七個(gè)最新的時(shí)間序列分析庫(kù)介紹和代碼示例-魔扣目錄

時(shí)間序列分析包括檢查隨著時(shí)間推移收集的數(shù)據(jù)點(diǎn)，目的是確定可以為未來(lái)預(yù)測(cè)提供信息的模式和趨勢(shì)。我們已經(jīng)介紹過(guò)很多個(gè)時(shí)間序列分析庫(kù)了，但是隨著時(shí)間推移，新的庫(kù)和更新也在不斷的出現(xiàn)，所以本文將分享8個(gè)目前比較常用的，用于處理時(shí)間序列問(wèn)題的Python/ target=_blank class=infotextkey>Python庫(kù)。他們是tsfresh, autots, darts, atspy, kats, sktime, greykite。

1、Tsfresh

Tsfresh在時(shí)間序列特征提取和選擇方面功能強(qiáng)大。它旨在自動(dòng)從時(shí)間序列數(shù)據(jù)中提取大量特征，并識(shí)別出最相關(guān)的特征。Tsfresh支持多種時(shí)間序列格式，可用于分類(lèi)、聚類(lèi)和回歸等各種應(yīng)用程序。

import pandas as pd
 from tsfresh import extract_features
 from tsfresh.utilities.dataframe_functions import make_forecasting_frame
 
 # Assume we have a time series dataset `data` with columns "time" and "value"
 data = pd.read_csv('data.csv')
 
 # We will use the last 10 points to predict the next point
 df_shift, y = make_forecasting_frame(data["value"], kind="value", max_timeshift=10, rolling_direction=1)
 
 # Extract relevant features using tsfresh
 X = extract_features(df_shift, column_id="id", column_sort="time", column_value="value", impute_function=impute)

2、AutoTS

autots是另一個(gè)用于時(shí)間序列預(yù)測(cè)的Python庫(kù)：

提供了單變量和多變量時(shí)間序列預(yù)測(cè)的各種算法，包括ARIMA, ETS, Prophet和DeepAR。
為最佳模型執(zhí)行自動(dòng)模型集成。
提供了上界和下界的置信區(qū)間預(yù)測(cè)。
通過(guò)學(xué)習(xí)最優(yōu)NaN imputation和異常值去除來(lái)處理數(shù)據(jù)。

from autots.datasets import load_monthly
 
 df_long = load_monthly(long=True)
 
 from autots import AutoTS
 
 model = AutoTS(
     forecast_length=3,
     frequency='infer',
     ensemble='simple',
     max_generations=5,
     num_validations=2,
 )
 model = model.fit(df_long, date_col='datetime', value_col='value', id_col='series_id')
 
 # Print the description of the best model
 print(model)

3、darts

darts(Data Analytics and Real-Time Systems)有多種時(shí)間序列預(yù)測(cè)模型，包括ARIMA、Prophet、指數(shù)平滑的各種變體，以及各種深度學(xué)習(xí)模型，如LSTMs、gru和tcn。Darts還具有用于交叉驗(yàn)證、超參數(shù)調(diào)優(yōu)和特征工程的內(nèi)置方法。

darts的一個(gè)關(guān)鍵特征是能夠進(jìn)行概率預(yù)測(cè)。這意味著，不僅可以為每個(gè)時(shí)間步驟生成單點(diǎn)預(yù)測(cè)，還可以生成可能結(jié)果的分布，從而更全面地理解預(yù)測(cè)中的不確定性。

import pandas as pd
 import matplotlib.pyplot as plt
 
 from darts import TimeSeries
 from darts.models import ExponentialSmoothing
 
 # Read data
 df = pd.read_csv("AirPassengers.csv", delimiter=",")
 
 # Create a TimeSeries, specifying the time and value columns
 series = TimeSeries.from_dataframe(df, "Month", "#Passengers")
 
 # Set aside the last 36 months as a validation series
 train, val = series[:-36], series[-36:]
 
 # Fit an exponential smoothing model, and make a (probabilistic)
 # prediction over the validation series’ duration
 model = ExponentialSmoothing()
 model.fit(train)
 prediction = model.predict(len(val), num_samples=1000)
 
 # Plot the median, 5th and 95th percentiles
 series.plot()
 prediction.plot(label="forecast", low_quantile=0.05, high_quantile=0.95)
 plt.legend()

4、AtsPy

atspy，可以簡(jiǎn)單地加載數(shù)據(jù)并指定要測(cè)試的模型，如下面的代碼所示。

# Importing packages
 import pandas as pd
 from atspy import AutomatedModel
 
 # Reading data
 df = pd.read_csv("AirPassengers.csv", delimiter=",")
 
 # Preprocessing data
 data.columns = ['month','Passengers']
 data['month'] = pd.to_datetime(data['month'],infer_datetime_format=True,format='%y%m')
 data.index = data.month
 df_air = data.drop(['month'], axis = 1)
 
 # Select the models you want to run:
 models = ['ARIMA','Prophet']
 run_models = AutomatedModel(df = df_air, model_list=models, forecast_len=10)

