class CifarExperimentRMSProp(CifarExperiment):
"""
Performs experiments over Cifar-10 dataset using RMSProp
"""
optimizer_name = const.TR_BASE
general_config = None
trainer = None
def _prepare_trainer(self):
tester = Tester(self.neural_net, self.data_input, self.input_tensor, self.output_tensor)
self.trainer = RMSPropTrainer(self.general_config, self.neural_net, self.data_input,
self.input_tensor, self.output_tensor, tester=tester, checkpoint=self.ckp_path)
def _prepare_config(self, str_optimizer: str, train_mode: TrainMode):
self.general_config = GeneralConfig(train_mode, 0.0001, self.summary_interval, self.ckp_interval,
config_name=str_optimizer, model_name=self.dataset_name)
# Creates configuration for 5 mega-batches
if train_mode == TrainMode.INCREMENTAL:
for i in range(5):
train_conf = MegabatchConfig(100, batch_size=128)
self.general_config.add_train_conf(train_conf)
elif train_mode == TrainMode.ACUMULATIVE:
train_confs = [MegabatchConfig(100, batch_size=128), MegabatchConfig(70, batch_size=128),
MegabatchConfig(55, batch_size=128), MegabatchConfig(40, batch_size=128),
MegabatchConfig(40, batch_size=128)]
self.general_config.train_configurations = train_confs
else:
raise OptionNotSupportedError("The requested Experiment class: {} doesn't support the requested training"
" mode: {}".format(self.__class__, train_mode))
def _prepare_config(self, str_optimizer: str, train_mode: TrainMode):
self.general_config = GeneralConfig(train_mode,
0.0001,
self.summary_interval,
self.ckp_interval,
config_name=str_optimizer,
model_name=self.dataset_name)
# Creates configuration for 5 mega-batches
if train_mode == TrainMode.INCREMENTAL:
for i in range(5):
train_conf = MegabatchConfig(60, batch_size=128)
self.general_config.add_train_conf(train_conf)
elif train_mode == TrainMode.ACUMULATIVE:
train_confs = [
MegabatchConfig(60, batch_size=128),
MegabatchConfig(50, batch_size=128),
MegabatchConfig(30, batch_size=128),
MegabatchConfig(30, batch_size=128),
MegabatchConfig(30, batch_size=128)
]
self.general_config.train_configurations = train_confs
else:
raise OptionNotSupportedError(
"The requested Experiment class: {} doesn't support the requested training"
" mode: {}".format(self.__class__, train_mode))