^jddlmZddlmZddlmZedeGddeZedeGd d eZd gZy ) )strict)PreTrainedConfig)auto_docstringz&PaddlePaddle/SLANeXt_wired_safetensors) checkpointc<eZdZUdZdZdZeed<dZeed<dZ eed<dZ eed <d Z eed <d Z eed <dZ eeezeeefzed<dZeed<dZeed<dZeezed<dZeed<dZeed<dZeed<dZeed<dZeed<dZeeeedfzed <d!Zeed"<y#)$SLANeXtVisionConfiga output_channels (`int`, *optional*, defaults to 256): Dimensionality of the output channels in the Patch Encoder. use_abs_pos (`bool`, *optional*, defaults to `True`): Whether to use absolute position embedding. use_rel_pos (`bool`, *optional*, defaults to `True`): Whether to use relative position embedding. window_size (`int`, *optional*, defaults to 14): Window size for relative position. global_attn_indexes (`list[int]`, *optional*, defaults to `[2, 5, 8, 11]`): The indexes of the global attention layers. mlp_dim (`int`, *optional*, defaults to 3072): The dimensionality of the MLP layer in the Transformer encoder. vision_configi hidden_sizeoutput_channels num_hidden_layersnum_attention_headsr num_channels image_size patch_sizegelu hidden_actgư>layer_norm_epsgattention_dropoutg|=initializer_rangeTqkv_bias use_abs_pos use_rel_pos window_size) .global_attn_indexesi mlp_dimN)__name__ __module__ __qualname____doc__base_config_keyr int__annotations__r rrrrrlisttuplerstrrfloatrrrboolrrrr$r%|/var/www/ramen.bs-engineer-server.com/venv/lib/python3.12/site-packages/transformers/models/slanext/configuration_slanext.pyr r s &OKOSs!!L#J46Jd3i%S/16J!NE!%(us{($u$HdKKK7DcU38_4DGSr3r ceZdZUdZdZdeiZdZeezdze d<dZ e e d<dZ e e d<d Z e e d <dZe e d <d Ze e d <fdZxZS) SLANeXtConfiga vision_config (`dict` or [`SLANeXtVisionConfig`], *optional*): Configuration for the vision encoder. If `None`, a default [`SLANeXtVisionConfig`] is used. post_conv_in_channels (`int`, *optional*, defaults to 256): Number of input channels for the post-encoder convolution layer. post_conv_out_channels (`int`, *optional*, defaults to 512): Number of output channels for the post-encoder convolution layer. out_channels (`int`, *optional*, defaults to 50): Vocabulary size for the table structure token prediction head, i.e., the number of distinct structure tokens the model can predict. hidden_size (`int`, *optional*, defaults to 512): Dimensionality of the hidden states in the attention GRU cell and the structure/location prediction heads. max_text_length (`int`, *optional*, defaults to 500): Maximum number of autoregressive decoding steps (tokens) for the structure and location decoder. slanextr Nr post_conv_in_channelsrpost_conv_out_channels2 out_channelsr imax_text_lengthc |jt|_n4t|jtrtdi|j|_t |di|y)Nr2)r r isinstancedictsuper __post_init__)selfkwargs __class__s r4rAzSLANeXtConfig.__post_init___sP    %!4!6D  **D 1!4!Jt7I7I!JD  ''r3)r&r'r(r) model_typer sub_configsr r?r,r8r+r9r;r r<rA __classcell__)rDs@r4r6r6Bsm J"$78K7;M4--4;!$3$"%C%L#KOS((r3r6N) huggingface_hub.dataclassesrconfiguration_utilsrutilsrr r6__all__r2r3r4rLsr,/3#CD!*!E!HCD ($ (E (F  r3