Save and load forecasters¶

Skforecast models can be easily saved and loaded from disk using the joblib library. Two handy functions, save_forecaster and load_forecaster are available to streamline this process. See below for a simple example.

A forecaster_id has been included when initializing the Forecaster, this may help to identify the target of the model.

✎ Note

Learn how to use forecaster models in production.

Libraries and data¶

In [1]:

# Libraries
# ==============================================================================
import numpy as np
import pandas as pd
from sklearn.ensemble import RandomForestRegressor

from skforecast.datasets import fetch_dataset
from skforecast.recursive import ForecasterRecursive
from skforecast.recursive import ForecasterRecursiveMultiSeries
from skforecast.utils import save_forecaster
from skforecast.utils import load_forecaster

# Libraries # ============================================================================== import numpy as np import pandas as pd from sklearn.ensemble import RandomForestRegressor from skforecast.datasets import fetch_dataset from skforecast.recursive import ForecasterRecursive from skforecast.recursive import ForecasterRecursiveMultiSeries from skforecast.utils import save_forecaster from skforecast.utils import load_forecaster

In [2]:

# Download data
# ==============================================================================
data = fetch_dataset(
    name="h2o", raw=True, kwargs_read_csv={"names": ["y", "date"], "header": 0}
)
data['date'] = pd.to_datetime(data['date'], format='%Y-%m-%d')
data = data.set_index('date')
data = data.asfreq('MS')

# Download data # ============================================================================== data = fetch_dataset( name="h2o", raw=True, kwargs_read_csv={"names": ["y", "date"], "header": 0} ) data['date'] = pd.to_datetime(data['date'], format='%Y-%m-%d') data = data.set_index('date') data = data.asfreq('MS')

h2o
---
Monthly expenditure ($AUD) on corticosteroid drugs that the Australian health
system had between 1991 and 2008.
Hyndman R (2023). fpp3: Data for Forecasting: Principles and Practice(3rd
Edition). http://pkg.robjhyndman.com/fpp3package/,https://github.com/robjhyndman
/fpp3package, http://OTexts.com/fpp3.
Shape of the dataset: (204, 2)

Save and load forecaster model¶

In [3]:

# Create and train forecaster
# ==============================================================================
forecaster = ForecasterRecursive(
                 regressor     = RandomForestRegressor(random_state=123),
                 lags          = 5,
                 forecaster_id = "forecaster_001"
             )

forecaster.fit(y=data['y'])
forecaster.predict(steps=3)

# Create and train forecaster # ============================================================================== forecaster = ForecasterRecursive( regressor = RandomForestRegressor(random_state=123), lags = 5, forecaster_id = "forecaster_001" ) forecaster.fit(y=data['y']) forecaster.predict(steps=3)

Out[3]:

2008-07-01    0.714526
2008-08-01    0.789144
2008-09-01    0.818433
Freq: MS, Name: pred, dtype: float64

In [4]:

# Save model
# ==============================================================================
save_forecaster(forecaster, file_name='forecaster_001.joblib', verbose=False)

# Save model # ============================================================================== save_forecaster(forecaster, file_name='forecaster_001.joblib', verbose=False)

In [5]:

# Load model
# ==============================================================================
forecaster_loaded = load_forecaster('forecaster_001.joblib', verbose=True)

# Load model # ============================================================================== forecaster_loaded = load_forecaster('forecaster_001.joblib', verbose=True)

=================== 
ForecasterRecursive 
=================== 
Regressor: RandomForestRegressor 
Lags: [1 2 3 4 5] 
Window features: None 
Window size: 5 
Exogenous included: False 
Exogenous names: None 
Transformer for y: None 
Transformer for exog: None 
Weight function included: False 
Differentiation order: None 
Training range: [Timestamp('1991-07-01 00:00:00'), Timestamp('2008-06-01 00:00:00')] 
Training index type: DatetimeIndex 
Training index frequency: MS 
Regressor parameters: 
    {'bootstrap': True, 'ccp_alpha': 0.0, 'criterion': 'squared_error', 'max_depth':
    None, 'max_features': 1.0, 'max_leaf_nodes': None, 'max_samples': None,
    'min_impurity_decrease': 0.0, 'min_samples_leaf': 1, 'min_samples_split': 2,
    'min_weight_fraction_leaf': 0.0, 'monotonic_cst': None, 'n_estimators': 100,
    'n_jobs': None, 'oob_score': False, 'random_state': 123, 'verbose': 0,
    'warm_start': False} 
fit_kwargs: {} 
Creation date: 2024-11-19 09:20:55 
Last fit date: 2024-11-19 09:20:55 
Skforecast version: 0.14.0 
Python version: 3.11.10 
Forecaster id: forecaster_001 

