尝试使用skflow运行一个非常简单的nn分类器。
classifier = skflow.TensorFlowDNNClassifier(
hidden_units=[10, 10, 10],
n_classes=10,
batch_size=100,
learning_rate=0.05)
print (data.train.images).shape
print (data.train.labels).shape
classifier.fit(data.train.images,data.train.labels)
输出是: (73257,3072) (73257,10)
,错误是:
in assert_same_rank
"Shapes %s and %s must have the same rank" % (self, other))
ValueError: Shapes (?, 10) and (?, 10, 10) must have the same rank
我真的不明白这里的问题是什么:(