> ## Documentation Index
> Fetch the complete documentation index at: https://nixtla-old-docs.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# BiTCN

Bidirectional Temporal Convolutional Network (BiTCN) is a forecasting
architecture based on two temporal convolutional networks (TCNs). The
first network (‘forward’) encodes future covariates of the time series,
whereas the second network (‘backward’) encodes past observations and
covariates. This method allows to preserve the temporal information of
sequence data, and is computationally more efficient than common RNN
methods (LSTM, GRU, …). As compared to Transformer-based methods, BiTCN
has a lower space complexity, i.e. it requires orders of magnitude less
parameters.

This model may be a good choice if you seek a small model (small amount
of trainable parameters) with few hyperparameters to tune (only 2).

**References**<br /> -[Olivier Sprangers, Sebastian Schelter, Maarten de
Rijke (2023). Parameter-Efficient Deep Probabilistic Forecasting.
International Journal of Forecasting 39, no. 1 (1 January 2023): 332–45.
URL:
https://doi.org/10.1016/j.ijforecast.2021.11.011.](https://doi.org/10.1016/j.ijforecast.2021.11.011)<br />
-[Shaojie Bai, Zico Kolter, Vladlen Koltun. (2018). An Empirical
Evaluation of Generic Convolutional and Recurrent Networks for Sequence
Modeling. Computing Research Repository, abs/1803.01271. URL:
https://arxiv.org/abs/1803.01271.](https://arxiv.org/abs/1803.01271)<br />
-[van den Oord, A., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O.,
Graves, A., Kalchbrenner, N., Senior, A. W., & Kavukcuoglu, K. (2016).
Wavenet: A generative model for raw audio. Computing Research
Repository, abs/1609.03499. URL: http://arxiv.org/abs/1609.03499.
arXiv:1609.03499.](https://arxiv.org/abs/1609.03499)<br />

<figure>
  <img src="https://mintcdn.com/nixtla-old-docs/x1r-gXtjoZEzq4L2/neuralforecast/imgs_models/bitcn.png?fit=max&auto=format&n=x1r-gXtjoZEzq4L2&q=85&s=c51d2901678d007d30bc190dd1caebb4" alt="Figure 1. Visualization of a stack of dilated causal convolutional layers." width="1024" height="358" data-path="neuralforecast/imgs_models/bitcn.png" />

  <figcaption aria-hidden="true">Figure 1. Visualization of a stack of
  dilated causal convolutional layers.</figcaption>
</figure>

## 1. Auxiliary Functions

***

<a href="https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/bitcn.py#L57" target="_blank" style={{ float: "right", fontSize: "smaller" }}>source</a>

### TCNCell

> ```text theme={null}
>  TCNCell (in_channels, out_channels, kernel_size, padding, dilation, mode,
>           groups, dropout)
> ```

*Temporal Convolutional Network Cell, consisting of CustomConv1D
modules.*

***

<a href="https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/bitcn.py#L18" target="_blank" style={{ float: "right", fontSize: "smaller" }}>source</a>

### CustomConv1d

> ```text theme={null}
>  CustomConv1d (in_channels, out_channels, kernel_size, padding=0,
>                dilation=1, mode='backward', groups=1)
> ```

*Forward- and backward looking Conv1D*

## 2. BiTCN

***

<a href="https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/bitcn.py#L87" target="_blank" style={{ float: "right", fontSize: "smaller" }}>source</a>

### BiTCN

> ```text theme={null}
>  BiTCN (h:int, input_size:int, hidden_size:int=16, dropout:float=0.5,
>         futr_exog_list=None, hist_exog_list=None, stat_exog_list=None,
>         exclude_insample_y=False, loss=MAE(), valid_loss=None,
>         max_steps:int=1000, learning_rate:float=0.001,
>         num_lr_decays:int=-1, early_stop_patience_steps:int=-1,
>         val_check_steps:int=100, batch_size:int=32,
>         valid_batch_size:Optional[int]=None, windows_batch_size=1024,
>         inference_windows_batch_size=1024, start_padding_enabled=False,
>         step_size:int=1, scaler_type:str='identity', random_seed:int=1,
>         drop_last_loader:bool=False, alias:Optional[str]=None,
>         optimizer=None, optimizer_kwargs=None, lr_scheduler=None,
>         lr_scheduler_kwargs=None, dataloader_kwargs=None,
>         **trainer_kwargs)
> ```

\*BiTCN

Bidirectional Temporal Convolutional Network (BiTCN) is a forecasting
architecture based on two temporal convolutional networks (TCNs). The
first network (‘forward’) encodes future covariates of the time series,
whereas the second network (‘backward’) encodes past observations and
covariates. This is a univariate model.