In [6]:

# Predict
# ==============================================================================
forecaster_loaded.predict(steps=3)

# Predict # ============================================================================== forecaster_loaded.predict(steps=3)

Out[6]:

2008-07-01    0.714526
2008-08-01    0.789144
2008-09-01    0.818433
Freq: MS, Name: pred, dtype: float64

In [7]:

# Forecaster identifier
# ==============================================================================
forecaster.forecaster_id

# Forecaster identifier # ============================================================================== forecaster.forecaster_id

Out[7]:

'forecaster_001'

Saving and Loading a Forecaster Model with Custom Features¶

Sometimes external objects are needed when creating a Forecaster. For example:

• Custom class to create window and custom features.

• A function to reduce the impact of some dates on the model, Weighted Time Series Forecasting.

For your code to work properly, these functions must be available in the environment where the Forecaster is loaded.

In [8]:

# Custom class to create rolling skewness features
# ==============================================================================
from scipy.stats import skew


class RollingSkewness():
    """
    Custom class to create rolling skewness features.
    """

    def __init__(self, window_sizes, features_names: list = 'rolling_skewness'):
        
        if not isinstance(window_sizes, list):
            window_sizes = [window_sizes]
        self.window_sizes = window_sizes
        self.features_names = features_names

    def transform_batch(self, X: pd.Series) -> pd.DataFrame:
        
        rolling_obj = X.rolling(window=self.window_sizes[0], center=False, closed='left')
        rolling_skewness = rolling_obj.skew()
        rolling_skewness = pd.DataFrame({
                               self.features_names: rolling_skewness
                           }).dropna()

        return rolling_skewness

    def transform(self, X: np.ndarray) -> np.ndarray:
        
        X = X[~np.isnan(X)]
        if len(X) > 0:
            rolling_skewness = np.array([skew(X, bias=False)])
        else:
            rolling_skewness = np.array([np.nan])
        
        return rolling_skewness

# Custom class to create rolling skewness features # ============================================================================== from scipy.stats import skew class RollingSkewness(): """ Custom class to create rolling skewness features. """ def __init__(self, window_sizes, features_names: list = 'rolling_skewness'): if not isinstance(window_sizes, list): window_sizes = [window_sizes] self.window_sizes = window_sizes self.features_names = features_names def transform_batch(self, X: pd.Series) -> pd.DataFrame: rolling_obj = X.rolling(window=self.window_sizes[0], center=False, closed='left') rolling_skewness = rolling_obj.skew() rolling_skewness = pd.DataFrame({ self.features_names: rolling_skewness }).dropna() return rolling_skewness def transform(self, X: np.ndarray) -> np.ndarray: X = X[~np.isnan(X)] if len(X) > 0: rolling_skewness = np.array([skew(X, bias=False)]) else: rolling_skewness = np.array([np.nan]) return rolling_skewness

In [9]:

# Custom function to create weights
# ==============================================================================
def custom_weights(index):
    """
    Return 0 if index is between 2004-01-01 and 2005-01-01.
    """
    weights = np.where(
                  (index >= '2004-01-01') & (index <= '2005-01-01'),
                   0,
                   1
              )

    return weights

# Custom function to create weights # ============================================================================== def custom_weights(index): """ Return 0 if index is between 2004-01-01 and 2005-01-01. """ weights = np.where( (index >= '2004-01-01') & (index <= '2005-01-01'), 0, 1 ) return weights

In [10]:

# Create and train forecaster
# ==============================================================================
window_features = RollingSkewness(window_sizes=3)

forecaster = ForecasterRecursive(
                 regressor       = RandomForestRegressor(random_state=123),
                 lags            = 3,
                 window_features = window_features,
                 weight_func     = custom_weights,
                 forecaster_id   = "forecaster_custom_features"
             )

forecaster.fit(y=data['y'])

