Skip to content

Commit

Permalink
Fix/update doc (#2511)
Browse files Browse the repository at this point in the history 
* fix: change num_loader_workers to dataloader_kwargs

* fix: mention the fact that lags are shifted for each step when multi_models=False

* fix: croston does not support future covariates without conformal prediction, changed it from FutureCovLocalModel to LocalModel

* update changelog

* feat: new issue template for question about doc or features

* fix: change template name

* fix: change the description of the multi_models argument in all the regression models

* fix: execution count of the modified cell

* fix: reduce the number of diff in the notebook

* fix: revert changes

* fix: relevage new argument
  • Loading branch information
@madtoinou
madtoinou authored Sep 2, 2024
1 parent 3e893cd commit 26c5f39
Show file tree
Hide file tree
Showing 13 changed files with 61 additions and 63 deletions.
16 changes: 16 additions & 0 deletions .github/ISSUE_TEMPLATE/question.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
---
name: General question
about: Ask clarification about the documentation or a feature
title: "[QUESTION]"
labels: question, triage
assignees: ''

---

**Describe the issue linked to the documentation**
A detailed description of the issue you are facing. Please include any specific sections of/link to the documentation or methods you are struggling with.

If relevant, include a code snippet to describe your attempt.

**Additional context**
Add any other context about the problem here.
8 changes: 6 additions & 2 deletions CHANGELOG.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,9 +13,12 @@ but cannot always guarantee backwards compatibility. Changes that may **break co

- Added `IQRDetector`, that allows to detect anomalies using the interquartile range algorithm. [#2441] by [Igor Urbanik](https://github.com/u8-igor).
- Added hyperparameters controlling the hidden layer sizes for the feature encoders in `TiDEModel`. [#2408](https://github.com/unit8co/darts/issues/2408) by [eschibli](https://github.com/eschibli).
- Made README's forecasting model support table more colorblind-friendly. [#2433](https://github.com/unit8co/darts/pull/2433)
- Updated the Ray Tune Hyperparameter Optimization example in the [user guide](https://unit8co.github.io/darts/userguide/hyperparameter_optimization.html) to work with the latest `ray` versions (`>=2.31.0`). [#2459](https://github.com/unit8co/darts/pull/2459) by [He Weilin](https://github.com/cnhwl).
- Added support for broadcasting to TimeSeries on component and sample level. [#2476](https://https://github.com/unit8co/darts/pull/2476) by [Joel L.](https://github.com/Joelius300).
- Various improvements in the documentation:
- Made README's forecasting model support table more colorblind-friendly. [#2433](https://github.com/unit8co/darts/pull/2433)
- Updated the Ray Tune Hyperparameter Optimization example in the [user guide](https://unit8co.github.io/darts/userguide/hyperparameter_optimization.html) to work with the latest `ray` versions (`>=2.31.0`). [#2459](https://github.com/unit8co/darts/pull/2459) by [He Weilin](https://github.com/cnhwl).
- Indicate that `multi_models=False` induce a lags shift for each step in `output_chunk_length` in `RegressionModel` and `LinearRegressionModel`. [#2511](https://github.com/unit8co/darts/pull/2511) by [Antoine Madrona](https://github.com/madtoinou).
- Added reference to `timeseries_generation.datetime_attribute_timeseries` in `TimeSeries.add_datetime_attribute` (0-indexing of encoding is enforced). [#2511](https://github.com/unit8co/darts/pull/2511) by [Antoine Madrona](https://github.com/madtoinou).

**Fixed**

Expand All @@ -26,6 +29,7 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
- Fixed a bug when using `fit()` with a `RegressionModel` that uses an underlying `model` which does not support `sample_weight`. [#2445](https://github.com/unit8co/darts/pull/2445) by [He Weilin](https://github.com/cnhwl).
- Fixed a bug preventing TimeSeries to be divided by xarray or ndarray. [#2476](https://https://github.com/unit8co/darts/pull/2476) by [Joel L.](https://github.com/Joelius300).
- Fixed a bug when using `save()` and `load()` with a `RegressionEnsembleModel` that ensembles any `TorchForecastingModel`. [#2437](https://github.com/unit8co/darts/issues/2437) by [He Weilin](https://github.com/cnhwl).
- Fixed a bug with `CrostonModel`, which actually does not support future covariates. [#2511](https://github.com/unit8co/darts/pull/2511) by [Antoine Madrona](https://github.com/madtoinou).

