二进制分类器出错了。始终预测1

Question

下面是张量流中的二元分类器。它总是预测1，因此报告100％的准确性。有人可以解释我哪里出错了吗？

def mlp(curr_data_batch, y1, i, ep):
    with tf.variable_scope("name"+str(i), reuse=True):             
        b = tf.get_variable("bias")
        W = tf.get_variable("weights")
        y = tf.nn.softmax(tf.matmul(x,W) + b)
        cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y), 
        reduction_indices=[1]))
        train_op = tf.train.GradientDescentOptimizer(learning_rate
        ).minimize(cross_entropy)       
        sess.run([train_op, y], feed_dict={x: curr_data_batch, y_: y1})
        correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1))
        accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
        print i, ep, ("Accuracy:", accuracy.eval({x: curr_data_batch, y_: 
        y1}, session=sess))

编辑：我刚刚看到权重没有得到更新。它是否因可变范围而发生？