lstm_tagger_old.py 文件源码

def train(args):
    model = LSTMTagger(args.model, args.word_emb_size, args.afix_emb_size,
            args.nlayers, args.hidden_dim, args.relu_dim, args.dropout_ratio)
    with open(args.model + "/params", "w") as f:
            log(args, f)
    if args.initmodel:
        print 'Load model from', args.initmodel
        chainer.serializers.load_npz(args.initmodel, model)

    if args.pretrained:
        print 'Load pretrained word embeddings from', args.pretrained
        model.load_pretrained_embeddings(args.pretrained)

    if args.gpu >= 0:
        chainer.cuda.get_device(args.gpu).use()
        model.to_gpu()

    train = LSTMTaggerDataset(args.model, args.train)
    train_iter = chainer.iterators.SerialIterator(train, args.batchsize)
    val = LSTMTaggerDataset(args.model, args.val)
    val_iter = chainer.iterators.SerialIterator(
            val, args.batchsize, repeat=False, shuffle=False)
    optimizer = chainer.optimizers.MomentumSGD(momentum=0.7)
    optimizer.setup(model)
    optimizer.add_hook(WeightDecay(1e-6))
    optimizer.add_hook(GradientClipping(5.))
    updater = training.StandardUpdater(train_iter, optimizer,
            device=args.gpu, converter=converter)
    trainer = training.Trainer(updater, (args.epoch, 'epoch'), args.model)

    val_interval = 2000, 'iteration'
    log_interval = 200, 'iteration'

    eval_model = model.copy()
    eval_model.train = False

    trainer.extend(extensions.Evaluator(
        val_iter, eval_model, converter, device=args.gpu), trigger=val_interval)
    trainer.extend(extensions.snapshot_object(
        model, 'model_iter_{.updater.iteration}'), trigger=val_interval)
    trainer.extend(extensions.LogReport(trigger=log_interval))
    trainer.extend(extensions.PrintReport([
        'epoch', 'iteration', 'main/loss', 'validation/main/loss',
        'main/accuracy', 'validation/main/accuracy',
    ]), trigger=log_interval)
    trainer.extend(extensions.ProgressBar(update_interval=10))

    trainer.run()