# Create and train forecaster # ============================================================================== window_features = RollingSkewness(window_sizes=3) forecaster = ForecasterRecursive( regressor = RandomForestRegressor(random_state=123), lags = 3, window_features = window_features, weight_func = custom_weights, forecaster_id = "forecaster_custom_features" ) forecaster.fit(y=data['y'])

⚠ Warning

The save_forecaster function will save the functions used to create the weights as a module (custom_weights.py). But the classes used to create the window features will not be saved. Therefore, you must ensure that these classes are available in the environment where the Forecaster is loaded.

In [11]:

# Save model and custom function
# ==============================================================================
save_forecaster(
    forecaster, 
    file_name = 'forecaster_custom_features.joblib', 
    save_custom_functions = True, 
    verbose = False
)

# Save model and custom function # ============================================================================== save_forecaster( forecaster, file_name = 'forecaster_custom_features.joblib', save_custom_functions = True, verbose = False )

c:\Users\jaesc2\Miniconda3\envs\skforecast_py11_2\Lib\site-packages\skforecast\utils\utils.py:1898: SaveLoadSkforecastWarning: The Forecaster includes custom user-defined classes in the `window_features` argument. These classes are not saved automatically when saving the Forecaster. Please ensure you save these classes manually and import them before loading the Forecaster.
    Custom classes: RollingSkewness
Visit the documentation for more information: https://skforecast.org/latest/user_guides/save-load-forecaster.html#saving-and-loading-a-forecaster-model-with-custom-features 
 You can suppress this warning using: warnings.simplefilter('ignore', category=SaveLoadSkforecastWarning)
  warnings.warn(

At this point, the RollingSkewness class is manually saved in a file called rolling_skewness.py. This file must be available in the environment where the Forecaster is loaded.

In [12]:

# Load model and custom function
# ==============================================================================
from rolling_skewness import RollingSkewness  # This file has to be generated manually
from custom_weights import custom_weights

forecaster_loaded = load_forecaster('forecaster_custom_features.joblib', verbose=True)

# Load model and custom function # ============================================================================== from rolling_skewness import RollingSkewness # This file has to be generated manually from custom_weights import custom_weights forecaster_loaded = load_forecaster('forecaster_custom_features.joblib', verbose=True)

=================== 
ForecasterRecursive 
=================== 
Regressor: RandomForestRegressor 
Lags: [1 2 3] 
Window features: ['rolling_skewness'] 
Window size: 3 
Exogenous included: False 
Exogenous names: None 
Transformer for y: None 
Transformer for exog: None 
Weight function included: True 
Differentiation order: None 
Training range: [Timestamp('1991-07-01 00:00:00'), Timestamp('2008-06-01 00:00:00')] 
Training index type: DatetimeIndex 
Training index frequency: MS 
Regressor parameters: 
    {'bootstrap': True, 'ccp_alpha': 0.0, 'criterion': 'squared_error', 'max_depth':
    None, 'max_features': 1.0, 'max_leaf_nodes': None, 'max_samples': None,
    'min_impurity_decrease': 0.0, 'min_samples_leaf': 1, 'min_samples_split': 2,
    'min_weight_fraction_leaf': 0.0, 'monotonic_cst': None, 'n_estimators': 100,
    'n_jobs': None, 'oob_score': False, 'random_state': 123, 'verbose': 0,
    'warm_start': False} 
fit_kwargs: {} 
Creation date: 2024-11-19 09:20:55 
Last fit date: 2024-11-19 09:20:55 
Skforecast version: 0.14.0 
Python version: 3.11.10 
Forecaster id: forecaster_custom_features 

In [13]:

# Predict using loaded forecaster
# ==============================================================================
forecaster_loaded.predict(steps=5)

# Predict using loaded forecaster # ============================================================================== forecaster_loaded.predict(steps=5)

Out[13]:

2008-07-01    0.808125
2008-08-01    0.859447
2008-09-01    0.933751
2008-10-01    0.950768
2008-11-01    0.914137
Freq: MS, Name: pred, dtype: float64

ForecasterRecursiveMultiSeries¶

When using a ForecasterRecursiveMultiSeries, the save_forecaster function will save a different module for each of the functions used to create the weights.