該包提供了一組完全自動(dòng)化的模型。包括:

5、kats

kats (kit to Analyze Time Series)是一個(gè)由Facebook(現(xiàn)在的Meta)開(kāi)發(fā)的Python庫(kù)。這個(gè)庫(kù)的三個(gè)核心特性是:

模型預(yù)測(cè):提供了一套完整的預(yù)測(cè)工具，包括10+個(gè)單獨(dú)的預(yù)測(cè)模型、集成、元學(xué)習(xí)模型、回溯測(cè)試、超參數(shù)調(diào)優(yōu)和經(jīng)驗(yàn)預(yù)測(cè)區(qū)間。

檢測(cè):Kats支持檢測(cè)時(shí)間序列數(shù)據(jù)中的各種模式的函數(shù)，包括季節(jié)性、異常、變化點(diǎn)和緩慢的趨勢(shì)變化。

特征提取和嵌入:Kats中的時(shí)間序列特征(TSFeature)提取模塊可以生成65個(gè)具有明確統(tǒng)計(jì)定義的特征，可應(yīng)用于大多數(shù)機(jī)器學(xué)習(xí)(ML)模型，如分類(lèi)和回歸。

# pip install kats
 
 import pandas as pd
 from kats.consts import TimeSeriesData
 from kats.models.prophet import ProphetModel, ProphetParams
 
 # Read data
 df = pd.read_csv("AirPassengers.csv", names=["time", "passengers"])
 
 # Convert to TimeSeriesData object
 air_passengers_ts = TimeSeriesData(air_passengers_df)
 
 # Create a model param instance
 params = ProphetParams(seasonality_mode='multiplicative')
 
 # Create a prophet model instance
 m = ProphetModel(air_passengers_ts, params)
 
 # Fit model simply by calling m.fit()
 m.fit()
 
 # Make prediction for next 30 month
 forecast = m.predict(steps=30, freq="MS")
 forecast.head()

6、Sktime

sktime是一個(gè)用于時(shí)間序列分析的庫(kù)，它構(gòu)建在scikit-learn之上，并遵循類(lèi)似的API，可以輕松地在兩個(gè)庫(kù)之間切換。下面是如何使用Sktime進(jìn)行時(shí)間序列分類(lèi)的示例:

from sktime.datasets import load_arrow_head
 from sktime.classification.compose import TimeSeriesForestClassifier
 from sktime.utils.sampling import train_test_split
 
 # Load ArrowHead dataset
 X, y = load_arrow_head(return_X_y=True)
 
 # Split data into train and test sets
 X_train, X_test, y_train, y_test = train_test_split(X, y)
 
 # Create and fit a time series forest classifier
 classifier = TimeSeriesForestClassifier(n_estimators=100)
 classifier.fit(X_train, y_train)
 
 # Predict labels for the test set
 y_pred = classifier.predict(X_test)
 
 # Print classification report
 from sklearn.metrics import classification_report
 print(classification_report(y_test, y_pred))

7、GreyKite

greykite是LinkedIn發(fā)布的一個(gè)時(shí)間序列預(yù)測(cè)庫(kù)。該庫(kù)可以處理復(fù)雜的時(shí)間序列數(shù)據(jù)，并提供一系列功能，包括自動(dòng)化特征工程、探索性數(shù)據(jù)分析、預(yù)測(cè)管道和模型調(diào)優(yōu)。

from greykite.common.data_loader import DataLoader
 from greykite.framework.templates.autogen.forecast_config import ForecastConfig
 from greykite.framework.templates.autogen.forecast_config import MetadataParam
 from greykite.framework.templates.forecaster import Forecaster
 from greykite.framework.templates.model_templates import ModelTemplateEnum
 
 # Defines inputs
 df = DataLoader().load_bikesharing().tail(24*90)  # Input time series (pandas.DataFrame)
 config = ForecastConfig(
      metadata_param=MetadataParam(time_col="ts", value_col="count"),  # Column names in `df`
      model_template=ModelTemplateEnum.AUTO.name,  # AUTO model configuration
      forecast_horizon=24,   # Forecasts 24 steps ahead
      coverage=0.95,         # 95% prediction intervals
  )
 
 # Creates forecasts
 forecaster = Forecaster()
 result = forecaster.run_forecast_config(df=df, config=config)
 
 # Accesses results
 result.forecast     # Forecast with metrics, diagnostics
 result.backtest     # Backtest with metrics, diagnostics
 result.grid_search  # Time series CV result
 result.model        # Trained model
 result.timeseries   # Processed time series with plotting functions

總結(jié)

我們可以看到，這些時(shí)間序列的庫(kù)主要功能有2個(gè)方向，一個(gè)是特征的生成，另外一個(gè)就是多種時(shí)間序列預(yù)測(cè)模型的集成，所以無(wú)論是處理單變量還是多變量數(shù)據(jù)，它們都可以滿足我們的需求，但是具體用那個(gè)還要看具體的需求和使用的習(xí)慣。