AntoineH/smart_augmentation
1
0
Fork 0
mirror of https://github.com/AntoineHX/smart_augmentation.git synced 2025-05-04 20:20:46 +02:00
Code Issues Projects Releases Packages Wiki Activity

Fix etat Train/Eval pour augmentation differee (Retester !)

Browse source
This commit is contained in:
Harle, Antoine (Contracteur) 2020-01-20 17:09:31 -05:00
parent 2d6d2f7397
commit d21a6bbf5c
4 changed files with 38 additions and 29 deletions
Download patch file Download diff file Expand all files Collapse all files

22
higher/datasets.py
View file

@ -7,31 +7,31 @@ TEST_SIZE = 300
#TEST_SIZE = 10000 #legerement +Rapide / + Consomation memoire ! #TEST_SIZE = 10000 #legerement +Rapide / + Consomation memoire !
download_data=False download_data=False
num_workers=4 #4 num_workers=2 #4
pin_memory=False #True :+ GPU memory / + Lent pin_memory=False #True :+ GPU memory / + Lent
#ATTENTION : Dataug (Kornia) Expect image in the range of [0, 1] #ATTENTION : Dataug (Kornia) Expect image in the range of [0, 1]
#transform_train = torchvision.transforms.Compose([ #transform_train = torchvision.transforms.Compose([
# torchvision.transforms.RandomHorizontalFlip(), # torchvision.transforms.RandomHorizontalFlip(),
# torchvision.transforms.ToTensor(), # torchvision.transforms.ToTensor(),
#torchvision.transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)), #CIFAR10 # torchvision.transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)), #CIFAR10
#]) #])
transform = torchvision.transforms.Compose([ transform = torchvision.transforms.Compose([
torchvision.transforms.ToTensor(), torchvision.transforms.ToTensor(),
#torchvision.transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)), #CIFAR10 #torchvision.transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)), #CIFAR10
]) ])
'''
data_train = torchvision.datasets.MNIST( #data_train = torchvision.datasets.MNIST(
"./data", train=True, download=True, # "./data", train=True, download=True,
transform=torchvision.transforms.Compose([ # transform=torchvision.transforms.Compose([
#torchvision.transforms.RandomAffine(degrees=180, translate=None, scale=None, shear=None, resample=False, fillcolor=0), # #torchvision.transforms.RandomAffine(degrees=180, translate=None, scale=None, shear=None, resample=False, fillcolor=0),
torchvision.transforms.ToTensor() # torchvision.transforms.ToTensor()
]) # ])
) #)
data_test = torchvision.datasets.MNIST( data_test = torchvision.datasets.MNIST(
"./data", train=False, download=True, transform=torchvision.transforms.ToTensor() "./data", train=False, download=True, transform=torchvision.transforms.ToTensor()
) )
'''
from torchvision.datasets.vision import VisionDataset from torchvision.datasets.vision import VisionDataset
from PIL import Image from PIL import Image

