回答此问题可获得 20 贡献值,回答如果被采纳可获得 50 分。
<p>我使用的是Keras的序列模型,具有密集层类型。我写了一个函数,它递归地计算预测,但是这些预测离我很远。我想知道什么是我的数据最好的激活函数。目前我正在使用硬乙状体函数。输出数据值的范围为5到25。输入数据具有形状(6,1),输出数据为单个值。当我绘制预测图时,它们从不减少。谢谢你的帮助!!在</p>
<pre><code># create and fit Multilayer Perceptron model
model = Sequential();
model.add(Dense(20, input_dim=look_back, activation='hard_sigmoid'))
model.add(Dense(16, activation='hard_sigmoid'))
model.add(Dense(1))
model.compile(loss='mean_squared_error', optimizer='adam')
model.fit(trainX, trainY, epochs=200, batch_size=2, verbose=0)
#function to predict using predicted values
numOfPredictions = 96;
for i in range(numOfPredictions):
temp = [[origAndPredictions[i,0],origAndPredictions[i,1],origAndPredictions[i,2],origAndPredictions[i,3],origAndPredictions[i,4],origAndPredictions[i,5]]]
temp = numpy.array(temp)
temp1 = model.predict(temp)
predictions = numpy.append(predictions, temp1, axis=0)
temp2 = []
temp2 = [[origAndPredictions[i,1],origAndPredictions[i,2],origAndPredictions[i,3],origAndPredictions[i,4],origAndPredictions[i,5],predictions[i,0]]]
temp2 = numpy.array(temp2)
origAndPredictions = numpy.vstack((origAndPredictions, temp2))
</code></pre>
<p><a href="https://i.stack.imgur.com/ZQvm9.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/ZQvm9.png" alt="enter image description here"/></a></p>
<p>更新:
我用这段代码实现了swish。在</p>
^{pr2}$
<p><a href="https://i.stack.imgur.com/qE3vb.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/qE3vb.png" alt="New plot of predictions using swish."/></a></p>
<p>更新:
使用此代码:</p>
^{3}$
<p><a href="https://i.stack.imgur.com/bt1nz.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/bt1nz.png" alt="enter image description here"/></a></p>
<p>谢谢你的帮助!!在</p>