o Zh@sdZddlmZddlmZddlmZddlmZddl m Z ddl m Z e eZGd d d eZGd d d e Zd d gZd S)zMobileViTV2 model configuration OrderedDict)Mapping)version)PretrainedConfig) OnnxConfig)loggingcsZeZdZdZdZdddddddd d d d gd d dgdgdddddffdd ZZS)MobileViTV2Configa This is the configuration class to store the configuration of a [`MobileViTV2Model`]. It is used to instantiate a MobileViTV2 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 MobileViTV2 [apple/mobilevitv2-1.0](https://huggingface.co/apple/mobilevitv2-1.0) architecture. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: num_channels (`int`, *optional*, defaults to 3): The number of input channels. image_size (`int`, *optional*, defaults to 256): The size (resolution) of each image. patch_size (`int`, *optional*, defaults to 2): The size (resolution) of each patch. expand_ratio (`float`, *optional*, defaults to 2.0): Expansion factor for the MobileNetv2 layers. hidden_act (`str` or `function`, *optional*, defaults to `"swish"`): The non-linear activation function (function or string) in the Transformer encoder and convolution layers. conv_kernel_size (`int`, *optional*, defaults to 3): The size of the convolutional kernel in the MobileViTV2 layer. output_stride (`int`, *optional*, defaults to 32): The ratio of the spatial resolution of the output to the resolution of the input image. classifier_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout ratio for attached classifiers. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. layer_norm_eps (`float`, *optional*, defaults to 1e-05): The epsilon used by the layer normalization layers. aspp_out_channels (`int`, *optional*, defaults to 512): Number of output channels used in the ASPP layer for semantic segmentation. atrous_rates (`List[int]`, *optional*, defaults to `[6, 12, 18]`): Dilation (atrous) factors used in the ASPP layer for semantic segmentation. aspp_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout ratio for the ASPP layer for semantic segmentation. semantic_loss_ignore_index (`int`, *optional*, defaults to 255): The index that is ignored by the loss function of the semantic segmentation model. n_attn_blocks (`List[int]`, *optional*, defaults to `[2, 4, 3]`): The number of attention blocks in each MobileViTV2Layer base_attn_unit_dims (`List[int]`, *optional*, defaults to `[128, 192, 256]`): The base multiplier for dimensions of attention blocks in each MobileViTV2Layer width_multiplier (`float`, *optional*, defaults to 1.0): The width multiplier for MobileViTV2. ffn_multiplier (`int`, *optional*, defaults to 2): The FFN multiplier for MobileViTV2. attn_dropout (`float`, *optional*, defaults to 0.0): The dropout in the attention layer. ffn_dropout (`float`, *optional*, defaults to 0.0): The dropout between FFN layers. Example: ```python >>> from transformers import MobileViTV2Config, MobileViTV2Model >>> # Initializing a mobilevitv2-small style configuration >>> configuration = MobileViTV2Config() >>> # Initializing a model from the mobilevitv2-small style configuration >>> model = MobileViTV2Model(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```Z mobilevitv2rg@Zswish g?g{Gz?gh㈵>i) )r r)r g?gc stjdi|||_||_||_||_||_||_||_| |_ | |_ ||_ ||_ ||_ ||_||_||_||_| |_| |_| |_||_dS)N)super__init__ num_channels image_size patch_size expand_ratio hidden_actconv_kernel_size output_strideinitializer_rangelayer_norm_eps n_attn_blocksbase_attn_unit_dimswidth_multiplierffn_multiplier ffn_dropout attn_dropoutclassifier_dropout_probaspp_out_channels atrous_ratesaspp_dropout_probsemantic_loss_ignore_index)selfrrrrrrrr'rr r(r)r*r+r!r"r#r$r&r%kwargs __class__rh/var/www/auris/lib/python3.10/site-packages/transformers/models/mobilevitv2/configuration_mobilevitv2.pyrcs* zMobileViTV2Config.__init__)__name__ __module__ __qualname____doc__Z model_typer __classcell__rrr.r0r s0Br c@sleZdZedZedeeee efffddZ edeeee efffddZ ede fddZ d S) MobileViTV2OnnxConfigz1.11returncCstddddddfgS)NZ pixel_valuesbatchrheightwidth)rr rrr,rrr0inputsszMobileViTV2OnnxConfig.inputscCs8|jdkrtdddifgStdddifdddifgS)Nzimage-classificationZlogitsrr8Zlast_hidden_stateZ pooler_output)taskrr<rrr0outputss zMobileViTV2OnnxConfig.outputscCsdS)Ng-C6?rr<rrr0atol_for_validationsz)MobileViTV2OnnxConfig.atol_for_validationN)r1r2r3rparseZtorch_onnx_minimum_versionpropertyrstrintr=r?floatr@rrrr0r6s   r6N)r4 collectionsrtypingr packagingrZconfiguration_utilsrZonnxrutilsr Z get_loggerr1loggerr r6__all__rrrr0s       w