loss.backward()
optimizer.step()
if len(unitary_params)>0:
unitary_optimizer.step()
n_total = len(outputs)
n_correct = (outputs.argmax(dim = -1) == b_targets.argmax(dim = -1)).sum().item()
train_acc = n_correct/n_total
#Update Progress Bar
pbar.update(params.batch_size)
ordered_dict={'acc': train_acc, 'loss':loss.item()}
pbar.set_postfix(ordered_dict=ordered_dict)
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='running experiments on multimodal datasets.')
parser.add_argument('-config', action = 'store', dest = 'config_file', help = 'please enter configuration file.',default = 'config/run.ini')
args = parser.parse_args()
params = Params()
params.parse_config(args.config_file)
params.config_file = args.config_file
set_seed(params)
params.device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
run(params)
