def main():
model_settings = [Baseline2, Baseline1]
BNs = [True, False]
dropouts = [True, False]
augments = [True, False]
idx = 0
for BN in BNs[::-1]:
for aug in augments:
for dropout in dropouts:
for config in model_settings:
print("BN", BN)
print("Augmentation", aug)
print("dropout", dropout)
print("config", config)
settings = config()
settings.BATCH_NORM = BN
settings.DROPOUT = dropout
settings.AUGMENTATION = aug
expdir = "experiments/train/baseline/" + str(idx) + "/"
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS,
test_every=test_every,
num_epochs_cv=num_epochs_cv,
print_every=print_every,
small_debug=SMALL_DEBUG_DOM)
idx += 1
def main():
layer = 6
cardinality = 4
print("Layers", layer)
print("Cardinality", cardinality)
kwargs = {"layers": layer, "cardinality": cardinality}
settings = ResNeXt(**kwargs)
settings.AUGMENTATION = True
settings.DEBUG = False
settings.GPU_DEVICE = GPU_DEVICE
settings.SEED = 19
settings.export_env_vars()
expdir = exp_base
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS,
test_every=test_every,
num_epochs_cv=num_epochs_cv,
learning_rate=LR,
print_every=print_every,
small_debug=SMALL_DEBUG_DOM)
def main():
res_layers = [3, 9, 27]
cardinalities = [1, 8, 32]
idx = 0
for layer in res_layers:
for cardinality in cardinalities:
print("Layers", layer)
print("Cardinality", cardinality)
kwargs = {"layers": layer, "cardinality": cardinality}
settings = ResNeXt(**kwargs)
settings.AUGMENTATION = True
settings.DEBUG = False
expdir = exp_base + str(idx) + "/"
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS,
test_every=test_every,
num_epochs_cv=num_epochs_cv,
learning_rate=LR,
print_every=print_every,
small_debug=SMALL_DEBUG_DOM)
idx += 1
def test_AE_train_3d(self, tmpdir):
"""Test no exception thrown"""
epochs = 1
settings = self.__settings(tmpdir)
expdir = tmpdir.mkdir("experiments/")
trainer = TrainAE(settings, str(expdir))
model = trainer.train(epochs, num_workers=0)
def main():
print("updated")
model_settings = [BaselineBlock, Baseline2, Baseline1,]
augments = [True, False]
idx = 0
for aug in augments:
for config in model_settings:
print("Augmentation", aug)
print("model", config)
settings = config()
settings.AUGMENTATION = aug
settings.DEBUG = False
expdir = exp_base + str(idx) + "/"
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS, test_every=test_every, num_epochs_cv=num_epochs_cv,
learning_rate = LR, print_every=print_every, small_debug=SMALL_DEBUG_DOM)
idx += 1
def main():
res_layers = [3, 9, 27]
cardinalities = [1, 8, 32]
idx = 0
layer = 3
cardinality = 1
expdir = exp_base + str(0) + "/"
print("Layers", layer)
print("Cardinality", cardinality)
kwargs = {"layers": layer, "cardinality": cardinality}
_, settings = ML_utils.load_model_and_settings_from_dir(exp_load)
settings.AE_MODEL_FP = model_fp
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx) + "/"
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS,
test_every=test_every,
num_epochs_cv=num_epochs_cv,
learning_rate=LR,
print_every=print_every,
small_debug=SMALL_DEBUG_DOM)
def check_train_load_DA(config,
config_kwargs,
small_debug=True,
all_data=False,
activation=None,
params={
"var": VAR,
"tol": TOL
}):
expdir = EXPDIR
try:
if not config_kwargs:
config_kwargs = {}
assert isinstance(config_kwargs, dict)
settings = config(**config_kwargs)
settings.DEBUG = False
if activation:
settings.ACTIVATION = activation
calc_DA_MAE = RUN_DA_IN_TRAINING
num_epochs_cv = 0
print_every = 1
test_every = 1
lr = 0.0002
print(settings.__class__.__name__)
if config_kwargs:
print(list([(k, v) for (k, v) in config_kwargs.items()]))
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS,
learning_rate=lr,
test_every=test_every,
num_epochs_cv=num_epochs_cv,
print_every=print_every,
small_debug=small_debug)
if PRINT_MODEL:
print(model.layers_encode)
#test loading
model, settings = ML_utils.load_model_and_settings_from_dir(expdir)
model.to(ML_utils.get_device()) #TODO
x_fp = settings.get_X_fp(True) #force init X_FP
res_AE = run_DA_batch(settings, model, all_data, expdir, params)
print(res_AE.head(10))
shutil.rmtree(expdir, ignore_errors=False, onerror=None)
except Exception as e:
try:
shutil.rmtree(expdir, ignore_errors=False, onerror=None)
raise e
except Exception as z:
raise e
def test_AE_train_3D_DA(self, tmpdir):
"""Test no exception thrown"""
epochs = 1
settings = self.__settings(tmpdir, force_init= True)
expdir = tmpdir.mkdir("experiments/")
calc_DA_MAE = True
trainer = TrainAE(settings, str(expdir))
model = trainer.train(epochs, num_workers=0, calc_DA_MAE=calc_DA_MAE)
def main():
blocks = [
"NeXt",
"vanilla",
"CBAM_vanilla",
"CBAM_NeXt",
]
Cstd = 64