**Parameters:**<br /> `h`: int, forecast horizon.<br /> `input_size`: int,
considered autorregresive inputs (lags), y=\[1,2,3,4] input\_size=2 ->
lags=\[1,2].<br /> `hidden_size`: int=16, units for the TCN’s hidden
state size.<br /> `dropout`: float=0.1, dropout rate used for the dropout
layers throughout the architecture.<br /> `futr_exog_list`: str list,
future exogenous columns.<br /> `hist_exog_list`: str list, historic
exogenous columns.<br /> `stat_exog_list`: str list, static exogenous
columns.<br /> `exclude_insample_y`: bool=False, the model skips the
autoregressive features y\[t-input\_size:t] if True.<br /> `loss`: PyTorch
module, instantiated train loss class from [losses
collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).<br />
`valid_loss`: PyTorch module=`loss`, instantiated valid loss class from
[losses
collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).<br />
`max_steps`: int=1000, maximum number of training steps.<br />
`learning_rate`: float=1e-3, Learning rate between (0, 1).<br />
`num_lr_decays`: int=-1, Number of learning rate decays, evenly
distributed across max\_steps.<br /> `early_stop_patience_steps`: int=-1,
Number of validation iterations before early stopping.<br />
`val_check_steps`: int=100, Number of training steps between every
validation loss check.<br /> `batch_size`: int=32, number of different
series in each batch.<br /> `valid_batch_size`: int=None, number of
different series in each validation and test batch, if None uses
batch\_size.<br /> `windows_batch_size`: int=1024, number of windows to
sample in each training batch, default uses all.<br />
`inference_windows_batch_size`: int=1024, number of windows to sample in
each inference batch, -1 uses all.<br /> `start_padding_enabled`:
bool=False, if True, the model will pad the time series with zeros at
the beginning, by input size.<br /> `step_size`: int=1, step size between
each window of temporal data.<br /> `scaler_type`: str=‘identity’, type of
scaler for temporal inputs normalization see [temporal
scalers](https://nixtla.github.io/neuralforecast/common.scalers.html).<br />
`random_seed`: int=1, random\_seed for pytorch initializer and numpy
generators.<br /> `drop_last_loader`: bool=False, if True
`TimeSeriesDataLoader` drops last non-full batch.<br /> `alias`: str,
optional, Custom name of the model.<br /> `optimizer`: Subclass of
‘torch.optim.Optimizer’, optional, user specified optimizer instead of
the default choice (Adam).<br /> `optimizer_kwargs`: dict, optional, list
of parameters used by the user specified `optimizer`.<br />
`lr_scheduler`: Subclass of ‘torch.optim.lr\_scheduler.LRScheduler’,
optional, user specified lr\_scheduler instead of the default choice
(StepLR).<br /> `lr_scheduler_kwargs`: dict, optional, list of parameters
used by the user specified `lr_scheduler`.<br /> `dataloader_kwargs`:
dict, optional, list of parameters passed into the PyTorch Lightning
dataloader by the `TimeSeriesDataLoader`. <br /> `**trainer_kwargs`: int,
keyword trainer arguments inherited from [PyTorch Lighning’s
trainer](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.trainer.trainer.Trainer.html?highlight=trainer).<br />

**References**<br />

* [Olivier Sprangers, Sebastian Schelter, Maarten de Rijke (2023).
  Parameter-Efficient Deep Probabilistic Forecasting. International
  Journal of Forecasting 39, no. 1 (1 January 2023): 332–45. URL:
  https://doi.org/10.1016/j.ijforecast.2021.11.011.](https://doi.org/10.1016/j.ijforecast.2021.11.011)<br />\*

***

### BiTCN.fit

> ```text theme={null}
>  BiTCN.fit (dataset, val_size=0, test_size=0, random_seed=None,
>             distributed_config=None)
> ```

\*Fit.

