I am trying to train a LSTM like,
from __future__ import print_function
import mxnet as mx
import numpy as np
from mxnet import nd, autograd, sym
from mxnet import gluon
ctx = mx.cpu()
LIMIT = 20
data = np.array([(s, 1) for s in spanish_sentences[LIMIT]] + [(s, 0) for s in english_sentences[LIMIT]])
layer = mx.gluon.rnn.LSTM(100, 3)
net = mx.gluon.nn.Dense(2)
softmax_cross_entropy = gluon.loss.SoftmaxCrossEntropyLoss()
layer.initialize(ctx=ctx)
net.collect_params().initialize(ctx=ctx)
trainer = gluon.Trainer(net.collect_params(), 'sgd', {'learning_rate': .1})
for epoch in range(10):
np.random.shuffle(data)
losses = []
for s, l in data:
if len(s) == 0:
continue
x = nd.array([ord(c) for c in s]).reshape(shape=(-1, 1, 1))
y = nd.array([np.eye(2)[int(l)]])
with autograd.record():
output = layer(x)[output.shape[0]-1, :, :]
pred = net(output)
loss = softmax_cross_entropy(pred, y)
losses.append(loss.asscalar())
trainer.step(1, ignore_stale_grad=True)
print("Loss:", np.mean(losses), "+-", np.std(losses))
But I am getting an error,
---------------------------------------------------------------------------
MXNetError Traceback (most recent call last)
<ipython-input-31-12ab8d4ad733> in <module>()
30 output = layer(x)[output.shape[0]-1, :, :]
31 pred = net(output)
---> 32 loss = softmax_cross_entropy(pred, y)
33 losses.append(loss.asscalar())
34 trainer.step(1, ignore_stale_grad=True)
... Stack trace ...
MXNetError: Shape inconsistent, Provided=(1,2), inferred shape=(1,1)
What is being done wrong? When I test the shape of pred and y, I get that they are both equal to (1, 2). I don't know why it is expecting (1, 1).
It was rather simple. SoftMaxCrossEntropy()(pred, label) expects shapes pred.shape = (BATCH_SIZE, N_LABELS) and label.shape = (BATCH_SIZE,).
So y = nd.array([l]) fixed it.