sigmoid = False
activations = ["GDN", "prelu"]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for activation in activations:
for block in blocks:
kwargs = {
"model_name": "Tucodec",
"block_type": block,
"Cstd": Cstd,
"sigmoid": sigmoid,
"activation": activation
}
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
if activation == "prelu":
if block == "NeXt" or block == "vanilla":
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = CLIC(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
structures = [(8, 3), (1, 27)] #(cardinality, layers)
substructures = ["ResNeXt3", "RDB3"]
blocks = ["NeXt", "vanilla"]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for block in blocks:
for substruct in substructures:
for struct in structures:
idx += 1
#split work across 2 gpus:
if idx - 1 < 6 and GPU_DEVICE == 1:
continue
elif idx - 1 >= 4 and GPU_DEVICE == 0:
continue
(cardinality, layers) = struct
kwargs = {
"layers": layers,
"cardinality": cardinality,
"block_type": block,
"module_type": substruct
}
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
kwargs = {
"cardinality": 1,
"block_type": "RNAB",
"sigmoid": True,
"module_type": "Bespoke",
"attenuation": False
}
layers = [
1,
2,
4,
8,
]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for layer in layers:
kwargs[
"subBlock"] = "NeXt" #this performed slightly better on first case
kwargs["layers"] = layer
idx += 1
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
blocks = [
"NeXt",
"vanilla",
"CBAM_vanilla",
"CBAM_NeXt",
]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for block in blocks:
kwargs = {
"block_type": block,
"Cstd": 32,
"aug_scheme": 0,
"activation": "prelu"
}
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = GRDNBaseline(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
layers = [3, 6, 9, 18, 27]
cardinalities = [1, 4, 8, 16, 32]
substructure = "ResNeXt3"
block = "NeXt"
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for layer in layers:
for cardinality in cardinalities:
idx_ = idx
idx += 1
#split work across 4 gpus:
if idx_ % 4 != GPU_DEVICE:
continue
kwargs = {
"layers": layer,
"cardinality": cardinality,
"block_type": block,
"module_type": substructure
}
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx_) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
kwargs = {
"model_name": "Tucodec",
"block_type": "NeXt",
"Cstd": 64,
"sigmoid": False,
"activation": "prelu"
}
aug_schemes = list(range(1, 5))
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for aug_scheme in aug_schemes:
kwargs["aug_scheme"] = aug_scheme
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = CLIC(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
activations = ["relu"]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
expt.LR *= 0.1
idx = 0
for act in activations:
kwargs = {
"layers": 0,
"cardinality": 2,
"aug_scheme": 4,
"activation": act
}
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
blocks = ["NeXt", "vanilla"]
channels = [32, 64]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for block in blocks:
for Cstd in channels:
kwargs = {
"model_name": "Tucodec",
"block_type": block,
"Cstd": Cstd
}
idx += 1
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = CLIC(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
blocks = ["CBAM_vanilla", "vanilla", "CBAM_NeXt"]
kwargs = {"cardinality": 1, "block_type": "RAB",
"sigmoid": True, "module_type": "Bespoke",
"attenuation": False, "layers": 4,
"aug_scheme": 0}
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for block in blocks:
kwargs["subBlock"] = block
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx_) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS, test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate = expt.LR, print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
blocks = ["NeXt", "vanilla"]
kwargs = {
"layers": 1,
"cardinality": 1,
"block_type": "RNAB",
"module_type": "Bespoke"
}
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
idx = 0
for block in blocks:
kwargs["subBlock"] = block
idx += 1
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
#of form (layers, cardinality)
param_vals = [(3, 1), (3, 8), (3, 32), (27, 32)]
param_vals_2 = [
(9, 1),
(9, 8),
(9, 32),
(27, 1),
(27, 8),
]
param_options = [param_vals, param_vals_2]
idx = 0
params = param_options[PARAM_IDX]
for param in params:
layer, cardinality = param
print("Layers", layer)
print("Cardinality", cardinality)
kwargs = {"layers": layer, "cardinality": cardinality}
settings = ResNeXt(**kwargs)
settings.AUGMENTATION = True
settings.DEBUG = False
expdir = exp_base + str(idx) + "/"
