如何在 Keras 中为 LSTM 的多个输入重新构建数据帧？

Question

I am trying to predict the temperature for the given area (its integer number from 1 to 142) for the given date and time.

The problem is that I have CSV with the following columns:

DateTime,AreaID,Temperature

How to reframe the data-frame for LSTM (Apologise as I am a new bee for the LSTM)?

For the information, I have data for two months with a measured by the period of every 5 minutes.

I have coded LSTM for Input DateTime. But I want to include AreaID too. to predict Temperature.

The dataset created for the Training and Testing sets are using the following code block:

    dataset = dataset.temperature.values #numpy.ndarray
    dataset = dataset.astype('float32')
    dataset = np.reshape(dataset, (-1, 1))
    scaler = MinMaxScaler(feature_range=(0, 1))
    dataset = scaler.fit_transform(dataset)
    train_size = int(len(dataset) * 0.80)
    test_size = len(dataset) - train_size
    train, test = dataset[0:train_size,:], dataset[train_size:len(dataset),:]
    def create_dataset(dataset, look_back=1):
        X, Y = [], []
        for i in range(len(dataset)-look_back-1):
            a = dataset[i:(i+look_back), 0]
            X.append(a)
            Y.append(dataset[i + look_back, 0])
        return np.array(X), np.array(Y)

    look_back = 30
    X_train, Y_train = create_dataset(train, look_back)
    X_test, Y_test = create_dataset(test, look_back)

    # reshape input to be [samples, time steps, features]
    X_train = np.reshape(X_train, (X_train.shape[0], 1, X_train.shape[1]))
    X_test = np.reshape(X_test, (X_test.shape[0], 1, X_test.shape[1]))

Before this, The sample code have sorted the data frame based on DateTime like:

dataset.sort_values('timestamp', inplace=True, ascending=True)

I want to change LSTM to take two inputs 1. DateTime 2. AreaID

& One Output: 1. Temperature

How to code LSTM for this requirements? (Please help me I am a new bee in the area of neural network)

Answer 1

Just for hint.

You prepare new dataset into x_train and y_train Take an starting 60 days to train the model and predict 61th days thats my logic

X_train=[]
y_train=[]
count=0
for i in range(60,train.shape[0]):
    count=count+1
    X_train.append(df[i-60:i])
    y_train.append(train['targetcol'][i])