valid_sampler = BucketingSampler(valid_data, batch_size=args.batch_size)
valid_loader = AudioDataLoader(pad_token_id=0, dataset=valid_data, num_workers=args.num_workers)
valid_loader_list.append(valid_loader)
start_epoch = 0
metrics = None
loaded_args = None
if args.continue_from != "":
logging.info("Continue from checkpoint:" + args.continue_from)
model, vocab, opt, epoch, metrics, loaded_args = load_model(args.continue_from)
start_epoch = (epoch) # index starts from zero
verbose = args.verbose
else:
if args.model == "TRFS":
model = init_transformer_model(args, vocab, is_factorized=args.is_factorized, r=args.r)
opt = init_optimizer(args, model, "noam")
else:
logging.info("The model is not supported, check args --h")
loss_type = args.loss
if USE_CUDA:
model = model.cuda()
logging.info(model)
num_epochs = args.epochs
print("Parameters: {}(trainable), {}(non-trainable)".format(compute_num_params(model)[0], compute_num_params(model)[1]))
trainer = Trainer()
trainer.train(model, vocab, train_loader, train_sampler, valid_loader_list, opt, loss_type, start_epoch, num_epochs, args, metrics, early_stop=args.early_stop)
constant.args.continue_from)
start_epoch = epoch # index starts from zero
verbose = constant.args.verbose
if loaded_args != None:
# Unwrap nn.DataParallel
if loaded_args.parallel:
logging.info("unwrap from DataParallel")
model = model.module
# Parallelize the batch
if args.parallel:
model = nn.DataParallel(model, device_ids=args.device_ids)
else:
if constant.args.model == "TRFS":
model = init_transformer_model(constant.args, label2id, id2label)
opt = init_optimizer(constant.args, model, "noam")
else:
logging.info("The model is not supported, check args --h")
loss_type = args.loss
if constant.USE_CUDA:
model = model.cuda(0)
logging.info(model)
num_epochs = constant.args.epochs
trainer = Trainer()
trainer.train(model, train_loader, train_sampler, valid_loader_list, opt, loss_type, start_epoch, num_epochs, label2id, id2label, metrics)
valid_data = SpectrogramDataset(audio_conf=audio_conf, manifest_filepath=val_manifest, label2id=label2id,
normalize=True, augment=False)
valid_sampler = BucketingSampler(valid_data, batch_size=args.batch_size)
valid_loader = AudioDataLoader(valid_data, num_workers=args.num_workers, batch_sampler=valid_sampler)
valid_loaders.append(valid_loader)
start_epoch = 0
metrics = None
if constant.args.continue_from != "":
print("Continue from checkpoint:", constant.args.continue_from)
model, opt, epoch, metrics, loaded_args, label2id, id2label = load_model(
constant.args.continue_from)
start_epoch = (epoch-1) # index starts from zero
verbose = constant.args.verbose
else:
if constant.args.model == "TRFS":
model = init_transformer_model(constant.args, label2id, id2label)
else:
print("The model is not supported, check args --h")
opt = init_optimizer(constant.args, model)
if constant.USE_CUDA:
print("use cuda")
model = model.cuda()
# print(model)
num_epochs = constant.args.epochs
trainer = MultiTrainer()
trainer.train(model, train_loader, train_sampler, valid_loaders, opt, loss_type, start_epoch, num_epochs, label2id, id2label, metrics, logger)