使用 Chainer 训练 MLP 时出错

Question

我正在尝试训练和测试一个简单的多层感知器，就像在第一个 Chainer 教程中一样，但使用我自己的数据集而不是 MNIST。这是我正在使用的代码（主要来自教程）：

class MLP(Chain):
    def __init__(self, n_units, n_out):
        super(MLP, self).__init__()
        with self.init_scope():
            self.l1 = L.Linear(None, n_units)
            self.l2 = L.Linear(None, n_units)
            self.l3 = L.Linear(None, n_out)
    def __call__(self, x):
        h1 = F.relu(self.l1(x))
        h2 = F.relu(self.l2(h1))
        y = self.l3(h2)
        return y

X, X_test, y, y_test, xHeaders, yHeaders = load_train_test_data('xHeuristicData.csv', 'yHeuristicData.csv')

print 'dataset shape   X:', X.shape, '  y:', y.shape

model = MLP(100, 1)
optimizer = optimizers.SGD()
optimizer.setup(model)

train = tuple_dataset.TupleDataset(X, y)
test = tuple_dataset.TupleDataset(X_test, y_test)

train_iter = iterators.SerialIterator(train, batch_size=100, shuffle=True)
test_iter = iterators.SerialIterator(test, batch_size=100, repeat=False, shuffle=False)
updater = training.StandardUpdater(train_iter, optimizer)
trainer = training.Trainer(updater, (10, 'epoch'), out='result')

trainer.extend(extensions.Evaluator(test_iter, model))
trainer.extend(extensions.LogReport())
trainer.extend(extensions.PrintReport(['epoch', 'main/accuracy', 'validation/main/accuracy']))
trainer.extend(extensions.ProgressBar())

trainer.run()

print 'Predicted value for a test example'
print model(X_test[0])

我没有训练和打印预测值，而是在 "trainer.run()" 处收到以下错误：

dataset shape   X: (1003, 116)   y: (1003,)
Exception in main training loop: __call__() takes exactly 2 arguments (3 given)
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/trainer.py", line 299, in run
    update()
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/updater.py", line 223, in update
    self.update_core()
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/updater.py", line 234, in update_core
    optimizer.update(loss_func, *in_arrays)
  File "/usr/local/lib/python2.7/dist-packages/chainer/optimizer.py", line 534, in update
    loss = lossfun(*args, **kwds)
Will finalize trainer extensions and updater before reraising the exception.
Traceback (most recent call last):
  File "trainHeuristicChainer.py", line 76, in <module>
    trainer.run()
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/trainer.py", line 313, in run
    six.reraise(*sys.exc_info())
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/trainer.py", line 299, in run
    update()
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/updater.py", line 223, in update
    self.update_core()
  File "/usr/local/lib/python2.7/dist-packages/chainer/training/updater.py", line 234, in update_core
    optimizer.update(loss_func, *in_arrays)
  File "/usr/local/lib/python2.7/dist-packages/chainer/optimizer.py", line 534, in update
    loss = lossfun(*args, **kwds)
TypeError: __call__() takes exactly 2 arguments (3 given)

我不知道如何处理这个错误。我已经使用其他框架成功地训练了类似的网络，但我对 Chainer 很感兴趣，因为它与 PyPy 兼容。

此处提供包含文件的 tgz：https://mega.nz/#!wwsBiSwY!g72pC5ZgekeMiVr-UODJOqQfQZZU3lCqm9Er2jH4UD8

Answer 1

您正在将 (X, y) 的元组发送到 MLP，而实现的 __call__ 仅接受 x.

您可以将实现修改为

class MLP(Chain):
    def __init__(self, n_units, n_out):
        super(MLP, self).__init__()
        with self.init_scope():
            self.l1 = L.Linear(None, n_units)
            self.l2 = L.Linear(None, n_units)
            self.l3 = L.Linear(None, n_out)
    def __call__(self, x, y):
        h1 = F.relu(self.l1(x))
        h2 = F.relu(self.l2(h1))
        predict = self.l3(h2)
        loss = F.squared_error(predict, y)
        // or you can write it on your own as follows
        // loss = F.sum(F.square(predict - y))
        return loss

chainer 中可能与其他框架不同，默认情况下标准更新程序假定 __call__ 是损失函数。所以调用 model(X, y) 会 return 损失当前的 mini-batch。这就是为什么 chainer 教程引入了另一个 Classifier class 来计算损失函数并保持 MLP 简单。分类器在 MNIST 中有意义，但不适合您的任务，因此您需要自行实现损失函数。

当你完成训练后，你可以只保存模型实例（也许通过在训练器中添加 snapshot_object 的扩展）。

要使用保存的模型，就像在测试中一样，您必须在 class 中编写另一个方法，可能命名为 test，使用与当前 __call__ 相同的代码，这手头只有 X 输入，因此不需要其他 y。

此外，如果你不喜欢在MLP中添加任何额外的方法class，让它变得纯粹，那么你需要自己编写updater并更自然地计算损失函数。继承标准的比较简单，可以这样写，

class MyUpdater(chainer.training.StandardUpdater):
    def __init__(self, data_iter, model, opt, device=-1):
        super(MyUpdater, self).__init__(data_iter, opt, device=device)
        self.mlp = model

    def update_core(self):
        batch = self.get_iterator('main').next()
        x, y = self.converter(batch, self.device)
        predict = self.mlp(x)
        loss = F.squared_error(predict, y)
        self.mlp.cleargrads()
        loss.backward()
        self.get_iterator('main').update()

updater = MyUpdater(train_iter, model, optimizer)

使用 Chainer 训练 MLP 时出错

Error training a MLP using Chainer

chainer