sub_modules

Implementation of nn.Modules for TimeXer model.

Classes

AttentionLayer(attention, d_model, n_heads)	Attention layer that combines query, key, and value projections with an attention mechanism.
DataEmbedding_inverted(c_in, d_model[, ...])	Data embedding module for time series data.
EnEmbedding(n_vars, d_model, patch_len, dropout)	Encoder embedding module for time series data.
Encoder(layers[, norm_layer, projection])	Encoder module for the TimeXer model.
EncoderLayer(self_attention, ...[, d_ff, ...])	Encoder layer for the TimeXer model.
FlattenHead(n_vars, nf, target_window[, ...])	Flatten head for the output of the model.
FullAttention([mask_flag, factor, scale, ...])	Full attention mechanism with optional masking and dropout. :param mask_flag: Whether to apply masking. :type mask_flag: bool :param factor: Factor for scaling the attention scores. :type factor: int :param scale: Scaling factor for attention scores. :type scale: float :param attention_dropout: Dropout rate for attention scores. :type attention_dropout: float :param output_attention: Whether to output attention weights. :type output_attention: bool :param use_efficient_attention: Whether to use PyTorch's native, optimized Scaled Dot Product Attention implementation which can reduce computation time and memory consumption for longer sequences. PyTorch automatically selects the optimal backend (FlashAttention-2, Memory-Efficient Attention, or their own C++ implementation) based on user's input properties, hardware capabilities, and build configuration. :type use_efficient_attention: bool.
PositionalEmbedding(d_model[, max_len])	Positional embedding module for time series data.
TriangularCausalMask(B, L[, device])	Triangular causal mask for attention mechanism.