The `fit` method, optimizes the neural network’s weights using the
initialization parameters (`learning_rate`, `windows_batch_size`, …) and
the `loss` function as defined during the initialization. Within `fit`
we use a PyTorch Lightning `Trainer` that inherits the initialization’s
`self.trainer_kwargs`, to customize its inputs, see [PL’s trainer
arguments](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.trainer.trainer.Trainer.html?highlight=trainer).

The method is designed to be compatible with SKLearn-like classes and in
particular to be compatible with the StatsForecast library.

By default the `model` is not saving training checkpoints to protect
disk memory, to get them change `enable_checkpointing=True` in
`__init__`.

**Parameters:**<br /> `dataset`: NeuralForecast’s
[`TimeSeriesDataset`](https://nixtlaverse.nixtla.io/neuralforecast/tsdataset.html#timeseriesdataset),
see
[documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).<br />
`val_size`: int, validation size for temporal cross-validation.<br />
`random_seed`: int=None, random\_seed for pytorch initializer and numpy
generators, overwrites model.\_\_init\_\_’s.<br /> `test_size`: int, test
size for temporal cross-validation.<br />\*

***

### BiTCN.predict

> ```text theme={null}
>  BiTCN.predict (dataset, test_size=None, step_size=1, random_seed=None,
>                 quantiles=None, **data_module_kwargs)
> ```

\*Predict.

Neural network prediction with PL’s `Trainer` execution of
`predict_step`.

**Parameters:**<br /> `dataset`: NeuralForecast’s
[`TimeSeriesDataset`](https://nixtlaverse.nixtla.io/neuralforecast/tsdataset.html#timeseriesdataset),
see
[documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).<br />
`test_size`: int=None, test size for temporal cross-validation.<br />
`step_size`: int=1, Step size between each window.<br /> `random_seed`:
int=None, random\_seed for pytorch initializer and numpy generators,
overwrites model.\_\_init\_\_’s.<br /> `quantiles`: list of floats,
optional (default=None), target quantiles to predict. <br />
`**data_module_kwargs`: PL’s TimeSeriesDataModule args, see
[documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule).\*

## Usage Example

```python theme={null}
import pandas as pd
import matplotlib.pyplot as plt

from neuralforecast import NeuralForecast
from neuralforecast.losses.pytorch import GMM
from neuralforecast.models import BiTCN
from neuralforecast.utils import AirPassengersPanel, AirPassengersStatic

Y_train_df = AirPassengersPanel[AirPassengersPanel.ds<AirPassengersPanel['ds'].values[-12]] # 132 train
Y_test_df = AirPassengersPanel[AirPassengersPanel.ds>=AirPassengersPanel['ds'].values[-12]].reset_index(drop=True) # 12 test

fcst = NeuralForecast(
    models=[
            BiTCN(h=12,
                input_size=24,
                loss=GMM(n_components=7, level=[80,90]),
                max_steps=100,
                scaler_type='standard',
                futr_exog_list=['y_[lag12]'],
                hist_exog_list=None,
                stat_exog_list=['airline1'],
                windows_batch_size=2048,
                val_check_steps=10,
                early_stop_patience_steps=-1,
                ),     
    ],
    freq='ME'
)
fcst.fit(df=Y_train_df, static_df=AirPassengersStatic)
forecasts = fcst.predict(futr_df=Y_test_df)

# Plot quantile predictions
Y_hat_df = forecasts.reset_index(drop=False).drop(columns=['unique_id','ds'])
plot_df = pd.concat([Y_test_df, Y_hat_df], axis=1)
plot_df = pd.concat([Y_train_df, plot_df])

plot_df = plot_df[plot_df.unique_id=='Airline1'].drop('unique_id', axis=1)
plt.plot(plot_df['ds'], plot_df['y'], c='black', label='True')
plt.plot(plot_df['ds'], plot_df['BiTCN-median'], c='blue', label='median')
plt.fill_between(x=plot_df['ds'][-12:], 
                 y1=plot_df['BiTCN-lo-90'][-12:].values,
                 y2=plot_df['BiTCN-hi-90'][-12:].values,
                 alpha=0.4, label='level 90')
plt.legend()
plt.grid()
```