In [14]:

# Data download
# ==============================================================================
data = fetch_dataset(name="items_sales")
data.head()

# Data download # ============================================================================== data = fetch_dataset(name="items_sales") data.head()

items_sales
-----------
Simulated time series for the sales of 3 different items.
Simulated data.
Shape of the dataset: (1097, 3)

Out[14]:

	item_1	item_2	item_3
date
2012-01-01	8.253175	21.047727	19.429739
2012-01-02	22.777826	26.578125	28.009863
2012-01-03	27.549099	31.751042	32.078922
2012-01-04	25.895533	24.567708	27.252276
2012-01-05	21.379238	18.191667	20.357737

In [15]:

# Custom function to create weights for each item
# ==============================================================================
def custom_weights_item_1(index):
    """
    Return 0 if index is between 2012-01-01 and 2012-06-01.
    """
    weights = np.where(
        (index >= '2012-01-01') & (index <= '2012-06-01'), 0, 1
    )

    return weights

def custom_weights_item_2(index):
    """
    Return 0 if index is between 2012-04-01 and 2013-01-01.
    """
    weights = np.where(
        (index >= '2012-04-01') & (index <= '2013-01-01'), 0, 1
    )

    return weights

def custom_weights_item_3(index):
    """
    Return 0 if index is between 2012-06-01 and 2013-01-01.
    """
    weights = np.where(
        (index >= '2012-06-01') & (index <= '2013-01-01'), 0, 1
    )

    return weights

# Custom function to create weights for each item # ============================================================================== def custom_weights_item_1(index): """ Return 0 if index is between 2012-01-01 and 2012-06-01. """ weights = np.where( (index >= '2012-01-01') & (index <= '2012-06-01'), 0, 1 ) return weights def custom_weights_item_2(index): """ Return 0 if index is between 2012-04-01 and 2013-01-01. """ weights = np.where( (index >= '2012-04-01') & (index <= '2013-01-01'), 0, 1 ) return weights def custom_weights_item_3(index): """ Return 0 if index is between 2012-06-01 and 2013-01-01. """ weights = np.where( (index >= '2012-06-01') & (index <= '2013-01-01'), 0, 1 ) return weights

In [16]:

# Custom class to create rolling skewness features (multi-series)
# ==============================================================================
from scipy.stats import skew


class RollingSkewnessMultiSeries():
    """
    Custom class to create rolling skewness features for multiple series.
    """

    def __init__(self, window_sizes, features_names: list = 'rolling_skewness'):
        
        if not isinstance(window_sizes, list):
            window_sizes = [window_sizes]
        self.window_sizes = window_sizes
        self.features_names = features_names

    def transform_batch(self, X: pd.Series) -> pd.DataFrame:
        
        rolling_obj = X.rolling(window=self.window_sizes[0], center=False, closed='left')
        rolling_skewness = rolling_obj.skew()
        rolling_skewness = pd.DataFrame({
                               self.features_names: rolling_skewness
                           }).dropna()

        return rolling_skewness

    def transform(self, X: np.ndarray) -> np.ndarray:
        
        X_dim = X.ndim
        if X_dim == 1:
            n_series = 1  # Only one series
            X = X.reshape(-1, 1)
        else:
            n_series = X.shape[1]  # Series (levels) to be predicted (present in last_window)
        
        n_stats = 1  # Only skewness is calculated
        rolling_skewness = np.full(
            shape=(n_series, n_stats), fill_value=np.nan, dtype=float
        )
        for i in range(n_series):
            if len(X) > 0:
                rolling_skewness[i, :] = skew(X[:, i], bias=False)
            else:
                rolling_skewness[i, :] = np.nan      

        if X_dim == 1:
            rolling_skewness = rolling_skewness.flatten()  
        
