qyiSrSSKJr SSKJr SSKJr \"SS9\"SS \555r\"SS9\"S S \555r\"SS9\"S S \555r \"SS9\"SS\555r /SQr g)zSAM model configuration)strict)PreTrainedConfig)auto_docstringzfacebook/sam-vit-huge) checkpointc^\rSrSr%SrSrSr\\S'Sr \\ \-\ \\4-\S'Sr \\ \-\ \\4-\S 'Sr \\S 'S r\\S 'S r\\S'Sr\\S'U4SjrSrU=r$)SamPromptEncoderConfigz mask_input_channels (`int`, *optional*, defaults to 16): The number of channels to be fed to the `MaskDecoder` module. num_point_embeddings (`int`, *optional*, defaults to 4): The number of point embeddings to be used. prompt_encoder_config hidden_size image_size patch_sizemask_input_channelsnum_point_embeddingsgelu hidden_actư>layer_norm_epsc b>URUR-Ul[TU]"S0UD6 gN)rrimage_embedding_sizesuper __post_init__selfkwargs __class__s f/home/mande/repo/quber/.venv/lib/python3.13/site-packages/transformers/models/sam/configuration_sam.pyr$SamPromptEncoderConfig.__post_init__*s($(OOt$F! '')r)__name__ __module__ __qualname____firstlineno____doc__base_config_keyr int__annotations__rlisttuplerrrrstrrfloatr__static_attributes__ __classcell__r"s@r#r r s.OK48Jd3i%S/1846Jd3i%S/16!! !#!J NE ((r%r c\rSrSr%SrSrSr\\S'Sr \ \S'Sr \\S 'S r \\S 'S r \\S 'S r\\S'Sr\\S'Sr\\S'Sr\\S'Sr\\S'Srg)SamMaskDecoderConfig/a mlp_dim (`int`, *optional*, defaults to 2048): Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder. attention_downsample_rate (`int`, *optional*, defaults to 2): The downsampling rate of the attention layer. num_multimask_outputs (`int`, *optional*, defaults to 3): The number of outputs from the `SamMaskDecoder` module. In the Segment Anything paper, this is set to 3. iou_head_depth (`int`, *optional*, defaults to 3): The number of layers in the IoU head module. iou_head_hidden_dim (`int`, *optional*, defaults to 256): The dimensionality of the hidden states in the IoU head module. mask_decoder_configr r relurimlp_dimnum_hidden_layersnum_attention_headsattention_downsample_raternum_multimask_outputsiou_head_depthiou_head_hidden_dimrrrN)r&r'r(r)r*r+r r,r-rr0r:r<r>r?r@rArBrr1r2rr%r#r6r6/sw ,OKJGSs  %&s&!"3"NC"" NE r%r6c^\rSrSr%SrSrSrSr\\ S'Sr \\ S'S r \\ S 'S r \\ S 'S r \\ S 'Sr\\\-\\\4-\ S'Sr\\\-\\\4-\ S'Sr\\ S'Sr\\ S'Sr\\-\ S'Sr\\ S'Sr\\ S'Sr\\ S'Sr\\ S'Sr\\ S'S r\\ S!'S"r\\\\S#4-\ S$'S%r\\ S&'S'r \S'-\ S('U4S)jr!S*r"U=r#$)+SamVisionConfigMa output_channels (`int`, *optional*, defaults to 256): Dimensionality of the output channels in the Patch Encoder. use_rel_pos (`bool`, *optional*, defaults to `True`): Whether to use relative position embedding. window_size (`int`, *optional*, defaults to 14): Window size for relative position. global_attn_indexes (`list[int]`, *optional*, defaults to `[2, 5, 8, 11]`): The indexes of the global attention layers. num_pos_feats (`int`, *optional*, defaults to 128): The dimensionality of the position embedding. mlp_dim (`int`, *optional*): The dimensionality of the MLP layer