
for t in running_processes:
t.terminate()
if stop_event is not None:
stop_event.set()
for t in running_processes:
t.join()