3a770cbc97085c2cd4eaa0a46b2bc037f35389c2,utils.py,,evaluate_accuracy,#Any#Any#Any#,77

Before Change

        data_iterator.reset()
    for i, batch in enumerate(data_iterator):
        data, label = _get_batch(batch, ctx)
        output = net(data)
        acc += accuracy(output, label)
    return acc / (i+1)

def train(train_data, test_data, net, loss, trainer, ctx, num_epochs, print_batches=None):
    Train a network
    for epoch in range(num_epochs):

After Change

            data.shape[0])

def evaluate_accuracy(data_iterator, net, ctx=[mx.cpu()]):
    acc = nd.array([0])
    n = 0.
    if isinstance(data_iterator, mx.io.MXDataIter):
        data_iterator.reset()
    for batch in data_iterator:
        data, label, batch_size = _get_batch(batch, ctx)
        for X, y in zip(data, label):
            acc += nd.sum(net(X).argmax(axis=1)==y).copyto(mx.cpu())
        acc.wait_to_read() // don"t push too many operators into backend
        n += batch_size
    return acc.asscalar() / n

def train(train_data, test_data, net, loss, trainer, ctx, num_epochs, print_batches=None):
    Train a network
    if isinstance(ctx, mx.Context):

Instances