**Dependencies**

Expand Down
2 changes: 1 addition & 1 deletion README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -239,7 +239,7 @@ on bringing more models and features.
| [Prophet](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.prophet_model.html#darts.models.forecasting.prophet_model.Prophet) | [Prophet repo](https://github.com/facebook/prophet) | ✅ 🔴 | 🔴 ✅ 🔴 | ✅ 🔴 | 🔴 |
| [FFT](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.fft.html#darts.models.forecasting.fft.FFT) (Fast Fourier Transform) | | ✅ 🔴 | 🔴 🔴 🔴 | 🔴 🔴 | 🔴 |
| [KalmanForecaster](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.kalman_forecaster.html#darts.models.forecasting.kalman_forecaster.KalmanForecaster) using the Kalman filter and N4SID for system identification | [N4SID paper](https://people.duke.edu/~hpgavin/SystemID/References/VanOverschee-Automatica-1994.pdf) | ✅ ✅ | 🔴 ✅ 🔴 | ✅ 🔴 | 🔴 |
| [Croston](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.croston.html#darts.models.forecasting.croston.Croston) method | | ✅ 🔴 | 🔴 ✅ 🔴 | 🔴 🔴 | 🔴 |
| [Croston](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.croston.html#darts.models.forecasting.croston.Croston) method | | ✅ 🔴 | 🔴 🔴 🔴 | 🔴 🔴 | 🔴 |
| **Global Baseline Models**<br/>([GlobalForecastingModel](https://unit8co.github.io/darts/userguide/covariates.html#global-forecasting-models-gfms)) | | | | | |
| [GlobalNaiveAggregate](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.global_baseline_models.html#darts.models.forecasting.global_baseline_models.GlobalNaiveAggregate) | | ✅ ✅ | 🔴 🔴 🔴 | 🔴 🔴 | ✅ |
| [GlobalNaiveDrift](https://unit8co.github.io/darts/generated_api/darts.models.forecasting.global_baseline_models.html#darts.models.forecasting.global_baseline_models.GlobalNaiveDrift) | | ✅ ✅ | 🔴 🔴 🔴 | 🔴 🔴 | ✅ |
Expand Down
5 changes: 3 additions & 2 deletions darts/models/forecasting/catboost_model.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -122,8 +122,9 @@ def encode_year(idx):
Control the randomness in the fitting procedure and for sampling.
Default: ``None``.
multi_models
If True, a separate model will be trained for each future lag to predict. If False, a single model is
trained to predict at step 'output_chunk_length' in the future. Default: True.
If True, a separate model will be trained for each future lag to predict. If False, a single model
is trained to predict all the steps in 'output_chunk_length' (features lags are shifted back by
`output_chunk_length - n` for each step `n`). Default: True.
use_static_covariates
Whether the model should use static covariate information in case the input `series` passed to ``fit()``
contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
Expand Down
56 changes: 11 additions & 45 deletions darts/models/forecasting/croston.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,25 +3,22 @@
--------------
"""

from typing import Optional

from statsforecast.models import TSB as CrostonTSB
from statsforecast.models import CrostonClassic, CrostonOptimized, CrostonSBA

from darts.logging import raise_if, raise_if_not
from darts.models.forecasting.forecasting_model import (
FutureCovariatesLocalForecastingModel,
LocalForecastingModel,
)
from darts.timeseries import TimeSeries


class Croston(FutureCovariatesLocalForecastingModel):
class Croston(LocalForecastingModel):
def __init__(
self,
version: str = "classic",
alpha_d: float = None,
alpha_p: float = None,
add_encoders: Optional[dict] = None,
):
"""An implementation of the `Croston method
<https://otexts.com/fpp3/counts.html>`_ for intermittent
Expand All @@ -46,30 +43,6 @@ def __init__(
For the "tsb" version, the alpha smoothing parameter to apply on demand.
alpha_p
For the "tsb" version, the alpha smoothing parameter to apply on probability.
add_encoders
A large number of future covariates can be automatically generated with `add_encoders`.
This can be done by adding multiple pre-defined index encoders and/or custom user-made functions that
will be used as index encoders. Additionally, a transformer such as Darts' :class:`Scaler` can be added to
transform the generated covariates. This happens all under one hood and only needs to be specified at
model creation.
Read :meth:`SequentialEncoder <darts.dataprocessing.encoders.SequentialEncoder>` to find out more about
``add_encoders``. Default: ``None``. An example showing some of ``add_encoders`` features:
.. highlight:: python
.. code-block:: python
def encode_year(idx):
return (idx.year - 1950) / 50
add_encoders={
'cyclic': {'future': ['month']},
'datetime_attribute': {'future': ['hour', 'dayofweek']},
'position': {'future': ['relative']},
'custom': {'future': [encode_year]},
'transformer': Scaler(),
'tz': 'CET'
}
..
References
----------
Expand All @@ -96,7 +69,7 @@ def encode_year(idx):
[461.7666],
[461.7666]])
"""
super().__init__(add_encoders=add_encoders)
super().__init__(add_encoders=None)
raise_if_not(
version.lower() in ["classic", "optimized", "sba", "tsb"],
'The provided "version" parameter must be set to "classic", "optimized", "sba" or "tsb".',
Expand All @@ -123,37 +96,30 @@ def encode_year(idx):
def supports_multivariate(self) -> bool:
return False

def _fit(self, series: TimeSeries, future_covariates: Optional[TimeSeries] = None):
super()._fit(series, future_covariates)
def fit(self, series: TimeSeries):
super().fit(series)
self._assert_univariate(series)
series = self.training_series

