分析模型的准确性

Question

我让这个模型用于多变量多步序列预测。

def build_model(train,n_input):
    train_x, train_y = to_supervised(train, n_input)
    verbose, epochs, batch_size = 1, 60,20
    n_timesteps, n_features, n_outputs = train_x.shape[1], train_x.shape[2], train_y.shape[1]
    train_y = train_y.reshape((train_y.shape[0], train_y.shape[1], 1))
    model = Sequential()
    model.add(LSTM(200, activation='relu', input_shape=(n_timesteps, n_features)))
    model.add(RepeatVector(n_outputs))
    model.add(LSTM(200, activation='relu', return_sequences=True))
    model.add(TimeDistributed(Dense(100, activation='relu')))
    model.add(TimeDistributed(Dense(1)))
    model.compile(loss='mse', optimizer='adam', metrics=['accuracy'])
    model.fit(train_x, train_y, epochs=epochs, batch_size=batch_size, verbose=verbose)
    return model

我得到了这个结果：

Epoch 59/60
5516/5516 [==============================] - 21s 4ms/step - loss: 1.8988 - acc: 0.5636: 1s - loss: 1.
Epoch 60/60
5516/5516 [==============================] - 22s 4ms/step - loss: 1.8556 - acc: 0.5685

R2得分：

r2_score(test[:,-1], pred)
0.8688880951315198

从以上结果来看，我有一些疑问：

1）如何测量模型的准确性？是采取训练准确性（此处为loss: 1.8556 - acc: 0.5685还是采取r2 score来衡量准确性。

2）如何提高我的准确性？我再次尝试使用LSTM -CNN获得类似结果。

请帮助我提高准确性。