26
higher/dataug.py
View file

@ -747,21 +747,22 @@ class Data_augV5(nn.Module): #Optimisation jointe (mag, proba)
max_mag_reg = reg_factor * F.mse_loss(mags, target=self._reg_tgt.to(mags.device), reduction='mean') max_mag_reg = reg_factor * F.mse_loss(mags, target=self._reg_tgt.to(mags.device), reduction='mean')
return max_mag_reg return max_mag_reg
def train(self, mode=None): def train(self, mode=True):
""" Set the module training mode. """ Set the module training mode.
Args: Args:
mode (bool): Wether to learn the parameter of the module. None would not change mode. (default: None) mode (bool): Wether to learn the parameter of the module. None would not change mode. (default: None)
""" """
if mode is None : #if mode is None :
mode=self._data_augmentation # mode=self._data_augmentation
self.augment(mode=mode) #Inutile si mode=None self.augment(mode=mode) #Inutile si mode=None
super(Data_augV5, self).train(mode) super(Data_augV5, self).train(mode)
return self
def eval(self): def eval(self):
""" Set the module to evaluation mode. """ Set the module to evaluation mode.
""" """
self.train(mode=False) return self.train(mode=False)
def augment(self, mode=True): def augment(self, mode=True):
""" Set the augmentation mode. """ Set the augmentation mode.
@ -1266,7 +1267,7 @@ class Augmented_model(nn.Module):
Attributes: Attributes:
_mods (nn.ModuleDict): A dictionary containing the modules. _mods (nn.ModuleDict): A dictionary containing the modules.
_data_augmentation (bool): Wether data augmentation is used. _data_augmentation (bool): Wether data augmentation should be used.
""" """
def __init__(self, data_augmenter, model): def __init__(self, data_augmenter, model):
"""Init Augmented Model. """Init Augmented Model.
@ -1308,22 +1309,25 @@ class Augmented_model(nn.Module):
self._data_augmentation=mode self._data_augmentation=mode
self._mods['data_aug'].augment(mode) self._mods['data_aug'].augment(mode)
def train(self, mode=None): def train(self, mode=True):
""" Set the module training mode. """ Set the module training mode.
Args: Args:
mode (bool): Wether to learn the parameter of the module. None would not change mode. (default: None) mode (bool): Wether to learn the parameter of the module. (default: None)
""" """
if mode is None : #if mode is None :
mode=self._data_augmentation # mode=self._data_augmentation
self._mods['data_aug'].augment(mode)
super(Augmented_model, self).train(mode) super(Augmented_model, self).train(mode)
self._mods['data_aug'].augment(mode=self._data_augmentation) #Restart if needed data augmentation
return self return self
def eval(self): def eval(self):
""" Set the module to evaluation mode. """ Set the module to evaluation mode.
""" """
return self.train(mode=False) #return self.train(mode=False)
super(Augmented_model, self).train(mode=False)
self._mods['data_aug'].augment(mode=False)
return self
def items(self): def items(self):
"""Return an iterable of the ModuleDict key/value pairs. """Return an iterable of the ModuleDict key/value pairs.

5
higher/test_dataug.py
View file

@ -68,7 +68,7 @@ if __name__ == "__main__":
} }
n_inner_iter = 1 n_inner_iter = 1
epochs = 150 epochs = 150
dataug_epoch_start=0 dataug_epoch_start=10
optim_param={ optim_param={
'Meta':{ 'Meta':{
'optim':'Adam', 'optim':'Adam',
@ -87,8 +87,6 @@ if __name__ == "__main__":
#model = MobileNetV2(num_classes=10) #model = MobileNetV2(num_classes=10)
#model = WideResNet(num_classes=10, wrn_size=32) #model = WideResNet(num_classes=10, wrn_size=32)
model = Higher_model(model) #run_dist_dataugV3
#### Classic #### #### Classic ####
if 'classic' in tasks: if 'classic' in tasks:
t0 = time.process_time() t0 = time.process_time()
@ -171,6 +169,7 @@ if __name__ == "__main__":
t0 = time.process_time() t0 = time.process_time()
tf_dict = {k: TF.TF_dict[k] for k in tf_names} tf_dict = {k: TF.TF_dict[k] for k in tf_names}
model = Higher_model(model) #run_dist_dataugV3
aug_model = Augmented_model(Data_augV5(TF_dict=tf_dict, N_TF=2, mix_dist=0.8, fixed_prob=False, fixed_mag=False, shared_mag=False), model).to(device) aug_model = Augmented_model(Data_augV5(TF_dict=tf_dict, N_TF=2, mix_dist=0.8, fixed_prob=False, fixed_mag=False, shared_mag=False), model).to(device)
#aug_model = Augmented_model(RandAug(TF_dict=tf_dict, N_TF=2), model).to(device) #aug_model = Augmented_model(RandAug(TF_dict=tf_dict, N_TF=2), model).to(device)

