def main():
args = parse_args()
inputFileNameUsage = args['input_file_usage']
inputFileNameOperative = args['input_file_operative']
outputApplicationsFileName = args['output_file_applications']
outputUsersFileName = args['output_file_users']
predictionOutputFileName = args['output_file_applications_prediction']
utils.log_config()
logger = logging.getLogger(__name__)
startTime = datetime.datetime.now()
# exported to global scope for debugging purposes
global df
df = data_helper.import_data(inputFileNameUsage)
global odf
if inputFileNameOperative:
odf = data_helper.import_operative_data(inputFileNameOperative)
else:
odf = None
logger.info("N of events: {}, from {} to {} ".format(
len(df), df['datetime'].min(), df['datetime'].max()))
create_user_summary(outputUsersFileName)
create_application_summary(outputApplicationsFileName)
create_prediction_summary(predictionOutputFileName)
print_stats(startTime)
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--hp_file', type=str, default='hparams.py')
args = parser.parse_args()
hp.configure(args.hp_file)
fill_variables(hp)
log_config(hp)
os.makedirs(hp.save_dir, exist_ok=True)
n_gpus = torch.cuda.device_count()
args.__setattr__('n_gpus', n_gpus)
if n_gpus > 1:
run_distributed(run_training, args, hp)
else:
run_training(0, args, hp, None)
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('--hp_file', metavar='FILE', default='hparams.py')
args = parser.parse_args()
#overwrite_hparams(args)
hp.configure(args.hp_file)
fill_variables()
hp.save_dir = os.path.join(hp.save_dir, 'LM')
os.makedirs(hp.save_dir, exist_ok=True)
if hp.debug_mode == 'tensorboard':
writer = SummaryWriter(f'{hp.save_dir}/logs/{hp.comment}')
log_config()
model = Model_lm(hp)
model.apply(init_weight)
if torch.cuda.device_count() > 1:
# multi-gpu configuration
ngpu = torch.cuda.device_count()
device_ids = list(range(ngpu))
model = torch.nn.DataParallel(model, device_ids)
model.to(DEVICE)
optimizer = torch.optim.Adam(model.parameters(), weight_decay=1e-5)
load_epoch = 0
if hp.load_checkpoints:
else:
start_epoch = 0
step = 1
pytorch_total_params = sum(p.numel() for p in model.parameters())
print('params = {0:.2f}M'.format(pytorch_total_params / 1000 / 1000))
train_epoch(model, optimizer, args, hp, step=step, start_epoch=start_epoch, rank=rank)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--hp_file', type=str, default='hparams.py')
parser.add_argument('--debug', action='store_true')
args = parser.parse_args()
hp.configure(args.hp_file)
fill_variables(hp)
log_config(hp)
os.makedirs(hp.save_dir, exist_ok=True)
# # multi-gpu setup
# if torch.cuda.device_count() > 1:
# # multi-gpu configuration
# ngpu = torch.cuda.device_count()
# device_ids = list(range(ngpu))
# model = torch.nn.DataParallel(model, device_ids)
# model.cuda()
# else:
# model.to(DEVICE)
n_gpus = torch.cuda.device_count()
args.__setattr__('n_gpus', n_gpus)