        return rolling_skewness

# Custom class to create rolling skewness features (multi-series) # ============================================================================== from scipy.stats import skew class RollingSkewnessMultiSeries(): """ Custom class to create rolling skewness features for multiple series. """ def __init__(self, window_sizes, features_names: list = 'rolling_skewness'): if not isinstance(window_sizes, list): window_sizes = [window_sizes] self.window_sizes = window_sizes self.features_names = features_names def transform_batch(self, X: pd.Series) -> pd.DataFrame: rolling_obj = X.rolling(window=self.window_sizes[0], center=False, closed='left') rolling_skewness = rolling_obj.skew() rolling_skewness = pd.DataFrame({ self.features_names: rolling_skewness }).dropna() return rolling_skewness def transform(self, X: np.ndarray) -> np.ndarray: X_dim = X.ndim if X_dim == 1: n_series = 1 # Only one series X = X.reshape(-1, 1) else: n_series = X.shape[1] # Series (levels) to be predicted (present in last_window) n_stats = 1 # Only skewness is calculated rolling_skewness = np.full( shape=(n_series, n_stats), fill_value=np.nan, dtype=float ) for i in range(n_series): if len(X) > 0: rolling_skewness[i, :] = skew(X[:, i], bias=False) else: rolling_skewness[i, :] = np.nan if X_dim == 1: rolling_skewness = rolling_skewness.flatten() return rolling_skewness

⚠ Warning

When weight_func is a dict and does not contain any of the series, for instance:

# Weights are not included for item_2
weight_func_dict = {
    'item_1': custom_weights_item_1,
    'item_3': custom_weights_item_3
}

You must create a function that returns all 1's as weights of that series.

def custom_weights_all_1(index):
    """
    Return 1 for all elements in the index.
    """
    weights = np.ones(len(index))
    return weights

# item_2 dummy weights
weight_func_dict = {
    'item_1': custom_weights_item_1,
    'item_2': custom_weights_all_1,
    'item_3': custom_weights_item_3
}

In [17]:

# Create and train ForecasterRecursiveMultiSeries
# ==============================================================================
window_features = RollingSkewnessMultiSeries(window_sizes=3)
weight_func_dict = {
    'item_1': custom_weights_item_1,
    'item_2': custom_weights_item_2,
    'item_3': custom_weights_item_3
}

forecaster = ForecasterRecursiveMultiSeries(
                 regressor       = RandomForestRegressor(random_state=123),
                 lags            = 3,
                 window_features = window_features,
                 encoding        = 'ordinal',
                 weight_func     = weight_func_dict
             )

forecaster.fit(series=data)
forecaster

# Create and train ForecasterRecursiveMultiSeries # ============================================================================== window_features = RollingSkewnessMultiSeries(window_sizes=3) weight_func_dict = { 'item_1': custom_weights_item_1, 'item_2': custom_weights_item_2, 'item_3': custom_weights_item_3 } forecaster = ForecasterRecursiveMultiSeries( regressor = RandomForestRegressor(random_state=123), lags = 3, window_features = window_features, encoding = 'ordinal', weight_func = weight_func_dict ) forecaster.fit(series=data) forecaster

Out[17]:

ForecasterRecursiveMultiSeries

General Information

• Regressor: RandomForestRegressor
• Lags: [1 2 3]
• Window features: ['rolling_skewness']
• Window size: 3
• Series encoding: ordinal
• Exogenous included: False
• Weight function included: True
• Series weights: None
• Differentiation order: None
• Creation date: 2024-11-19 09:20:56
• Last fit date: 2024-11-19 09:20:57
• Skforecast version: 0.14.0
• Python version: 3.11.10
• Forecaster id: None

Exogenous Variables

None

Data Transformations

• Transformer for series: None
• Transformer for exog: None

Training Information

• Series names (levels): item_1, item_2, item_3
• Training range: 'item_1': ['2012-01-01', '2015-01-01'], 'item_2': ['2012-01-01', '2015-01-01'], 'item_3': ['2012-01-01', '2015-01-01']
• Training index type: DatetimeIndex
• Training index frequency: D

Regressor Parameters

{'bootstrap': True, 'ccp_alpha': 0.0, 'criterion': 'squared_error', 'max_depth': None, 'max_features': 1.0, 'max_leaf_nodes': None, 'max_samples': None, 'min_impurity_decrease': 0.0, 'min_samples_leaf': 1, 'min_samples_split': 2, 'min_weight_fraction_leaf': 0.0, 'monotonic_cst': None, 'n_estimators': 100, 'n_jobs': None, 'oob_score': False, 'random_state': 123, 'verbose': 0, 'warm_start': False}