in the Transformer encoder. If `None`, defaults to `mlp_ratio * hidden_size`. Example: ```python >>> from transformers import ( ... SamVisionConfig, ... SamVisionModel, ... ) >>> # Initializing a SamVisionConfig with `"facebook/sam-vit-huge"` style configuration >>> configuration = SamVisionConfig() >>> # Initializing a SamVisionModel (with random weights) from the `"facebook/sam-vit-huge"` style configuration >>> model = SamVisionModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` vision_configsam_vision_modelir r output_channels r<r>r num_channelsrrrrrrrrgattention_dropoutg|=initializer_rangeTqkv_biasg@ mlp_ratio use_abs_pos use_rel_pos window_size)r;r= .global_attn_indexes num_pos_featsNr:c >URc"[URUR-5O URUlURS-Ul[ TU]"S0UD6 g)Nr;r)r:r,r rNscalerrrs r#rSamVisionConfig.__post_init__sRAEAUs4++dnn<=[_[g[g %%*  ''r%)r:rY)$r&r'r(r)r*r+ model_typer r,r-rHr<r>rJrr.r/rrr0rr1rKrLrMboolrNrOrPrRrUrWr:rr2r3r4s@r#rDrDMs% B&O#JKOSs!!L#48Jd3i%S/1846Jd3i%S/16J!NE!%(us{($u$HdIuKKK7DcU38_4DM3GS4Z((r%rDc^\rSrSr%SrSr\\\S.r Sr \ \ -S-\ S'Sr\ \ -S-\ S'Sr\ \ -S-\ S'S r\\ S 'S r\\ S 'U4S jrSrU=r$) SamConfiga prompt_encoder_config (Union[`dict`, `SamPromptEncoderConfig`], *optional*): Dictionary of configuration options used to initialize [`SamPromptEncoderConfig`]. mask_decoder_config (Union[`dict`, `SamMaskDecoderConfig`], *optional*): Dictionary of configuration options used to initialize [`SamMaskDecoderConfig`]. Example: ```python >>> from transformers import ( ... SamVisionConfig, ... SamPromptEncoderConfig, ... SamMaskDecoderConfig, ... SamModel, ... ) >>> # Initializing a SamConfig with `"facebook/sam-vit-huge"` style configuration >>> configuration = SamConfig() >>> # Initializing a SamModel (with random weights) from the `"facebook/sam-vit-huge"` style configuration >>> model = SamModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config >>> # We can also initialize a SamConfig from a SamVisionConfig, SamPromptEncoderConfig, and SamMaskDecoderConfig >>> # Initializing SAM vision, SAM Q-Former and language model configurations >>> vision_config = SamVisionConfig() >>> prompt_encoder_config = SamPromptEncoderConfig() >>> mask_decoder_config = SamMaskDecoderConfig() >>> config = SamConfig(vision_config, prompt_encoder_config, mask_decoder_config) ```sam)r r8rFNrFr r8g{Gz?rLTtie_word_embeddingsc *>[UR[5(a[S0URD6UlOURc[5Ul[UR[5(a[ S0URD6UlOURc[ 5Ul[UR [5(a[S0UR D6UlOUR c[5Ul[TU]$"S0UD6 gr) isinstancerFdictrDr r r8r6rrrs r#rSamConfig.__post_init__s d(($ / /!0!F43E3E!FD     '!0!2D  d00$ 7 7)?)]$B\B\)]D &  ' ' /)?)AD & d.. 5 5';'Wd>V>V'WD $  % % -';'=D $ ''r%)r8r rF)r&r'r(r)r*r[r r6rD sub_configsrFrdrr-r r8rLr1rar\rr2r3r4s@r#r^r^s! FJ!73(K 59M4**T18<@4"22T9@:> 0047>#u# $$((r%r^)r^r6r rDN) r*huggingface_hub.dataclassesrconfiguration_utilsrutilsrr r6rDr^__all__rr%r#rks.3#23(-(4(.23!+!4!823<(&<(4<(~23A( A(4A(H ]r%