14
higher/train_utils.py
View file

@ -827,6 +827,7 @@ def run_dist_dataugV3(model, opt_param, epochs=1, inner_it=0, dataug_epoch_start
device = next(model.parameters()).device device = next(model.parameters()).device
log = [] log = []
dl_val_it = iter(dl_val) dl_val_it = iter(dl_val)
val_loss=None
high_grad_track = True high_grad_track = True
if inner_it == 0: #No HP optimization if inner_it == 0: #No HP optimization
@ -909,10 +910,8 @@ def run_dist_dataugV3(model, opt_param, epochs=1, inner_it=0, dataug_epoch_start
if(high_grad_track and i>0 and i%inner_it==0): #Perform Meta step if(high_grad_track and i>0 and i%inner_it==0): #Perform Meta step
#print("meta") #print("meta")
val_loss = compute_vaLoss(model=model, dl_it=dl_val_it, dl=dl_val) + model['data_aug'].reg_loss() val_loss = compute_vaLoss(model=model, dl_it=dl_val_it, dl=dl_val) + model['data_aug'].reg_loss()
#print_graph(val_loss) #to visualize computational graph #print_graph(val_loss) #to visualize computational graph
val_loss.backward() val_loss.backward()
torch.nn.utils.clip_grad_norm_(model['data_aug'].parameters(), max_norm=10, norm_type=2) #Prevent exploding grad with RNN torch.nn.utils.clip_grad_norm_(model['data_aug'].parameters(), max_norm=10, norm_type=2) #Prevent exploding grad with RNN
@ -920,7 +919,7 @@ def run_dist_dataugV3(model, opt_param, epochs=1, inner_it=0, dataug_epoch_start
meta_opt.step() meta_opt.step()
#Adjust Hyper-parameters #Adjust Hyper-parameters
model['data_aug'].adjust_param(soft=True) #Contrainte sum(proba)=1 model['data_aug'].adjust_param(soft=False) #Contrainte sum(proba)=1
if hp_opt: if hp_opt:
for param_group in diffopt.param_groups: for param_group in diffopt.param_groups:
for param in list(opt_param['Inner'].keys())[1:]: for param in list(opt_param['Inner'].keys())[1:]:
@ -949,6 +948,7 @@ def run_dist_dataugV3(model, opt_param, epochs=1, inner_it=0, dataug_epoch_start
accuracy, test_loss =test(model) accuracy, test_loss =test(model)
model.train() model.train()
print(model['data_aug']._data_augmentation)
#### Log #### #### Log ####
param = [{'p': p.item(), 'm':model['data_aug']['mag'].item()} for p in model['data_aug']['prob']] if model['data_aug']._shared_mag else [{'p': p.item(), 'm': m.item()} for p, m in zip(model['data_aug']['prob'], model['data_aug']['mag'])] param = [{'p': p.item(), 'm':model['data_aug']['mag'].item()} for p in model['data_aug']['prob']] if model['data_aug']._shared_mag else [{'p': p.item(), 'm': m.item()} for p, m in zip(model['data_aug']['prob'], model['data_aug']['mag'])]
data={ data={
@ -989,7 +989,13 @@ def run_dist_dataugV3(model, opt_param, epochs=1, inner_it=0, dataug_epoch_start
print('Starting Data Augmention...') print('Starting Data Augmention...')
dataug_epoch_start = epoch dataug_epoch_start = epoch
model.augment(mode=True) model.augment(mode=True)
if inner_it != 0: high_grad_track = True if inner_it != 0: #Rebuild diffopt if needed
high_grad_track = True
diffopt = model['model'].get_diffopt(
inner_opt,
grad_callback=(lambda grads: clip_norm(grads, max_norm=10)),
track_higher_grads=high_grad_track)
#Data sample saving #Data sample saving
try: try: