PyTorch搭建ANN實(shí)現(xiàn)時(shí)間序列風(fēng)速預(yù)測(cè)

更新時(shí)間：2022年05月11日 15:36:11 作者：Cyril_KI

這篇文章主要為大家介紹了PyTorch搭建ANN實(shí)現(xiàn)時(shí)間序列風(fēng)速預(yù)測(cè)，有需要的朋友可以借鑒參考下，希望能夠有所幫助，祝大家多多進(jìn)步，早日升職加薪

數(shù)據(jù)集

數(shù)據(jù)集為Barcelona某段時(shí)間內(nèi)的氣象數(shù)據(jù)，其中包括溫度、濕度以及風(fēng)速等。本文將簡(jiǎn)單搭建來(lái)對(duì)風(fēng)速進(jìn)行預(yù)測(cè)。

特征構(gòu)造

對(duì)于風(fēng)速的預(yù)測(cè)，除了考慮歷史風(fēng)速數(shù)據(jù)外，還應(yīng)該充分考慮其余氣象因素的影響。因此，我們根據(jù)前24個(gè)時(shí)刻的風(fēng)速+下一時(shí)刻的其余氣象數(shù)據(jù)來(lái)預(yù)測(cè)下一時(shí)刻的風(fēng)速。

數(shù)據(jù)處理

1.數(shù)據(jù)預(yù)處理

數(shù)據(jù)預(yù)處理階段，主要將某些列上的文本數(shù)據(jù)轉(zhuǎn)為數(shù)值型數(shù)據(jù)，同時(shí)對(duì)原始數(shù)據(jù)進(jìn)行歸一化處理。文本數(shù)據(jù)如下所示：

經(jīng)過(guò)轉(zhuǎn)換后，上述各個(gè)類別分別被賦予不同的數(shù)值，比如"sky is clear"為0，"few clouds"為1。

def load_data():
    global Max, Min
    df = pd.read_csv('Barcelona/Barcelona.csv')
    df.drop_duplicates(subset=[df.columns[0]], inplace=True)
    # weather_main
    listType = df['weather_main'].unique()
    df.fillna(method='ffill', inplace=True)
    dic = dict.fromkeys(listType)
    for i in range(len(listType)):
        dic[listType[i]] = i
    df['weather_main'] = df['weather_main'].map(dic)
    # weather_description
    listType = df['weather_description'].unique()
    dic = dict.fromkeys(listType)
    for i in range(len(listType)):
        dic[listType[i]] = i
    df['weather_description'] = df['weather_description'].map(dic)
    # weather_icon
    listType = df['weather_icon'].unique()
    dic = dict.fromkeys(listType)
    for i in range(len(listType)):
        dic[listType[i]] = i
    df['weather_icon'] = df['weather_icon'].map(dic)
    # print(df)
    columns = df.columns
    Max = np.max(df['wind_speed'])  # 歸一化
    Min = np.min(df['wind_speed'])
    for i in range(2, 17):
        column = columns[i]
        if column == 'wind_speed':
            continue
        df[column] = df[column].astype('float64')
        if len(df[df[column] == 0]) == len(df):  # 全0
            continue
        mx = np.max(df[column])
        mn = np.min(df[column])
        df[column] = (df[column] - mn) / (mx - mn)
    # print(df.isna().sum())
    return df

2.數(shù)據(jù)集構(gòu)造

利用當(dāng)前時(shí)刻的氣象數(shù)據(jù)和前24個(gè)小時(shí)的風(fēng)速數(shù)據(jù)來(lái)預(yù)測(cè)當(dāng)前時(shí)刻的風(fēng)速：

def nn_seq():
    """
    :param flag:
    :param data: 待處理的數(shù)據(jù)
    :return: X和Y兩個(gè)數(shù)據(jù)集,X=[當(dāng)前時(shí)刻的year，month, hour, day, lowtemp, hightemp, 前一天當(dāng)前時(shí)刻的負(fù)荷以及前23小時(shí)負(fù)荷]
                              Y=[當(dāng)前時(shí)刻負(fù)荷]
    """
    print('處理數(shù)據(jù):')
    data = load_data()
    speed = data['wind_speed']
    speed = speed.tolist()
    speed = torch.FloatTensor(speed).view(-1)
    data = data.values.tolist()
    seq = []
    for i in range(len(data) - 30):
        train_seq = []
        train_label = []
        for j in range(i, i + 24):
            train_seq.append(speed[j])
        # 添加溫度、濕度、氣壓等信息
        for c in range(2, 7):
            train_seq.append(data[i + 24][c])
        for c in range(8, 17):
            train_seq.append(data[i + 24][c])
        train_label.append(speed[i + 24])
        train_seq = torch.FloatTensor(train_seq).view(-1)
        train_label = torch.FloatTensor(train_label).view(-1)
        seq.append((train_seq, train_label))
    # print(seq[:5])
    Dtr = seq[0:int(len(seq) * 0.5)]
    Den = seq[int(len(seq) * 0.50):int(len(seq) * 0.75)]
    Dte = seq[int(len(seq) * 0.75):len(seq)]
    return Dtr, Den, Dte

