def new_fresh_model(args, DEVICE):
"""
Reload a new-fresh model.
useful to restart training with certainty that the model is at the same
exact state at t=0.
:param args: object.
:param DEVICE: torch device.
:return:
"""
ALLOW_MULTIGPUS = check_if_allow_multgpu_mode()
NBRGPUS = torch.cuda.device_count()
# ========================= Instantiate models =============================
model = instantiate_models(args)
# Check if we are using a user specific pre-trained model other than our
# pre-defined pre-trained models. This can be used to to EVALUATE a
# trained model. You need to set args.max_epochs to -1 so no training is
# performed. This is a hack to avoid creating other function to deal with
# LATER-evaluation after this code is done. This script is intended for
# training. We evaluate at the end. However, if you missed something
# during the training/evaluation (for example plot something over the
# predicted images), you do not need to re-train the model. You can 1.
# specify the path to the pre-trained model. 2. Set max_epochs to -1. 3.
# Set strict to True. By doing this, we load the pre-trained model, and,
# we skip the training loop, fast-forward to the evaluation.
if args.model['path_pre_trained'] not in [None, 'None']:
warnings.warn("You have asked to load a specific pre-trained "
"model from {} .... [OK]".format(
args.model['path_pre_trained']))
model = load_pre_pretrained_model(
model=model,
path_file=args.model['path_pre_trained'],
strict=args.model['strict'],
freeze_classifier=args.freeze_classifier
)
# Check if there are multiple GPUS.
if ALLOW_MULTIGPUS:
model = MyDataParallel(model)
if args.batch_size < NBRGPUS:
warnings.warn("You asked for MULTIGPU mode. However, "
"your batch size {} is smaller than the number of "
"GPUs available {}. This is fine in practice. "
"However, some GPUs will be idol. "
"This is just a warning .... "
"[OK]".format(args.batch_size, NBRGPUS))
model.to(DEVICE)
# freeze the classifier if needed
if args.freeze_classifier:
warnings.warn("You asked to freeze the classifier."
"We are going to do it right now.")
model.freeze_cl()
assert model.assert_cl_is_frozen(), "Something is wrong"
# ==========================================================================
return model
# thread-safe.
import reproducibility
import constants
ACTIVATE_SYNC_BN = True
# Override ACTIVATE_SYNC_BN using variable environment in Bash:
# $ export ACTIVATE_SYNC_BN="True" ----> Activate
# $ export ACTIVATE_SYNC_BN="False" ----> Deactivate
if "ACTIVATE_SYNC_BN" in os.environ.keys():
ACTIVATE_SYNC_BN = (os.environ['ACTIVATE_SYNC_BN'] == "True")
announce_msg("ACTIVATE_SYNC_BN was set to {}".format(ACTIVATE_SYNC_BN))
if check_if_allow_multgpu_mode() and ACTIVATE_SYNC_BN: # Activate Synch-BN.
from deeplearning.syncbn import nn as NN_Sync_BN
BatchNorm2d = NN_Sync_BN.BatchNorm2d
announce_msg("Synchronized BN has been activated. \n"
"MultiGPU mode has been activated. "
"{} GPUs".format(torch.cuda.device_count()))
else:
BatchNorm2d = nn.BatchNorm2d
if check_if_allow_multgpu_mode():
announce_msg("Synchronized BN has been deactivated.\n"
"MultiGPU mode has been activated. "
"{} GPUs".format(torch.cuda.device_count()))
else:
announce_msg("Synchronized BN has been deactivated.\n"
"MultiGPU mode has been deactivated. "
"{} GPUs".format(torch.cuda.device_count()))
# "Oxford-flowers-102"
DEBUG_MODE = False
# dataset to go fast. If True, we select only few samples for training
# , validation, and test.
FRQ = 2. # number of times to plot the train figures.
# =============================================
# Parse the inputs and deal with the yaml file.
# init. reproducibility.
# =============================================
args, args_dict, input_args = parse_input()
NBRGPUS = torch.cuda.device_count()
ALLOW_MULTIGPUS = check_if_allow_multgpu_mode()
if __name__ == "__main__":
init_time = dt.datetime.now()
# ==================================================
# Device, criteria, folders, output logs, callbacks.
# ==================================================
DEVICE = get_device(args)
CPUDEVICE = get_cpu_device()
CRITERION = instantiate_loss(args).to(DEVICE)
# plot freq.
ds = constants.CAM16
announce_msg("Processing dataset: {}".format(ds))
args = {
"baseurl": get_rootpath_2_dataset(Dict2Obj({'dataset': ds})),
"dataset": ds,
"fold_folder": "folds/{}".format(ds),
"img_extension": "jpg",
"path_encoding": "folds/{}/encoding-origine.yaml".format(ds)
}
# Convert masks into binary masks: already done.
# create_bin_mask_Oxford_flowers_102(Dict2Obj(args))
reproducibility.init_seed()
al_split_camelyon16(Dict2Obj(args))
# get_stats(Dict2Obj(args), split=0, fold=0, subset='train')
if __name__ == "__main__":
check_if_allow_multgpu_mode()
# ==========================================================================
# ACTIVE LEARNING
# ==========================================================================
do_glas()
do_Caltech_UCSD_Birds_200_2011()
# do_Oxford_flowers_102()
# do_camelyon16()
# ==========================================================================
# END: ACTIVE LEARNING
# ==========================================================================