def main():
# capture the config path from the run arguments
# then process the json configuration file
try:
args = get_args()
config = process_config(args.config)
except:
print("missing or invalid arguments")
exit(0)
# create the experiments dirs
create_dirs([config.summary_dir, config.checkpoint_dir])
# create tensorflow session
sess = tf.Session()
# create an instance of the model you want
model = MyModel(config)
# create your data generator
data = DataGenerator(config)
# create tensorboard logger
logger = Logger(sess, config)
# create trainer and pass all the previous components to it
trainer = SimpleTrainer(sess, model, data, config, logger)
# here you train your model
trainer.train()
trainer.validate()
def main():
# capture the config path from the run arguments
# then process the json configuration file
try:
args = get_args()
config = process_config(args.config)
except:
print("missing or invalid arguments")
exit(0)
# create the experiments dirs
create_dirs([config.summary_dir, config.checkpoint_dir])
# create tensorflow session
sess = tf.Session()
# create an instance of the model you want
model = ExampleModel(config)
# create your data generator
data = DataGenerator(config)
# create tensorboard logger
logger = Logger(sess, config)
# create trainer and pass all the previous components to it
trainer = SimpleTrainer(sess, model, data, config, logger)
# here you train your model
trainer.train()
def main():
# capture the config path from the run arguments
# then process the json configuration file
try:
args = get_args()
config = process_config(args.config)
except:
print("missing or invalid arguments")
exit(0)
# create the experiments dirs
create_dirs([config.summary_dir, config.checkpoint_dir])
# create tensorflow session
sess = tf.Session()
# create an instance of the model you want
model = MyModel(config)
# create your data generator
data = DataGenerator(config)
# create tensorboard logger
logger = Logger(sess, config)
# create trainer and pass all the previous components to it
trainer = SimpleTrainer(sess, model, data, config, logger)
saverExternal = tf.train.Saver(var_list=tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='external'))
saverExternal.restore(sess, "experiments/model_8/model8.ckpt_2")
# here you train your model
trainer.train()
trainer.validate()
def main():
cfg = config()
print('Loading Data...')
data_loader = CaptchaDataLoader(cfg)
print('Loading Model...')
model = SimpleCaptchaModel(cfg)
print('Loading Trainer...')
trainer = SimpleTrainer(model.model, data_loader.get_train_data(), cfg)
trainer.train()
def main():
modalities = [
DatasetFeaturesSet.VIDEO_SCENE_R2PLUS1_FEATURES,
DatasetFeaturesSet.VIDEO_FACE,
DatasetFeaturesSet.AUDIO,
DatasetFeaturesSet.PULSE,
DatasetFeaturesSet.VIDEO_SCENE,
]
fusion_types = ['sum', 'concatenation', 'fbp']
for i in range((len(modalities))):
for fusion in fusion_types:
processor = MultimodalDatasetFeaturesProcessor(modalities_list=modalities[i:])
data_manager = DataManager(
tf_record_path=config.DATASET_TF_RECORDS_PATH,
batch_size=1)
model = MultimodalModel(
fusion_type=fusion,
modalities_list=modalities[i:],
fc_units=64,
first_layer=128,
second_layer=64,
learning_rate=0.0001,
cp_dir=config.CHECKPOINT_DIR + "/" + fusion + "_" + modalities[i].name,
cp_name=config.CHECKPOINT_NAME
)
_, epoch = model.load()
trainer = SimpleTrainer(
dataset_processor=processor,
model=model,
data=data_manager,
board_path=config.TENSORBOARD_DIR + "/" + fusion + "_" + modalities[i].name,
log_freq=config.LOG_AND_SAVE_FREQ_BATCH,
num_epochs=2,
initial_epoch=epoch,
create_dirs_flag=True
)
metrics = trainer.train()
print(metrics)
def main():
# capture the config path from the run arguments
# then process the json configuration file
try:
args = get_args()
config = process_config(args.config)
except:
print("missing or invalid arguments")
exit(0)
# create the experiments dirs
create_dirs([config.summary_dir, config.checkpoint_dir])
# create tensorflow session
sess = tf.Session()
# create an instance of the model you want
model = MyModel(config)
# create your data generator
data = DataGenerator(config)
# create tensorboard logger
logger = Logger(sess, config)
# create trainer and pass all the previous components to it
trainer = SimpleTrainer(sess, model, data, config, logger)
saverExternal = tf.train.Saver(var_list=tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='external'))
saverExternal.restore(sess, "experiments/model_8/model8.ckpt_2")
# here you train your model
trainer.train()
trainer.validate()
def main():
modalities = [
DatasetFeaturesSet.AUDIO, DatasetFeaturesSet.VIDEO_FACE,
DatasetFeaturesSet.VIDEO_SCENE, DatasetFeaturesSet.PULSE,
DatasetFeaturesSet.VIDEO_SCENE_R2PLUS1_FEATURES
]
for modality in modalities:
processor = MultimodalDatasetFeaturesProcessor(
modalities_list=[modality])
data_manager = DataManager(
tf_record_path=config.DATASET_TF_RECORDS_PATH, batch_size=1)
model = UnimodalModel(
modality=modality,
fc_units=64,
first_layer=128,
second_layer=64,
learning_rate=0.001,
pretrained_model_path=modality.config.extractor.config.
pretrained_path if modality.config.extractor is not None else None,
cp_dir=config.CHECKPOINT_DIR,
cp_name=config.CHECKPOINT_NAME)
_, epoch = model.load()
trainer = SimpleTrainer(dataset_processor=processor,
model=model,
data=data_manager,
board_path=config.TENSORBOARD_DIR,
log_freq=config.LOG_AND_SAVE_FREQ_BATCH,
num_epochs=2,
initial_epoch=epoch,
create_dirs_flag=True)
metrics = trainer.train()
print(metrics)
import argparse
import yaml
from trainers.base_trainer import BaseTrainer
from trainers.rgb_only_trainer import RGBOnlyTrainer
from trainers.simple_trainer import SimpleTrainer
from trainers.local_global_trainer import LocalGlobalTrainer
parser = argparse.ArgumentParser()
parser.add_argument("--config", type=str, help='path to config file')
try:
args = parser.parse_args()
except IOError as msg:
parser.error(str(msg))
with open("configs/" + args.config, 'r') as f:
params = yaml.load(f, Loader=yaml.FullLoader)
if params["type"] == "simple":
trainer = SimpleTrainer(params)
elif params["type"] == "local_global":
trainer = LocalGlobalTrainer(params)
elif params["type"] == "rgb":
trainer = RGBOnlyTrainer(params)
trainer.train()