i-zdZddlZddlZddlmZddlmZddlmZedeGd d eZ d gZ y) zUniSpeech model configurationN)strict)PreTrainedConfig)auto_docstringz"microsoft/unispeech-large-1500h-cv) checkpointcleZdZUdZdZdZeed<dZeed<dZ eed<dZ eed <d Z eed <d Z e ed <dZeezed<dZeezed<dZeezed<dZeezed<dZeezed<dZeezed<dZeezed<dZeed<dZeed<dZe ed<d Ze ed<dZeeeedfzed <d!Zeeeedfzed"<d#Zeeeedfzed$<d%Ze ed&<d'Z!eed(<d)Z"eed*<d%Z#e ed+<d,Z$e ed-<d.Z%eezed/<d0Z&eed1<d2Z'eed3<dZ(eezed4<d0Z)eed5<d6Z*eed7<d8Z+eed9<d2Z,eed:<dZ-eed;<dZ/eed?<d>Z0eed@<dZ1eedA<dBZ2e edC<d%Z3e edD<d%Z4e edE<d>Z5eedF<dGZ6eedH<d6Z7edIzedJ<dKZ8edIzedL<d2Z9eeezdIzedM<dNZ:eezedO<fdPZ;dQZxZ?S)SUniSpeechConfiga` feat_proj_dropout (`float`, *optional*, defaults to 0.0): The dropout probability for output of the feature encoder. feat_quantizer_dropout (`float`, *optional*, defaults to 0.0): The dropout probability for the output of the feature encoder that's used by the quantizer. final_dropout (`float`, *optional*, defaults to 0.1): The dropout probability for the final projection layer of [`UniSpeechForCTC`]. feat_extract_norm (`str`, *optional*, defaults to `"group"`): The norm to be applied to 1D convolutional layers in feature encoder. One of `"group"` for group normalization of only the first 1D convolutional layer or `"layer"` for layer normalization of all 1D convolutional layers. feat_extract_activation (`str, *optional*, defaults to `"gelu"`): The non-linear activation function (function or string) in the 1D convolutional layers of the feature extractor. If string, `"gelu"`, `"relu"`, `"selu"` and `"gelu_new"` are supported. conv_dim (`tuple[int]` or `list[int]`, *optional*, defaults to `(512, 512, 512, 512, 512, 512, 512)`): A tuple of integers defining the number of input and output channels of each 1D convolutional layer in the feature encoder. The length of *conv_dim* defines the number of 1D convolutional layers. conv_stride (`tuple[int]` or `list[int]`, *optional*, defaults to `(5, 2, 2, 2, 2, 2, 2)`): A tuple of integers defining the stride of each 1D convolutional layer in the feature encoder. The length of *conv_stride* defines the number of convolutional layers and has to match the length of *conv_dim*. conv_kernel (`tuple[int]` or `list[int]`, *optional*, defaults to `(10, 3, 3, 3, 3, 2, 2)`): A tuple of integers defining the kernel size of each 1D convolutional layer in the feature encoder. The length of *conv_kernel* defines the number of convolutional layers and has to match the length of *conv_dim*. conv_bias (`bool`, *optional*, defaults to `False`): Whether the 1D convolutional layers have a bias. num_conv_pos_embeddings (`int`, *optional*, defaults to 128): Number of convolutional positional embeddings. Defines the kernel size of 1D convolutional positional embeddings layer. num_conv_pos_embedding_groups (`int`, *optional*, defaults to 16): Number of groups of 1D convolutional positional embeddings layer. do_stable_layer_norm (`bool`, *optional*, defaults to `False`): Whether to apply *stable* layer norm architecture of the Transformer encoder. `do_stable_layer_norm is True` corresponds to applying layer norm before the attention layer, whereas `do_stable_layer_norm is False` corresponds to applying layer norm after the attention layer. apply_spec_augment (`bool`, *optional*, defaults to `True`): Whether to apply *SpecAugment* data augmentation to the outputs of the feature encoder. For reference see [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://huggingface.co/papers/1904.08779). mask_time_prob (`float`, *optional*, defaults to 0.05): Percentage (between 0 and 1) of all feature vectors along the time axis which will be masked. The masking procedure generates ''mask_time_prob*len(time_axis)/mask_time_length'' independent masks over the axis. If reasoning from the probability of each feature vector to be chosen as the start of the vector span to be masked, *mask_time_prob* should be `prob_vector_start*mask_time_length`. Note that overlap may decrease the actual percentage of masked vectors. This is only relevant if `apply_spec_augment is True`. mask_time_length (`int`, *optional*, defaults to 10): Length of vector span along the time axis. mask_time_min_masks (`int`, *optional*, defaults to 2): The minimum number of masks of length `mask_feature_length` generated along the time axis, each time step, irrespectively of `mask_feature_prob`. Only relevant if ''mask_time_prob*len(time_axis)/mask_time_length < mask_time_min_masks'' mask_feature_prob (`float`, *optional*, defaults to 0.0): Percentage (between 0 and 1) of all feature vectors along the feature axis which will be masked. The masking procedure generates ''mask_feature_prob*len(feature_axis)/mask_time_length'' independent masks over the axis. If reasoning from the probability of each feature vector to be chosen as the start of the vector span to be masked, *mask_feature_prob* should be `prob_vector_start*mask_feature_length`. Note that overlap may decrease the actual percentage of masked vectors. This is only relevant