if seed is not None:
train_config["seed"] = seed
train_config["gqcnn"]["seed"] = seed
if tensorboard_port is not None:
train_config["tensorboard_port"] = tensorboard_port
gqcnn_params = train_config["gqcnn"]
# Create a unique output folder based on the date and time.
if save_datetime:
# Create output dir.
unique_name = time.strftime("%Y%m%d-%H%M%S")
output_dir = os.path.join(output_dir, unique_name)
utils.mkdir_safe(output_dir)
# Set visible devices.
if "gpu_list" in train_config:
gqcnn_utils.set_cuda_visible_devices(train_config["gpu_list"])
# Fine-tune the network.
start_time = time.time()
gqcnn = get_gqcnn_model(backend)(gqcnn_params)
trainer = get_gqcnn_trainer(backend)(gqcnn,
dataset_dir,
split_name,
output_dir,
train_config,
name=name)
trainer.finetune(model_dir)
logger.info("Total Fine-tuning Time: " +
str(utils.get_elapsed_time(time.time() - start_time)))
from gqcnn import get_gqcnn_model, get_gqcnn_trainer, utils as gqcnn_utils
#%%
logger = Logger.get_logger('tools/train.py')
dataset_dir = '/home/ai/git/gqcnn/data/training/Dexnet-2.0_testTraining'
train_config = YamlConfig('cfg/train_dex-net_2.0.yaml')
gqcnn_params = train_config['gqcnn']
#%%
print(os.path.join(dataset_dir, 'config.json'))
config_filename = os.path.join(dataset_dir, 'config.json')
print(config_filename)
# print(os.getcwd())
#%%
open(config_filename, 'r')
config = json.load(open(config_filename, 'r'))
#%%
start_time = time.time()
gqcnn = get_gqcnn_model('tf')(gqcnn_params)
#%%
trainer = get_gqcnn_trainer('tf')(gqcnn,
'data/training/Dexnet-2.0_testTraining',
'image_wise', 'models/', train_config,
'GQCNN-2.0_Training_from_Scratch')
#%%
trainer.train()
#%%
logger.info('Total Training Time: ' +
str(utils.get_elapsed_time(time.time() - start_time)))