烤宽面条NN奇怪的行为准确和重量收敛

Question

!!无法用图片来解释!! 我有一个神经网络，用文字新闻提供，用BagOfWOrds和标签{0,1}编码。我用卷积对文本进行分类。一切似乎都在起作用，但是，如果我有一些小数据（20 000条消息），列车的准确性不会收敛，测试的准确性本身就很准确（X轴是批次，绿色是测试，蓝色是火车）：

然后，我想象了我的NN中每一层的wieghts，它让我感到惊讶和困惑（X轴是批次）：

并且

1。我真的无法解释，为什么权重收集在第20批并且之后不会改变，而根据准确性，它们应该是!!

2。为什么测试准确性的行为如此奇怪（绿色线） 我希望这只是代码...

topic_input = lasagne.layers.InputLayer(shape=(None, v_train.shape[1]), input_var=v_t)
embedding = lasagne.layers.EmbeddingLayer(topic_input, input_size=len(token_to_id)+1, output_size=32)
WHAT = lasagne.layers.DimshuffleLayer(embedding, [0,2,1])
conv_1 = lasagne.layers.Conv1DLayer(WHAT, num_filters=15, filter_size=4)
conv_2 = lasagne.layers.Conv1DLayer(conv_1, num_filters=5, filter_size=3)
dense_1 = lasagne.layers.DenseLayer(conv_2, 30)
dense_2 = lasagne.layers.DenseLayer(dense_1, 5)
dense_3 = lasagne.layers.DenseLayer(dense_2, 1, nonlinearity=lasagne.nonlinearities.sigmoid)

weights = lasagne.layers.get_all_params(dense_3,trainable=True)
prediction = lasagne.layers.get_output(dense_3)
loss = lasagne.objectives.binary_crossentropy(prediction, target).mean()
updates = lasagne.updates.adam(loss, weights, learning_rate=0.01)
accuracy = lasagne.objectives.binary_accuracy(prediction, target).mean()

train_func = theano.function([v_t, target], [loss, prediction, accuracy, weights[7]], updates=updates)
acc_func = theano.function([v_t, target], [accuracy, prediction])