qyiSrSSKJr SSKJr SSKJrJr \R"\ 5r \"SS9\"SS \555r \"SS9\"S S \555r \"SS9\"S S \555r /SQrg)zMllama model configuration)strict)PreTrainedConfig)auto_docstringloggingzmeta-llama/Llama-3.2-11B-Vision) checkpointc^\rSrSr%SrSrSrSS0rSr\ \ S'S r \ \ S 'S r \ \ S 'S r\ \ S'Sr\ \ S'Sr\ \ S'Sr\ \ S'Sr\ \ S'Sr\ \\ -\\ \ 4-\ S'Sr\ \\ -\\ \ 4-\ S'Sr\\ S'Sr\ \ S'Sr\\ S-\ S'Sr\\\ S-\ S 'S!r\\ S"'U4S#jrS$r\S%\ 4S&j5r S'r!U=r"$)(MllamaVisionConfigaV num_global_layers (`int`, *optional*, defaults to 8): Number of global layers in the Transformer encoder. Vision model has a second transformer encoder, called global. vision_output_dim (`int`, *optional*, defaults to 7680): Dimensionality of the vision model output. Includes output of transformer encoder with intermediate layers and global transformer encoder. max_num_tiles (`int`, *optional*, defaults to 4): Maximum number of tiles for image splitting. intermediate_layers_indices (`list[int]`, *optional*, defaults to [3, 7, 15, 23, 30]): Indices of intermediate layers of transformer encoder from which to extract and output features. These output features are concatenated with final hidden state of transformer encoder. supported_aspect_ratios (`list[list[int]]`, *optional*): List of supported aspect ratios for image splitting. If not specified, the default supported aspect ratios are [[1, 1], [1, 2], [1, 3], [1, 4], [2, 1], [2, 2], [3, 1], [4, 1]] for `max_num_tiles=4`. Example: ```python >>> from transformers import MllamaVisionConfig, MllamaVisionModel >>> # Initializing a Llama config >>> config = MllamaVisionConfig() >>> # Initializing a vision model from the mllama-11b style configuration >>> model = MllamaVisionModel(config) >>> # Accessing the model configuration >>> configuration = model.config ```mllama_vision_model vision_confignum_attention_headsattention_headsi hidden_sizegelu hidden_act num_hidden_layersnum_global_layersr num_channelsiintermediate_sizeivision_output_dimi image_size patch_sizeh㈵>norm_eps max_num_tilesNintermediate_layers_indicessupported_aspect_ratios{Gz?initializer_rangec >URcSS/SS/SS/SS/SS/SS/SS/SS//UlURc /SQUl[TU] "S0UD6 g)Nrr )r)r#r"super __post_init__selfkwargs __class__s l/home/mande/repo/quber/.venv/lib/python3.13/site-packages/transformers/models/mllama/configuration_mllama.pyr/ MllamaVisionConfig.__post_init__Msv  ' ' /-.FQFQFQFQPQFUVXYTZ]^`a\befhidj+kD (  + + 3/AD , ''c URSS/SS/SS/SS/SS/SS/SS/SS//:XaURS:wa [S5egg)zOPart of `@strict`-powered validation. Validates the architecture of the config.r'r(rr z;max_num_tiles must be 4 for default supported aspect ratiosN)r#r! ValueErrorr1s r4validate_architecture(MllamaVisionConfig.validate_architectureUsr  ( (aVaVaVaVaQRVVWYZU[^_ab]cfgijek,l l""a'Z[ [( mr6returnc,[UR5$)N)lenr#r9s r4max_aspect_ratio_id&MllamaVisionConfig.max_aspect_ratio_id]s4//00r6)r"r#)#__name__ __module__ __qualname____firstlineno____doc__ model_typebase_config_key attribute_maprint__annotations__rstrrrrrrrrlisttuplerrfloatr!r"r#r%r/r:propertyr?