cCiPSSKrSSKJr SSKJrJrJr SSKrSSKJ s J r SSKJ r SSK J r SSKJr SSKJrJrJrJr SS KJr SS KJr SS KJrJrJr SS KJrJr SS K J!r! SSK"J#r#J$r$J%r%J&r&J'r' SSK(J)r) SSK*J+r+ SSK,J-r-J.r.J/r/ \&R`"\15r2\$"SS\55r3\\$"SS9"SS\555r4\\$"SS9"SS\555r5\\$"SS9"SS\555r6"S S!\ Rn5r8S"\RrS#\:S$\Rr4S%jr;S^S&\ RnS'\RrS(\RrS)\RrS*\\RrS+\"S1S2\ Rn5r?"S3S4\5r@"S5S6\ Rn5rA"S7S8\ Rn5rB"S9S:\ Rn5rC"S;S<\5rD\$"S=S>\355rE"S?S@\ Rn5rF"SASB\ Rn5rG"SCSD\ Rn5rH"SESF\ Rn5rI"SGSH\ Rn5rJ"SISJ\ Rn5rK"SKSL\ Rn5rL"SMSN\ Rn5rM"SOSP\ Rn5rN\$"SQS9"SRSS\355rO"STSU\ Rn5rP"SVSW\ Rn5rQ\$"SXS9"SYSZ\355rR"S[S\\3\5rS/S]QrTg)_N) dataclass)CallableOptionalUnion)nn)ACT2FN)Cache)%ClassifierFreeGuidanceLogitsProcessorGenerationMixinGenerationModeLogitsProcessorList)GenerateDecoderOnlyOutput)GradientCheckpointingLayer)BaseModelOutputBaseModelOutputWithPooling ModelOutput)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuplelogging torch_int)check_model_inputs) AutoModel) JanusConfigJanusVisionConfigJanusVQVAEConfigcH\rSrSr%\\S'SrSrSS/rSS/r Sr Sr Sr S r S rg ) JanusPreTrainedModel/configmodelTLlamaDecoderLayerJanusVisionEncoderLayerpast_key_values causal_maskFN)__name__ __module__ __qualname____firstlineno__r __annotations__base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_can_compile_fullgraph!_supports_param_buffer_assignment__static_attributes__r,b/home/james-whalen/.local/lib/python3.13/site-packages/transformers/models/janus/modeling_janus.pyr$r$/sB &*#,.GH#4m"DN!(-%r;r$z9 Base class for Janus VQ-VAE mode model outputs. ) custom_introcj\rSrSr%SrSr\\R\ S'Sr \\R\ S'Sr g)JanusVQVAEOutput=z decoded_pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`): Reconstructed pixel values after encoding and decoding the input. embedding_loss (`torch.FloatTensor`): Embedding loss. Ndecoded_pixel_valuesembedding_lossr,) r-r.r/r0__doc__rArtorch FloatTensorr1rBr:r,r;r<r?r?=s49=(5#4#45<26NHU../6r;r?zy Base class for Janus model's outputs that may also contain a past key/values (to speed up sequential decoding). c\rSrSr%SrSr\\R\ S'Sr \\ \ S'Sr \\ \R\ S'Sr\\ \R\ S'Sr\\ \R\ S'S rg) JanusBaseModelOutputWithPastOaa last_hidden_state (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`): Sequence of hidden-states at the output of the last layer of the model. If `past_key_values` is used only the last hidden-state of the sequences of shape `(batch_size, 1, hidden_size)` is output. past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): It is a [`~cache_utils.Cache`] instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if `config.is_encoder_decoder=True` in the cross-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. image_hidden_states (`tuple(torch.FloatTensor)`, *optional*): Tuple of `torch.FloatTensor` (one for the output of the image embeddings, `(batch_size, num_images, sequence_length, hidden_size)`. image_hidden_states of the model produced by the vision encoder, and optionally by the perceiver Nlast_hidden_stater* hidden_states attentionsimage_hidden_statesr,)r-r.r/r0rCrIrrDrEr1r*r rJtuplerKrLr:r,r;r<rGrGOs|&6:x 1 129'+OXe_+8B%(9(9":;Br;rGzQ Base class for Janus causal language model (or autoregressive) outputs. c\rSrSr%SrSr\\R\ S'Sr \\R\ S'Sr \\ \ S'Sr \\\R\ S'Sr\\\R\ S'Sr\\\R\ S 'S rg) JanusCausalLMOutputWithPastpa1 loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): It is a [`~cache_utils.Cache`] instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. image_hidden_states (`tuple(torch.FloatTensor)`, *optional*): Tuple of `torch.FloatTensor` (one for the output of the image embeddings, `(batch_size, num_images, sequence_length, hidden_size)`. image_hidden_states of the model produced by the vision encoder, and optionally by the perceiver Nlosslogitsr*rJrKrLr,)r-r.r/r0rCrQrrDrEr1rRr*r rJrMrKrLr:r,r;r<rOrOps")-D(5$$ %,*.FHU&& '.'+OXe_+8B%(9(9":;Br;rOc^\rSrSrS\4U4SjjrS\RS\S\S\R4Sjr S S \RS \ S\R4S jjr S r U=r $)JanusVisionEmbeddingsr&c^>[TU]5 XlURUlUR UlUR Ul[R"URURUR UR SS9Ul UR UR -S-Ul URUl [R"URUR5UlURS[ R""UR5R%S5SS9 g)Nvalid) in_channels out_channels kernel_sizestridepaddingr position_ids)rF) persistent)super__init__r& hidden_size embed_dim image_size patch_sizerConv2d num_channelspatch_embedding num_patches num_positions Embeddingposition_embeddingregister_bufferrDarangeexpandselfr& __class__s r<raJanusVisionEmbeddings.__init__s  ++ ++ ++!yy++??  !OOt>1D!--"$,,t/A/A4>>"R ^U\\$:L:L-M-T-TU\-]jopr; embeddingsheightwidthreturncURSnURRRSn[RR 5(d%XE:Xa X#:XaURUR 5$URRRS5nURSnX R-nX0R-n [US-5n URSXU5nURSSSS5n[RRUX4SSS 9nURSSSS5RSSU5nU$) a This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher resolution images. This method is also adapted to support torch.jit tracing and no class embeddings. Adapted from: - https://github.com/facebookresearch/dino/blob/de9ee3df6cf39fac952ab558447af1fa1365362a/vision_transformer.py#L174-L194, and - https://github.com/facebookresearch/dinov2/blob/e1277af2ba9496fbadf7aec6eba56e8d882d1e35/dinov2/models/vision_transformer.py#L179-L211 rrr^g?rrbicubicF)sizemode align_corners)shaperlweightrDjit is_tracingr] unsqueezererreshapepermuter functional interpolateview) rqrtrurvrirjpatch_pos_embeddim new_height new_widthsqrt_num_positionss r<interpolate_pos_encoding.JanusVisionEmbeddings.interpolate_pos_encodings:!&&q) //66<