Fit Kwargs

{}

🛈 API Reference    🗎 User Guide

In [18]:

# Save model and custom function
# ==============================================================================
save_forecaster(
    forecaster, 
    file_name = 'forecaster_multiseries_custom_features.joblib', 
    save_custom_functions = True, 
    verbose = False
)

# Save model and custom function # ============================================================================== save_forecaster( forecaster, file_name = 'forecaster_multiseries_custom_features.joblib', save_custom_functions = True, verbose = False )

c:\Users\jaesc2\Miniconda3\envs\skforecast_py11_2\Lib\site-packages\skforecast\utils\utils.py:1898: SaveLoadSkforecastWarning: The Forecaster includes custom user-defined classes in the `window_features` argument. These classes are not saved automatically when saving the Forecaster. Please ensure you save these classes manually and import them before loading the Forecaster.
    Custom classes: RollingSkewnessMultiSeries
Visit the documentation for more information: https://skforecast.org/latest/user_guides/save-load-forecaster.html#saving-and-loading-a-forecaster-model-with-custom-features 
 You can suppress this warning using: warnings.simplefilter('ignore', category=SaveLoadSkforecastWarning)
  warnings.warn(

In [19]:

# Load model and custom function
# ==============================================================================
from rolling_skewness import RollingSkewnessMultiSeries  # This file has to be generated manually
from custom_weights_item_1 import custom_weights_item_1
from custom_weights_item_2 import custom_weights_item_2
from custom_weights_item_3 import custom_weights_item_3

forecaster_loaded = load_forecaster(
    'forecaster_multiseries_custom_features.joblib', verbose=True
)

# Load model and custom function # ============================================================================== from rolling_skewness import RollingSkewnessMultiSeries # This file has to be generated manually from custom_weights_item_1 import custom_weights_item_1 from custom_weights_item_2 import custom_weights_item_2 from custom_weights_item_3 import custom_weights_item_3 forecaster_loaded = load_forecaster( 'forecaster_multiseries_custom_features.joblib', verbose=True )

============================== 
ForecasterRecursiveMultiSeries 
============================== 
Regressor: RandomForestRegressor 
Lags: [1 2 3] 
Window features: ['rolling_skewness'] 
Window size: 3 
Series encoding: ordinal 
Series names (levels): item_1, item_2, item_3 
Exogenous included: False 
Exogenous names: None 
Transformer for series: None 
Transformer for exog: None 
Weight function included: True 
Series weights: None 
Differentiation order: None 
Training range: 
    'item_1': ['2012-01-01', '2015-01-01'], 'item_2': ['2012-01-01', '2015-01-01'],
    'item_3': ['2012-01-01', '2015-01-01'] 
Training index type: DatetimeIndex 
Training index frequency: D 
Regressor parameters: 
    {'bootstrap': True, 'ccp_alpha': 0.0, 'criterion': 'squared_error', 'max_depth':
    None, 'max_features': 1.0, 'max_leaf_nodes': None, 'max_samples': None,
    'min_impurity_decrease': 0.0, 'min_samples_leaf': 1, 'min_samples_split': 2,
    'min_weight_fraction_leaf': 0.0, 'monotonic_cst': None, 'n_estimators': 100,
    'n_jobs': None, 'oob_score': False, 'random_state': 123, 'verbose': 0,
    'warm_start': False} 
fit_kwargs: {} 
Creation date: 2024-11-19 09:20:56 
Last fit date: 2024-11-19 09:20:57 
Skforecast version: 0.14.0 
Python version: 3.11.10 
Forecaster id: None 

In [20]:

# Predict using loaded forecaster
# ==============================================================================
forecaster_loaded.predict(steps=5, levels=None)  # Predict all levels

# Predict using loaded forecaster # ============================================================================== forecaster_loaded.predict(steps=5, levels=None) # Predict all levels

Out[20]:

	item_1	item_2	item_3
2015-01-02	14.818313	17.954045	19.676498
2015-01-03	14.961743	17.530592	19.207165
2015-01-04	18.349711	17.792810	19.919855
2015-01-05	18.639790	18.447346	22.158983
2015-01-06	17.254107	19.599428	22.687187