+ j"Rt^RIHt^RIHt^RIHt^RIHtH t ^RI H t H t ] P!]4t]!RR7]!RR ]444t]!RR7]!R R ]444t]!RR7]!R R ]444t.ROtR#)z InstructBLIP model configuration)strict)PreTrainedConfig)!MODEL_FOR_CAUSAL_LM_MAPPING_NAMES)auto_docstringlogging)CONFIG_MAPPING AutoConfigz"Salesforce/instructblip-flan-t5-xl) checkpointcda]tRt^toRtRtRtRtRt^'t ^t ^t ^t Rt RtRtR tR tV3R ltR tVtR #)InstructBlipVisionConfiga Example: ```python >>> from transformers import InstructBlipVisionConfig, InstructBlipVisionModel >>> # Initializing a InstructBlipVisionConfig with Salesforce/instructblip-flan-t5-xl style configuration >>> configuration = InstructBlipVisionConfig() >>> # Initializing a InstructBlipVisionModel (with random weights) from the Salesforce/instructblip-flan-t5-xl style configuration >>> model = InstructBlipVisionModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```instructblip_vision_model vision_configigelugư>gg|=Tc:<V^8dQh/S[;R&S[;R&S[;R&S[;R&S[S[S[,,S[S[S[3,,;R&S[S[S[,,S[S[S[3,,;R&S[;R&S[;R&S[S[,;R &S[;R &S[;R &#)  hidden_sizeintermediate_sizenum_hidden_layersnum_attention_heads image_size patch_size hidden_actlayer_norm_epsattention_dropoutinitializer_rangeqkv_bias)intlisttuplestrfloatbool)format __classdict__s"|/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/instructblip/configuration_instructblip.py __annotate__%InstructBlipVisionConfig.__annotate__s,-.!/012!34d3i%S/1756d3i%S/16789: ;<s{(=>$?@AN)__name__ __module__ __qualname____firstlineno____doc__ model_typebase_config_keyrrrrrrrrrrr__annotate_func____static_attributes____classdictcell__r$s@r%r r sZ -J%OK!!47J46JJ N%($HAr(r cpa]tRt^>toRtRtRtRtRt^ t ^ t Rt Rt Rt RtR tR tR t^t^tR tV3R ltRtVtR#)InstructBlipQFormerConfiga cross_attention_frequency (`int`, *optional*, defaults to 2): The frequency of adding cross-attention to the Transformer layers. encoder_hidden_size (`int`, *optional*, defaults to 1408): The hidden size of the hidden states for cross-attention. Examples: ```python >>> from transformers import InstructBlipQFormerConfig, InstructBlipQFormerModel >>> # Initializing a InstructBLIP Salesforce/instructblip-flan-t5-xl style configuration >>> configuration = InstructBlipQFormerConfig() >>> # Initializing a model (with random weights) from the Salesforce/instructblip-flan-t5-xl style configuration >>> model = InstructBlipQFormerModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```instructblip_qformerqformer_configi:wii rg?i{Gz?g-q=rc<V^8dQh/S[;R&S[;R&S[;R&S[;R&S[;R&S[;R&S[S[,;R&S[S[,;R&S[;R &S[;R &S[;R &S[R ,;R &S[;R&S[;R&#)r vocab_sizerrrrrhidden_dropout_probattention_probs_dropout_probmax_position_embeddingsrrN pad_token_idcross_attention_frequencyencoder_hidden_size)rr r!)r#r$s"r%r&&InstructBlipQFormerConfig.__annotate__>s456789:!;<!=>?@*AB#(#+3CD!&EF#GH!IJ* KL #&MN#Or(r)N)r*r+r,r-r.r/r0r;rrrrrr<r=r>rrr?r@rAr1r2r3r4s@r%r6r6>sm ((J&OJK!!J'*03 #&#!N L%&#Or(r6czaa]tRt^htoRtRtRR/tR]R]R] /t Rt Rt Rt ^ tRtR tR tV3R ltV3R ltR tVtV;t#)InstructBlipConfigaY qformer_config (`dict`, *optional*): Dictionary of configuration options used to initialize [`InstructBlipQFormerConfig`]. num_query_tokens (`int`, *optional*, defaults to 32): The number of query tokens passed through the Transformer. Example: ```python >>> from transformers import ( ... InstructBlipVisionConfig, ... InstructBlipQFormerConfig, ... OPTConfig, ... InstructBlipConfig, ... InstructBlipForConditionalGeneration, ... ) >>> # Initializing a InstructBlipConfig with Salesforce/instructblip-flan-t5-xl style configuration >>> configuration = InstructBlipConfig() >>> # Initializing a InstructBlipForConditionalGeneration (with random weights) from the Salesforce/instructblip-flan-t5-xl style configuration >>> model = InstructBlipForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config >>> # We can also initialize a InstructBlipConfig from a InstructBlipVisionConfig, InstructBlipQFormerConfig and any PreTrainedConfig >>> # Initializing InstructBLIP vision, InstructBLIP Q-Former and language model configurations >>> vision_config = InstructBlipVisionConfig() >>> qformer_config = InstructBlipQFormerConfig() >>> text_config = OPTConfig() >>> config = InstructBlipConfig(vision_config=vision_config, qformer_config=qformer_config, text_config=text_config) ``` instructblipimage_token_idimage_token_index text_configr8r Ng?r9c <VPf.\R,!4Vn\PR4M^\ VP\ 4'd?VPP RR4p\V,!R/VPBVnVPf&\4Vn\PR4M:\ VP\ 4'd\R/VPBVnVPf&\4Vn \PR4M:\ VP\ 4'd\R/VPBVn VPPVPn VPP\9Vn\ SV`D!R/VBR#)NoptzTtext_config is None. Initializing the text config with default values (`OPTConfig`).r/zWqformer_config is None. Initializing the InstructBlipQFormerConfig with default values.z[`vision_config` is `None`. initializing the `InstructBlipVisionConfig` with default values.r))rHrloggerinfo isinstancedictgetr8r6r r rrAr/ruse_decoder_only_language_modelsuper __post_init__)selfkwargstext_model_type __class__s&, r%rR InstructBlipConfig.__post_init__sP    #-e46D  KKn o (($ / /"..22<GO-o>RAQAQRD     &";"=D  KKq r ++T 2 2";"Rd>Q>Q"RD     %!9!;D  KKu v **D 1 1!9!ODrfs'.3J,-   H %?@/AB?@%$ 0%$A%$P?@M()M(AM(` Zr(