如何用XGBoost做時(shí)間序列預(yù)測(cè)？

# forecast monthlybirths with xgboostfrom numpy importasarrayfrom pandas importread_csvfrom pandas importDataFramefrom pandas importconcatfrom sklearn.metricsimport mean_absolute_errorfrom xgboost importXGBRegressorfrom matplotlib importpyplot # transform a timeseries dataset into a supervised learning datasetdefseries_to_supervised(data, n_in=1, n_out=1, dropnan=True):       n_vars = 1 if type(data) is list elsedata.shape[1]       df = DataFrame(data)       cols = list()       # input sequence (t-n, ... t-1)       for i in range(n_in, 0, -1):              cols.append(df.shift(i))       # forecast sequence (t, t+1, ... t+n)       for i in range(0, n_out):              cols.append(df.shift(-i))       # put it all together       agg = concat(cols, axis=1)       # drop rows with NaN values       if dropnan:              agg.dropna(inplace=True)       return agg.values # split a univariatedataset into train/test setsdeftrain_test_split(data, n_test):       return data[:-n_test, :], data[-n_test:,:] # fit an xgboost modeland make a one step predictiondef xgboost_forecast(train,testX):       # transform list into array       train = asarray(train)       # split into input and output columns       trainX, trainy = train[:, :-1], train[:,-1]       # fit model       model =XGBRegressor(objective='reg:squarederror', n_estimators=1000)       model.fit(trainX, trainy)       # make a one-step prediction       yhat = model.predict(asarray([testX]))       return yhat[0] # walk-forwardvalidation for univariate datadefwalk_forward_validation(data, n_test):       predictions = list()       # split dataset       train, test = train_test_split(data,n_test)       # seed history with training dataset       history = [x for x in train]       # step over each time-step in the testset       for i in range(len(test)):              # split test row into input andoutput columns              testX, testy = test[i, :-1],test[i, -1]              # fit model on history and make aprediction              yhat = xgboost_forecast(history,testX)              # store forecast in list ofpredictions              predictions.append(yhat)              # add actual observation tohistory for the next loop              history.append(test[i])              # summarize progress              print('>expected=%.1f,predicted=%.1f' % (testy, yhat))       # estimate prediction error       error = mean_absolute_error(test[:, 1],predictions)       return error, test[:, 1], predictions # load the datasetseries =read_csv('daily-total-female-births.csv', header=0, index_col=0)values = series.values# transform the timeseries data into supervised learningdata =series_to_supervised(values, n_in=3)# evaluatemae, y, yhat =walk_forward_validation(data, 12)print('MAE: %.3f' %mae)# plot expected vspreductedpyplot.plot(y,label='Expected')pyplot.plot(yhat,label='Predicted')pyplot.legend()pyplot.show()

# finalize model andmake a prediction for monthly births with xgboostfrom numpy importasarrayfrom pandas importread_csvfrom pandas importDataFramefrom pandas importconcatfrom xgboost importXGBRegressor # transform a timeseries dataset into a supervised learning datasetdefseries_to_supervised(data, n_in=1, n_out=1, dropnan=True):       n_vars = 1 if type(data) is list elsedata.shape[1]       df = DataFrame(data)       cols = list()       # input sequence (t-n, ... t-1)       for i in range(n_in, 0, -1):              cols.append(df.shift(i))       # forecast sequence (t, t+1, ... t+n)       for i in range(0, n_out):              cols.append(df.shift(-i))       # put it all together       agg = concat(cols, axis=1)       # drop rows with NaN values       if dropnan:              agg.dropna(inplace=True)       return agg.values # load the datasetseries =read_csv('daily-total-female-births.csv', header=0, index_col=0)values = series.values# transform the timeseries data into supervised learningtrain =series_to_supervised(values, n_in=3)# split into input andoutput columnstrainX, trainy =train[:, :-1], train[:, -1]# fit modelmodel =XGBRegressor(objective='reg:squarederror', n_estimators=1000)model.fit(trainX,trainy)# construct an inputfor a new preductionrow = values[-3:].flatten()# make a one-steppredictionyhat =model.predict(asarray([row]))print('Input: %s,Predicted: %.3f' % (row, yhat[0]))