self.model.fit(
y=series.values(copy=False).flatten(),
X=(
future_covariates.values(copy=False).flatten()
if future_covariates is not None
else None
),
# X can be used to passe future covariates only when conformal prediction is used
X=None,
)

return self

def _predict(
def predict(
self,
n: int,
future_covariates: Optional[TimeSeries] = None,
num_samples: int = 1,
verbose: bool = False,
):
super()._predict(n, future_covariates, num_samples)
super().predict(n, num_samples)
values = self.model.predict(
h=n,
X=(
future_covariates.values(copy=False).flatten()
if future_covariates is not None
else None
),
# X can be used to passe future covariates only when conformal prediction is used
X=None,
)["mean"]
return self._build_forecast_series(values)

Expand Down
5 changes: 3 additions & 2 deletions darts/models/forecasting/lgbm.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -130,8 +130,9 @@ def encode_year(idx):
Control the randomness in the fitting procedure and for sampling.
Default: ``None``.
multi_models
If True, a separate model will be trained for each future lag to predict. If False, a single model is
trained to predict at step 'output_chunk_length' in the future. Default: True.
If True, a separate model will be trained for each future lag to predict. If False, a single model
is trained to predict all the steps in 'output_chunk_length' (features lags are shifted back by
`output_chunk_length - n` for each step `n`). Default: True.
use_static_covariates
Whether the model should use static covariate information in case the input `series` passed to ``fit()``
contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
Expand Down
5 changes: 3 additions & 2 deletions darts/models/forecasting/linear_regression_model.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -128,8 +128,9 @@ def encode_year(idx):
no `likelihood` is set.
Default: ``None``.
multi_models
If True, a separate model will be trained for each future lag to predict. If False, a single model is
trained to predict at step 'output_chunk_length' in the future. Default: True.
If True, a separate model will be trained for each future lag to predict. If False, a single model
is trained to predict all the steps in 'output_chunk_length' (features lags are shifted back by
`output_chunk_length - n` for each step `n`). Default: True.
use_static_covariates
Whether the model should use static covariate information in case the input `series` passed to ``fit()``
contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
Expand Down
5 changes: 3 additions & 2 deletions darts/models/forecasting/random_forest.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -125,8 +125,9 @@ def encode_year(idx):
The maximum depth of the tree. If None, then nodes are expanded until all leaves are pure or until all
leaves contain less than min_samples_split samples.
multi_models
If True, a separate model will be trained for each future lag to predict. If False, a single model is
trained to predict at step 'output_chunk_length' in the future. Default: True.
If True, a separate model will be trained for each future lag to predict. If False, a single model
is trained to predict all the steps in 'output_chunk_length' (features lags are shifted back by
`output_chunk_length - n` for each step `n`). Default: True.
use_static_covariates
Whether the model should use static covariate information in case the input `series` passed to ``fit()``
contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
Expand Down
5 changes: 3 additions & 2 deletions darts/models/forecasting/regression_model.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -161,8 +161,9 @@ def encode_year(idx):
will be used per component in the multivariate series.
If None, defaults to: ``sklearn.linear_model.LinearRegression(n_jobs=-1)``.
multi_models
If True, a separate model will be trained for each future lag to predict. If False, a single model is
trained to predict at step 'output_chunk_length' in the future. Default: True.
If True, a separate model will be trained for each future lag to predict. If False, a single model
is trained to predict all the steps in 'output_chunk_length' (features lags are shifted back by
`output_chunk_length - n` for each step `n`). Default: True.
use_static_covariates
Whether the model should use static covariate information in case the input `series` passed to ``fit()``
contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
Expand Down
5 changes: 3 additions & 2 deletions darts/models/forecasting/xgboost.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -148,8 +148,9 @@ def encode_year(idx):
Control the randomness in the fitting procedure and for sampling.
Default: ``None``.
multi_models
If True, a separate model will be trained for each future lag to predict. If False, a single model is
trained to predict at step 'output_chunk_length' in the future. Default: True.
If True, a separate model will be trained for each future lag to predict. If False, a single model
is trained to predict all the steps in 'output_chunk_length' (features lags are shifted back by
`output_chunk_length - n` for each step `n`). Default: True.
use_static_covariates
Whether the model should use static covariate information in case the input `series` passed to ``fit()``
contain static covariates. If ``True``, and static covariates are available at fitting time, will enforce
Expand Down
6 changes: 6 additions & 0 deletions darts/timeseries.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3249,6 +3249,12 @@ def add_datetime_attribute(
This works only for deterministic time series (i.e., made of 1 sample).
Notes
-----
0-indexing is enforced across all the encodings, see
:meth:`datetime_attribute_timeseries() <darts.utils.timeseries_generation.datetime_attribute_timeseries>`
for more information.
Parameters
----------
attribute
Expand Down
Loading

0 comments on commit 26c5f39

Please sign in to comment.