if `apply_spec_augment is True`. mask_feature_length (`int`, *optional*, defaults to 10): Length of vector span along the feature axis. mask_feature_min_masks (`int`, *optional*, defaults to 0): The minimum number of masks of length `mask_feature_length` generated along the feature axis, each time step, irrespectively of `mask_feature_prob`. Only relevant if ''mask_feature_prob*len(feature_axis)/mask_feature_length < mask_feature_min_masks'' num_codevectors_per_group (`int`, *optional*, defaults to 320): Number of entries in each quantization codebook (group). num_codevector_groups (`int`, *optional*, defaults to 2): Number of codevector groups for product codevector quantization. contrastive_logits_temperature (`float`, *optional*, defaults to 0.1): The temperature *kappa* in the contrastive loss. num_negatives (`int`, *optional*, defaults to 100): Number of negative samples for the contrastive loss. codevector_dim (`int`, *optional*, defaults to 256): Dimensionality of the quantized feature vectors. proj_codevector_dim (`int`, *optional*, defaults to 256): Dimensionality of the final projection of both the quantized and the transformer features. diversity_loss_weight (`int`, *optional*, defaults to 0.1): The weight of the codebook diversity loss component. ctc_zero_infinity (`bool`, *optional*, defaults to `False`): Whether to zero infinite losses and the associated gradients of `torch.nn.CTCLoss`. Infinite losses mainly occur when the inputs are too short to be aligned to the targets. Only relevant when training an instance of [`UniSpeechForCTC`]. use_weighted_layer_sum (`bool`, *optional*, defaults to `False`): Whether to use a weighted average of layer outputs with learned weights. Only relevant when using an instance of [`UniSpeechForSequenceClassification`]. classifier_proj_size (`int`, *optional*, defaults to 256): Dimensionality of the projection before token mean-pooling for classification. num_ctc_classes (`int`, *optional*, defaults to 80): Specifies the number of classes (phoneme tokens and blank token) for phoneme-level CTC loss. Only relevant when using an instance of [`UniSpeechForPreTraining`]. replace_prob (`float`, *optional*, defaults to 0.5): Probability that transformer feature is replaced by quantized feature for pretraining. Example: ```python >>> from transformers import UniSpeechConfig, UniSpeechModel >>> # Initializing a UniSpeech facebook/unispeech-base-960h style configuration >>> configuration = UniSpeechConfig() >>> # Initializing a model (with random weights) from the facebook/unispeech-base-960h style configuration >>> model = UniSpeechModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` unispeech vocab_sizei hidden_size num_hidden_layersnum_attention_headsi intermediate_sizegelu hidden_actg?hidden_dropoutactivation_dropoutattention_dropoutgfeat_proj_dropoutfeat_quantizer_dropout final_dropout layerdropg{Gz?initializer_rangegh㈵>layer_norm_epsgroupfeat_extract_normfeat_extract_activation)r r r r r r .conv_dim)r#r#r#r#r# conv_stride) rrrrr#r# conv_kernelF conv_biasnum_conv_pos_embeddingsnum_conv_pos_embedding_groupsdo_stable_layer_normTapply_spec_augmentg?mask_time_probr%mask_time_lengthr#mask_time_min_masksmask_feature_probmask_feature_lengthrmask_feature_min_masksi@num_codevectors_per_groupnum_codevector_groupscontrastive_logits_temperatured num_negativescodevector_dimproj_codevector_dimdiversity_loss_weightmeanctc_loss_reductionctc_zero_infinityuse_weighted_layer_sumclassifier_proj_sizePnum_ctc_classesN pad_token_id bos_token_id eos_token_idg? replace_probc Vt|j|_t|di|S)N)lenr!num_feat_extract_layerssuper __post_init__)selfkwargs __class__s /var/www/vps2.regionflexible.com/Desarrollo/venv/lib/python3.12/site-packages/transformers/models/unispeech/configuration_unispeech.pyrNzUniSpeechConfig.__post_init__s''*4=='9$w$.v..c lt|j|jk7sDt|j|jk7s"t|j|jk7rNt dt|jdt|jdt|jdy)zOPart of `@strict`-powered validation. Validates the architecture of the config.zConfiguration for convolutional layers is incorrect. It is required that `len(config.conv_dim)` == `len(config.conv_stride)` == `len(config.conv_kernel)`, but is `len(config.conv_dim) = z`, `len(config.conv_stride) = z`, `len(config.conv_kernel) = z`.N)rKr$rLr&r! ValueErrorrOs rRvalidate_architecturez%UniSpeechConfig.validate_architectures!! "d&B&B BD$$%)E)EEDMM"d&B&BB &''Ec$JZJZF[E\]//243C3C/D.ERI CrSc`tjtj|jdS)NrE) functoolsreduceoperatormulr$rVs rRinputs_to_logits_ratioz&UniSpeechConfig.inputs_to_logits_ratios! d.>.>BBrS)@__name__ __module__ __qualname____doc__ model_typer int__annotations__r rrrrstrrfloatrrrrrrrrrrr!listtupler$r&r'boolr)r+r,r-r.r/r0r1r2r3r4r5r6r8r:r;r<r>r?r@rArCrDrFrGrHrNrWpropertyr] __classcell__)rQs@rRr r sj XJJKs!!!s!J"%NECK%&) )%(us{(%(us{(*-ECK-!$M53;$ Ius{ #u# NE $s$#)S),OHd3i%S/)O/DKcU38_,D/EKcU38_,EIt#&S&)+!3+!&$&##"&NECK&c  %(us{(!!"#C#%(s(!"3",/"E/M3NC""#&5&$$#t##(D( ###OS L#*  L#* +,L#S /D(, #L%#+#/ CCrSr ) rarYr[huggingface_hub.dataclassesrconfiguration_utilsrutilsrr __all__rJrSrRrpsU$.3#?@sC&sCAsCl  rS