- Nie, Y., Nguyen, N. H., Sinthong, P., & Kalagnanam, J. (2022). “A Time Series is Worth 64 Words: Long-term Forecasting with Transformers”
1. Backbone
Auxiliary Functions
source
get_activation_fn
source
Transpose
Transpose
Positional Encoding
source
positional_encoding
source
Coord1dPosEncoding
source
Coord2dPosEncoding
source
PositionalEncoding
Encoder
source
TSTEncoderLayer
TSTEncoderLayer
source
TSTEncoder
TSTEncoder
source
TSTiEncoder
TSTiEncoder
source
Flatten_Head
Flatten_Head
source
PatchTST_backbone
PatchTST_backbone
2. Model
source
PatchTST
*PatchTST The PatchTST model is an efficient Transformer-based model for multivariate time series forecasting. It is based on two key components: - segmentation of time series into windows (patches) which are served as input tokens to Transformer - channel-independence, where each channel contains a single univariate time series. Parameters:
h: int, Forecast horizon. input_size: int,
autorregresive inputs size, y=[1,2,3,4] input_size=2 ->
y_[t-2:t]=[1,2].stat_exog_list: str list, static exogenous
columns.hist_exog_list: str list, historic exogenous columns.futr_exog_list: str list, future exogenous columns.exclude_insample_y: bool=False, the model skips the autoregressive
features y[t-input_size:t] if True.encoder_layers: int, number
of layers for encoder.n_heads: int=16, number of multi-head’s
attention.hidden_size: int=128, units of embeddings and
encoders.linear_hidden_size: int=256, units of linear layer.dropout: float=0.1, dropout rate for residual connection.fc_dropout: float=0.1, dropout rate for linear layer.head_dropout: float=0.1, dropout rate for Flatten head layer.attn_dropout: float=0.1, dropout rate for attention layer.patch_len: int=32, length of patch. Note: patch_len = min(patch_len,
input_size + stride).stride: int=16, stride of patch.revin: bool=True, bool to use RevIn.revin_affine: bool=False,
bool to use affine in RevIn.revin_subtract_last: bool=False, bool
to use substract last in RevIn.activation: str=‘ReLU’, activation
from [‘gelu’,‘relu’].res_attention: bool=False, bool to use
residual attention.batch_normalization: bool=False, bool to use
batch normalization.learn_pos_embed: bool=True, bool to learn
positional embedding.loss: PyTorch module, instantiated train
loss class from losses
collection.valid_loss: PyTorch module=loss, instantiated valid loss class from
losses
collection.max_steps: int=1000, maximum number of training steps.learning_rate: float=1e-3, Learning rate between (0, 1).num_lr_decays: int=-1, Number of learning rate decays, evenly
distributed across max_steps.early_stop_patience_steps: int=-1,
Number of validation iterations before early stopping.val_check_steps: int=100, Number of training steps between every
validation loss check.batch_size: int=32, number of different
series in each batch.valid_batch_size: int=None, number of
different series in each validation and test batch, if None uses
batch_size.windows_batch_size: int=1024, number of windows to
sample in each training batch, default uses all.inference_windows_batch_size: int=1024, number of windows to sample in
each inference batch.start_padding_enabled: bool=False, if True,
the model will pad the time series with zeros at the beginning, by input
size.step_size: int=1, step size between each window of temporal
data.scaler_type: str=‘identity’, type of scaler for temporal
inputs normalization see temporal
scalers.random_seed: int, random_seed for pytorch initializer and numpy
generators.drop_last_loader: bool=False, if True
TimeSeriesDataLoader drops last non-full batch.alias: str,
optional, Custom name of the model.optimizer: Subclass of
‘torch.optim.Optimizer’, optional, user specified optimizer instead of
the default choice (Adam).optimizer_kwargs: dict, optional, list
of parameters used by the user specified optimizer.lr_scheduler: Subclass of ‘torch.optim.lr_scheduler.LRScheduler’,
optional, user specified lr_scheduler instead of the default choice
(StepLR).lr_scheduler_kwargs: dict, optional, list of parameters
used by the user specified lr_scheduler.dataloader_kwargs: dict, optional, list of parameters passed into the
PyTorch Lightning dataloader by the TimeSeriesDataLoader. **trainer_kwargs: int, keyword trainer arguments inherited from
PyTorch Lighning’s
trainer.References:
-Nie, Y., Nguyen, N. H., Sinthong, P., & Kalagnanam, J. (2022). “A Time Series is Worth 64 Words: Long-term Forecasting with Transformers”*
PatchTST.fit
*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.
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:dataset: NeuralForecast’s
TimeSeriesDataset,
see
documentation.val_size: int, validation size for temporal cross-validation.random_seed: int=None, random_seed for pytorch initializer and numpy
generators, overwrites model.__init__’s.test_size: int, test
size for temporal cross-validation.*
PatchTST.predict
*Predict. Neural network prediction with PL’s
Trainer execution of
predict_step.
Parameters:dataset: NeuralForecast’s
TimeSeriesDataset,
see
documentation.test_size: int=None, test size for temporal cross-validation.step_size: int=1, Step size between each window.random_seed:
int=None, random_seed for pytorch initializer and numpy generators,
overwrites model.__init__’s.quantiles: list of floats,
optional (default=None), target quantiles to predict. **data_module_kwargs: PL’s TimeSeriesDataModule args, see
documentation.*