o Zh@sdZddlmZddlmZddlmZddlmZddl m Z ddl m Z e eZGd d d eZGd d d e Zd d gZd S)zLeViT model configuration OrderedDict)Mapping)version)PretrainedConfig) OnnxConfig)loggingcsZeZdZdZdZddddddgdgd gd gd d gd gd dffdd ZZS) LevitConfiga This is the configuration class to store the configuration of a [`LevitModel`]. It is used to instantiate a LeViT model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the LeViT [facebook/levit-128S](https://huggingface.co/facebook/levit-128S) architecture. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: image_size (`int`, *optional*, defaults to 224): The size of the input image. num_channels (`int`, *optional*, defaults to 3): Number of channels in the input image. kernel_size (`int`, *optional*, defaults to 3): The kernel size for the initial convolution layers of patch embedding. stride (`int`, *optional*, defaults to 2): The stride size for the initial convolution layers of patch embedding. padding (`int`, *optional*, defaults to 1): The padding size for the initial convolution layers of patch embedding. patch_size (`int`, *optional*, defaults to 16): The patch size for embeddings. hidden_sizes (`List[int]`, *optional*, defaults to `[128, 256, 384]`): Dimension of each of the encoder blocks. num_attention_heads (`List[int]`, *optional*, defaults to `[4, 8, 12]`): Number of attention heads for each attention layer in each block of the Transformer encoder. depths (`List[int]`, *optional*, defaults to `[4, 4, 4]`): The number of layers in each encoder block. key_dim (`List[int]`, *optional*, defaults to `[16, 16, 16]`): The size of key in each of the encoder blocks. drop_path_rate (`int`, *optional*, defaults to 0): The dropout probability for stochastic depths, used in the blocks of the Transformer encoder. mlp_ratios (`List[int]`, *optional*, defaults to `[2, 2, 2]`): Ratio of the size of the hidden layer compared to the size of the input layer of the Mix FFNs in the encoder blocks. attention_ratios (`List[int]`, *optional*, defaults to `[2, 2, 2]`): Ratio of the size of the output dimension compared to input dimension of attention layers. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. Example: ```python >>> from transformers import LevitConfig, LevitModel >>> # Initializing a LeViT levit-128S style configuration >>> configuration = LevitConfig() >>> # Initializing a model (with random weights) from the levit-128S style configuration >>> model = LevitModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```Zlevitr)i) )rrr)rrrr)r r r g{Gz?c stjdi|||_||_||_||_||_||_||_| |_ | |_ | |_ ||_ | |_ | |_||_d| d|d| ddddgd| d|d| ddddgg|_dS)NZ Subsamplerrr r )super__init__ image_size num_channels kernel_sizestridepadding hidden_sizesnum_attention_headsdepthskey_dimdrop_path_rate patch_sizeattention_ratio mlp_ratioinitializer_rangeZdown_ops)selfrrrrrr!rrrrr r#r"r$kwargs __class__r\/var/www/auris/lib/python3.10/site-packages/transformers/models/levit/configuration_levit.pyrXs$ zLevitConfig.__init__)__name__ __module__ __qualname____doc__Z model_typer __classcell__rrr'r)r s$7r c@sJeZdZedZedeeee efffddZ ede fddZ dS)LevitOnnxConfigz1.11returncCstddddddfgS)NZ pixel_valuesbatchrheightwidth)rr r rrr%rrr)inputsszLevitOnnxConfig.inputscCsdS)Ng-C6?rr4rrr)atol_for_validationsz#LevitOnnxConfig.atol_for_validationN) r*r+r,rparseZtorch_onnx_minimum_versionpropertyrstrintr5floatr6rrrr)r/s   r/N)r- collectionsrtypingr packagingrZconfiguration_utilsrZonnxrutilsr Z get_loggerr*loggerr r/__all__rrrr)s       b