任意輸出其中一條數(shù)據(jù)：

(tensor([1.0000e+00, 1.0000e+00, 2.0000e+00, 1.0000e+00, 1.0000e+00, 1.0000e+00,
        1.0000e+00, 1.0000e+00, 0.0000e+00, 1.0000e+00, 5.0000e+00, 0.0000e+00,
        2.0000e+00, 0.0000e+00, 0.0000e+00, 5.0000e+00, 0.0000e+00, 2.0000e+00,
        2.0000e+00, 5.0000e+00, 6.0000e+00, 5.0000e+00, 5.0000e+00, 5.0000e+00,
        5.3102e-01, 5.5466e-01, 4.6885e-01, 1.0066e-03, 5.8000e-01, 6.6667e-01,
        0.0000e+00, 0.0000e+00, 0.0000e+00, 0.0000e+00, 9.9338e-01, 0.0000e+00,
        0.0000e+00, 0.0000e+00]), tensor([5.]))

數(shù)據(jù)被劃分為三部分：Dtr、Den以及Dte，Dtr用作訓(xùn)練集，Dte用作測(cè)試集。

ANN模型

1.模型訓(xùn)練

ANN模型搭建如下：

def ANN():
    Dtr, Den, Dte = nn_seq()
    my_nn = torch.nn.Sequential(
        torch.nn.Linear(38, 64),
        torch.nn.ReLU(),
        torch.nn.Linear(64, 128),
        torch.nn.ReLU(),
        torch.nn.Linear(128, 1),
    )
    model = my_nn.to(device)
    loss_function = nn.MSELoss().to(device)
    optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
    train_inout_seq = Dtr
    # 訓(xùn)練
    epochs = 50
    for i in range(epochs):
        print('當(dāng)前', i)
        for seq, labels in train_inout_seq:
            seq = seq.to(device)
            labels = labels.to(device)
            y_pred = model(seq)
            single_loss = loss_function(y_pred, labels)
            optimizer.zero_grad()
            single_loss.backward()
            optimizer.step()
        # if i % 2 == 1:
        print(f'epoch: {i:3} loss: {single_loss.item():10.8f}')
    print(f'epoch: {i:3} loss: {single_loss.item():10.10f}')
    state = {'model': model.state_dict(), 'optimizer': optimizer.state_dict(), 'epoch': epochs}
    torch.save(state, 'Barcelona' + ANN_PATH)

可以看到，模型定義的代碼段為：

my_nn = torch.nn.Sequential(
    torch.nn.Linear(38, 64),
    torch.nn.ReLU(),
    torch.nn.Linear(64, 128),
    torch.nn.ReLU(),
    torch.nn.Linear(128, 1),
)

第一層全連接層輸入維度為38（前24小時(shí)風(fēng)速+14種氣象數(shù)據(jù)），輸出維度為64；第二層輸入為64，輸出128；第三層輸入為128，輸出為1。

2.模型預(yù)測(cè)及表現(xiàn)

def ANN_predict(ann, test_seq):
    pred = []
    for seq, labels in test_seq:
        seq = seq.to(device)
        with torch.no_grad():
            pred.append(ann(seq).item())
    pred = np.array([pred])
    return pred

測(cè)試：

def test():
    Dtr, Den, Dte = nn_seq()
    ann = torch.nn.Sequential(
        torch.nn.Linear(38, 64),
        torch.nn.ReLU(),
        torch.nn.Linear(64, 128),
        torch.nn.ReLU(),
        torch.nn.Linear(128, 1),
    )
    ann = ann.to(device)
    ann.load_state_dict(torch.load('Barcelona' + ANN_PATH)['model'])
    ann.eval()
    pred = ANN_predict(ann, Dte)
    print(mean_absolute_error(te_y, pred2.T), np.sqrt(mean_squared_error(te_y, pred2.T)))

ANN在Dte上的表現(xiàn)如下表所示：