trainer = TrainAE(settings, expdir, calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(EPOCHS,
test_every=test_every,
num_epochs_cv=num_epochs_cv,
learning_rate=LR,
print_every=print_every,
small_debug=SMALL_DEBUG_DOM)
idx += 1
def retrain(dir, gpu_device, new_expdir, batch_sz=None):
"""This function accepts an expdir and returns an initialized TrainAE class"""
model, settings, prev_epoch = ML_utils.load_model_and_settings_from_dir(
dir, device_idx=gpu_device, return_epoch=True)
start_epoch = prev_epoch + 1
batch_sz = batch_sz if batch_sz is not None else settings.batch_sz
trainer = TrainAE(settings,
new_expdir,
batch_sz=batch_sz,
model=model,
start_epoch=start_epoch)
return trainer
def main():
activations = ["GDN", "relu"]
lr_factors = [1, 0.2]
resNextk3 = {
"layers": 3,
"cardinality": 8,
"block_type": "vanilla",
"module_type": "RDB3",
"aug_scheme": 0
}
rabkwargs = {
"cardinality": 1,
"block_type": "RAB",
"sigmoid": True,
"module_type": "Bespoke",
"attenuation": False,
"layers": 4,
"aug_scheme": 0,
"subBlock": "NeXt",
}
kwarg_lst = (
resNextk3,
rabkwargs,
)
models = (
ResStack3,
ResStack3,
)
assert len(models) == len(kwarg_lst)
idx = 0
for index, kwargs in enumerate(kwarg_lst):
for idx2, act in enumerate(activations):
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
expt.LR = expt.LR * lr_factors[idx2]
Model = models[index]
kwargs["activation"] = act
if act == "relu":
kwargs["aug_scheme"] = 4
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = Model(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
idx = 0
##################### 02c
structure = (4, 27) #(cardinality, layers)
substructures = ["ResNeXt3", "RDB3"]
blocks = ["CBAM_vanilla", "vanilla", "NeXt", "CBAM_NeXt"]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
expt.EPOCHS = 150
for substruct in substructures:
for block in blocks:
(cardinality, layers) = structure
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
kwargs = {
"layers": layers,
"cardinality": cardinality,
"block_type": block,
"module_type": substruct,
"aug_scheme": 0
}
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx_) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
################# 03c
blocks = ["CBAM_vanilla", "vanilla", "CBAM_NeXt"]
kwargs = {
"cardinality": 1,
"block_type": "RAB",
"sigmoid": True,
"module_type": "Bespoke",
"attenuation": False,
"layers": 4,
"aug_scheme": 0
}
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
expt.EPOCHS = 150
for block in blocks:
kwargs["subBlock"] = block
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = ResStack3(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx_) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
################################06a5
blocks = [
"NeXt",
"vanilla",
"CBAM_vanilla",
"CBAM_NeXt",
]
Cstd = 64
sigmoid = False
activations = ["relu"]
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
expt.EPOCHS = 300
for activation in activations:
for block in blocks:
kwargs = {
"model_name": "Tucodec",
"block_type": block,
"Cstd": Cstd,
"sigmoid": sigmoid,
"activation": activation,
"aug_scheme": 0
}
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = CLIC(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
trainer = TrainAE(settings, expdir, expt.calc_DA_MAE)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
def main():
activations = ["GDN", "relu"]
lr_factors = [1, 0.10]
resNextk1 = {
"layers": 27,
"cardinality": 4,
"block_type": "CBAM_vanilla",
"module_type": "RDB3",
"aug_scheme": 0
}
resNextk2 = {
"layers": 27,
"cardinality": 1,
"block_type": "CBAM_vanilla",
"module_type": "ResNeXt3",
"aug_scheme": 0
}
kwarg_lst = (
resNextk1,
resNextk2,
)
models = (
ResStack3,
ResStack3,
)
assert len(models) == len(kwarg_lst)
idx = 0
for index, kwargs in enumerate(kwarg_lst):
for idx2, act in enumerate(activations):
if TEST:
expt = ExptConfigTest()
else:
expt = ExptConfig()
expt.LR = expt.LR * lr_factors[idx2]
batch_sz = 16
# if act == "GDN" and "module_type" == "RDB3":
# batch_sz = 8
if act == "relu":
kwargs["aug_scheme"] = 4
Model = models[index]
kwargs["activation"] = act
idx_ = idx
idx += 1
if idx_ % NUM_GPU != GPU_DEVICE - GPU_OFFSET:
continue
for k, v in kwargs.items():
print("{}={}, ".format(k, v), end="")
print()
settings = Model(**kwargs)
settings.GPU_DEVICE = GPU_DEVICE
settings.export_env_vars()
expdir = exp_base + str(idx - 1) + "/"
print(expdir)
trainer = TrainAE(settings,
expdir,
expt.calc_DA_MAE,
batch_sz=batch_sz)
expdir = trainer.expdir #get full path
model = trainer.train(expt.EPOCHS,
test_every=expt.test_every,
num_epochs_cv=expt.num_epochs_cv,
learning_rate=expt.LR,
print_every=expt.print_every,
small_debug=expt.SMALL_DEBUG_DOM)
