|
|
| from fla.layers import ( |
| ABCAttention, |
| Attention, |
| BasedLinearAttention, |
| BitAttention, |
| Comba, |
| DeltaFormerAttention, |
| DeltaNet, |
| GatedDeltaNet, |
| GatedDeltaProduct, |
| GatedLinearAttention, |
| GatedSlotAttention, |
| HGRN2Attention, |
| HGRNAttention, |
| LightNetAttention, |
| LinearAttention, |
| LogLinearMamba2, |
| MesaNet, |
| MomAttention, |
| MultiheadLatentAttention, |
| MultiScaleRetention, |
| NativeSparseAttention, |
| PaTHAttention, |
| ReBasedLinearAttention, |
| RodimusAttention, |
| RWKV6Attention, |
| RWKV7Attention, |
| ) |
| from fla.models import ( |
| ABCForCausalLM, |
| ABCModel, |
| BitNetForCausalLM, |
| BitNetModel, |
| CombaForCausalLM, |
| CombaModel, |
| DeltaFormerForCausalLM, |
| DeltaFormerModel, |
| DeltaNetForCausalLM, |
| DeltaNetModel, |
| GatedDeltaNetForCausalLM, |
| GatedDeltaNetModel, |
| GatedDeltaProductForCausalLM, |
| GatedDeltaProductModel, |
| GLAForCausalLM, |
| GLAModel, |
| GSAForCausalLM, |
| GSAModel, |
| HGRN2ForCausalLM, |
| HGRN2Model, |
| HGRNForCausalLM, |
| HGRNModel, |
| LightNetForCausalLM, |
| LightNetModel, |
| LinearAttentionForCausalLM, |
| LinearAttentionModel, |
| LogLinearMamba2ForCausalLM, |
| LogLinearMamba2Model, |
| MesaNetForCausalLM, |
| MesaNetModel, |
| MLAForCausalLM, |
| MLAModel, |
| MomForCausalLM, |
| MomModel, |
| NSAForCausalLM, |
| NSAModel, |
| PaTHAttentionForCausalLM, |
| PaTHAttentionModel, |
| RetNetForCausalLM, |
| RetNetModel, |
| RodimusForCausalLM, |
| RodimusModel, |
| RWKV6ForCausalLM, |
| RWKV6Model, |
| RWKV7ForCausalLM, |
| RWKV7Model, |
| TransformerForCausalLM, |
| TransformerModel, |
| ) |
|
|
| __all__ = [ |
| 'ABCAttention', 'ABCForCausalLM', 'ABCModel', |
| 'Attention', 'TransformerForCausalLM', 'TransformerModel', |
| 'BasedLinearAttention', |
| 'BitAttention', 'BitNetForCausalLM', 'BitNetModel', |
| 'Comba', 'CombaForCausalLM', 'CombaModel', |
| 'DeltaNet', 'DeltaNetForCausalLM', 'DeltaNetModel', |
| 'DeltaFormerAttention', 'DeltaFormerForCausalLM', 'DeltaFormerModel', |
| 'GatedDeltaNet', 'GatedDeltaNetForCausalLM', 'GatedDeltaNetModel', |
| 'GatedDeltaProduct', 'GatedDeltaProductForCausalLM', 'GatedDeltaProductModel', |
| 'GatedLinearAttention', 'GLAForCausalLM', 'GLAModel', |
| 'GatedSlotAttention', 'GSAForCausalLM', 'GSAModel', |
| 'HGRNAttention', 'HGRNForCausalLM', 'HGRNModel', |
| 'HGRN2Attention', 'HGRN2ForCausalLM', 'HGRN2Model', |
| 'LightNetAttention', 'LightNetForCausalLM', 'LightNetModel', |
| 'LinearAttention', 'LinearAttentionForCausalLM', 'LinearAttentionModel', |
| 'LogLinearMamba2', 'LogLinearMamba2ForCausalLM', 'LogLinearMamba2Model', |
| 'MesaNet', 'MesaNetForCausalLM', 'MesaNetModel', |
| 'MomAttention', 'MomForCausalLM', 'MomModel', |
| 'MultiheadLatentAttention', 'MLAForCausalLM', 'MLAModel', |
| 'MultiScaleRetention', 'RetNetForCausalLM', 'RetNetModel', |
| 'NativeSparseAttention', 'NSAForCausalLM', 'NSAModel', |
| 'PaTHAttention', 'PaTHAttentionForCausalLM', 'PaTHAttentionModel', |
| 'ReBasedLinearAttention', |
| 'RodimusAttention', 'RodimusForCausalLM', 'RodimusModel', |
| 'RWKV6Attention', 'RWKV6ForCausalLM', 'RWKV6Model', |
| 'RWKV7Attention', 'RWKV7ForCausalLM', 'RWKV7Model', |
| ] |
|
|
| __version__ = '0.4.0' |
|
|
