724 KiB
724 KiB
Darts: Quickstart¶
https://unit8co.github.io/darts/quickstart/00-quickstart.html
In [29]:
%matplotlib inline
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from darts import TimeSeries
from darts.datasets import AirPassengersDataset
series = AirPassengersDataset().load()
Разделение данных¶
In [3]:
# по процентам
series1, series2 = series.split_before(0.75)
series1.plot()
series2.plot()
Out[3]:
In [4]:
# по дате
train, val = series.split_before(pd.Timestamp("19580101"))
train.plot(label="training")
val.plot(label="validation")
Out[4]:
Простая обработка¶
In [5]:
series_noise = TimeSeries.from_times_and_values(
series.time_index, np.random.randn(len(series))
)
(series / 2 + 20 * series_noise - 10).plot()
Out[5]:
In [6]:
series.map(np.log).plot()
Out[6]:
In [7]:
series.map(np.sqrt).stack(series).plot()
Out[7]:
In [28]:
series.map(lambda ts, x: x / ts.days_in_month).stack(series).plot()
Out[28]:
In [9]:
#Восстановление пропущенных значений
from darts.utils.missing_values import fill_missing_values, extract_subseries
values = np.arange(50, step=0.5)
values[10:30] = np.nan
values[60:95] = np.nan
series_ = TimeSeries.from_values(values)
fill_missing_values(series_).plot(label="without missing values")
(series_).plot(label="with missing values (shifted below)")
# extract_subseries(series_, mode="any") # выбирает существующие отрезки
Out[9]:
In [35]:
# проверка восстановления более сложного ВР -> восстановление значений линейно между последними значениями
series_ = series.copy()
values_ = series_.all_values(copy=False)
values_[10:30] = np.nan
values_[60:102] = np.nan
fill_missing_values(series_).plot(label="without missing values")
(series).plot(label="with missing values (shifted below)")
Out[35]:
Forecasting¶
In [10]:
#наивное прогнозирование
from darts.models import NaiveSeasonal
naive_model = NaiveSeasonal(K=1)
naive_model.fit(train)
naive_forecast = naive_model.predict(36)
#подбор помог сильно повысить точность благодаря сильно выраженной сезонности
naive_model_podbor = NaiveSeasonal(K=12)
naive_model_podbor.fit(train)
naive_forecast_podbor = naive_model_podbor.predict(36)
series.plot(label="actual")
naive_forecast.plot(label="naive forecast (K=1)")
naive_forecast_podbor.plot(label="naive forecast (K=12)")
Out[10]:
In [11]:
from darts.utils.statistics import plot_acf, check_seasonality
plot_acf(train, m=12, alpha=0.05, max_lag=24)
In [12]:
# проверка сезонности
for m in range(2, 25):
is_seasonal, period = check_seasonality(train, m=m, alpha=0.05)
if is_seasonal:
train_period = period
print("There is seasonality of order {}.".format(period))
In [13]:
# использование поиска сезонности для автоматической подстановки
seasonal_model = NaiveSeasonal(K=train_period)
seasonal_model.fit(train)
seasonal_forecast = seasonal_model.predict(36)
series.plot(label="actual")
seasonal_forecast.plot(label="naive forecast (K=12)")
Out[13]:
In [14]:
# Добавляем линейный тренд (drift)
from darts.models import NaiveDrift
drift_model = NaiveDrift()
drift_model.fit(train)
drift_forecast = drift_model.predict(36)
combined_forecast = drift_forecast + seasonal_forecast - train.last_value()
series.plot()
combined_forecast.plot(label="combined")
drift_forecast.plot(label="drift")
Out[14]:
Метрики качества¶
In [15]:
from darts.metrics import mape
print(
f"Mean absolute percentage error for the combined naive drift + seasonal: {mape(series, combined_forecast):.2f}%."
)
In [16]:
from darts.models import ExponentialSmoothing, TBATS, AutoARIMA, Theta
def eval_model(model):
model.fit(train)
forecast = model.predict(len(val))
print(f"model {model} obtains MAPE: {mape(val, forecast):.2f}%")
eval_model(ExponentialSmoothing())
eval_model(TBATS())
eval_model(AutoARIMA())
eval_model(Theta())
Theta method¶
Данный метод не рассматривался для экономии времени
Модели машинного обучения¶
Пример моделей:
- RegressionModel can wrap around any sklearn-compatible regression model to produce forecasts (it has its own section below).
- RNNModel is a flexible RNN implementation, which can be used like DeepAR.
- NBEATSModel implements the N-BEATS model.
- TFTModel implements the Temporal Fusion Transformer model.
- TCNModel implements temporal convolutional networks.
- …
In [17]:
from darts.datasets import AirPassengersDataset, MonthlyMilkDataset
series_air = AirPassengersDataset().load().astype(np.float32)
series_milk = MonthlyMilkDataset().load().astype(np.float32)
# set aside last 36 months of each series as validation set:
train_air, val_air = series_air[:-36], series_air[-36:]
train_milk, val_milk = series_milk[:-36], series_milk[-36:]
train_air.plot()
val_air.plot()
train_milk.plot()
val_milk.plot()
Out[17]:
In [21]:
# масштабирование
from darts.dataprocessing.transformers import Scaler
scaler = Scaler()
train_air_scaled, train_milk_scaled = scaler.fit_transform([train_air, train_milk])
train_air_scaled.plot()
train_milk_scaled.plot()
Out[21]:
Пример: N-BEATS¶
Нужные параметры для настройки
input_chunk_length: this is the “lookback window” of the model - i.e., how many time steps of history the neural network takes as input to produce its output in a forward pass.output_chunk_length: this is the “forward window” of the model - i.e., how many time steps of future values the neural network outputs in a forward pass.
In [22]:
from darts.models import NBEATSModel
model = NBEATSModel(input_chunk_length=24, output_chunk_length=12, random_state=42)
model.fit([train_air_scaled, train_milk_scaled], epochs=50, verbose=True);
In [23]:
pred_air, pred_milk = model.predict(series=[train_air_scaled, train_milk_scaled], n=36)
# scale back:
pred_air, pred_milk = scaler.inverse_transform([pred_air, pred_milk])
plt.figure(figsize=(10, 6))
series_air.plot(label="actual (air)")
series_milk.plot(label="actual (milk)")
pred_air.plot(label="forecast (air)")
pred_milk.plot(label="forecast (milk)")
Out[23]:
