cCi tSrSSKrSSKJr SSKJrJr SSKrSSKJr SSK J r SSK J r SS K Jr SS KJrJr SS KJrJrJrJrJrJr SS KJr S SKJr \"5(a SSKJrJr OSrSr\R@"\!5r"\\"SS9"SS\555r#\\"SS9"SS\555r$\\"SS9"SS\555r%"SS\RL5r'"SS\RL5r("S S!\RL5r)SDS"\RTS#\+S$\,S%\RT4S&jjr-"S'S(\RL5r."S)S*\RL5r/"S+S,\RL5r0"S-S.\RL5r1"S/S0\RL5r2"S1S2\RL5r3"S3S4\RL5r4"S5S6\RL5r5"S7S8\RL5r6\"S9S:\55r7\"S;S<\755r8\"S=S9"S>S?\755r9\"S@S9"SASB\7\55r:/SCQr;g)Ez9PyTorch Dilated Neighborhood Attention Transformer model.N) dataclass)OptionalUnion)nn)ACT2FN)BackboneOutput)PreTrainedModel) find_pruneable_heads_and_indicesprune_linear_layer) ModelOutputOptionalDependencyNotAvailableauto_docstringis_natten_availableloggingrequires_backends) BackboneMixin) DinatConfig) natten2dav natten2dqkrpbc[5eNrargskwargss b/home/james-whalen/.local/lib/python3.13/site-packages/transformers/models/dinat/modeling_dinat.pyrr, ,..c[5errrs rrr/rr zO Dinat encoder's outputs, with potential hidden states and attentions. ) custom_introc\rSrSr%SrSr\\R\ S'Sr \\ \RS4\ S'Sr \\ \RS4\ S'Sr \\ \RS4\ S'S rg) DinatEncoderOutput9a reshaped_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`): Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each stage) of shape `(batch_size, hidden_size, height, width)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs reshaped to include the spatial dimensions. Nlast_hidden_state. hidden_states attentionsreshaped_hidden_states)__name__ __module__ __qualname____firstlineno____doc__r&rtorch FloatTensor__annotations__r'tupler(r)__static_attributes__r*r rr$r$9s}6:x 1 129=AM8E%"3"3S"89:A:>Ju00#567>FJHU5+<+