__static_attributes__ __classcell__r3s@r4r r s" <'J%O*,=>MKJssOSL#!s!!s!47Jd3i%S/1746Jd3i%S/16HeM348cT!186:T$s)_t3:#u#(\1S11r6r cf^\rSrSr%SrSrSrSrSr\ \ S'Sr \ \ S 'S r \ \ S 'S r\ \ S 'Sr\ \ S'Sr\ \ S'Sr\ \ S'Sr\S-\ S'Sr\\ S'Sr\ \ S'Sr\\ S'Sr\\ S'Sr\\ S'Sr\\ S-\ S 'S!r\\ -\ S"'S#r\ \ S$'S%r\ \\ -S-\ S&'S'r \ S-\ S('U4S)jr!S*r"U=r#$)+MllamaTextConfigba cross_attention_layers (`list[int]`, *optional*): Indices of the cross attention layers. If not specified, will default to [3, 8, 13, 18, 23, 28, 33, 38]. Example: ```python >>> from transformers import MllamaTextModel, MllamaTextConfig >>> # Initializing a Mllama text config >>> config = MllamaTextConfig() >>> # Initializing a model from the Mllama text configuration >>> model = MllamaTextModel(config) >>> # Accessing the model configuration >>> configuration = model.config ```mllama_text_model text_configgA vocab_sizeirsilur(rrrrnum_key_value_headsi8rNrope_parametersr rms_norm_epsimax_position_embeddingsr$r%T use_cacheFtie_word_embeddingscross_attention_layersgdropouti bos_token_idi eos_token_idi pad_token_idc R>URc /SQUl[TU]"S0UD6 g)N)rr r+!&r-)rbr.r/r0s r4r/MllamaTextConfig.__post_init__s'  & & .*HD ' ''r6)rb)$rArBrCrDrErFrG default_thetarYrIrJrrrKrrr\rr]dictr^rNr_r%r`boolrarbrLrcrdrerfr/rPrQrRs@r4rTrTbs &%J#OMJKJs!!  #s##'OTD['L%#*S*#u#It %%/3DI,3GUS[L#+1L#S /D(1%L#*%((r6rTc^\rSrSr%SrSrSS0r\\S.r Sr \ \ -S-\ S'Sr\ \ -S-\ S 'S r\\ S'U4S jrSrU=r$) MllamaConfiga\ Example: ```python >>> from transformers import MllamaForConditionalGeneration, MllamaConfig, MllamaVisionConfig, MllamaTextConfig >>> # Initializing a CLIP-vision config >>> vision_config = MllamaVisionConfig() >>> # Initializing a Llama config >>> text_config = MllamaTextConfig() >>> # Initializing a mllama-11b style configuration >>> configuration = MllamaConfig(vision_config, text_config) >>> # Initializing a model from the mllama-11b style configuration >>> model = MllamaForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```mllamaimage_token_idimage_token_index)rWr Nr rWrXc >URc%[5Ul[RS5 O9[ UR[ 5(a[S0URD6UlUR c%[5Ul[RS5 O9[ UR [ 5(a[S0UR D6Ul[TU]$"S0UD6 g)Nz9vision_config is None, using default mllama vision configz5text_config is None, using default mllama text configr-) r r loggerinfo isinstancerorWrTr.r/r0s r4r/MllamaConfig.__post_init__s    %!3!5D  KKS T **D 1 1!3!Id6H6H!ID     #/1D  KKO P (($ / //C$2B2BCD  ''r6)rArBrCrDrErFrHrTr sub_configsr rorrJrWrvrIr/rPrQrRs@r4rrrrsh ,J-M#3EWXK48M4**T1826K((4/6#s# ( (r6rr)rrrTr N)rEhuggingface_hub.dataclassesrconfiguration_utilsrutilsrr get_loggerrArxr rTrr__all__r-r6r4rs!.3,   H %<=E1)E1>E1P<=.('.(>.(b<=.(#.(>.(b Er6