T^(hBdZddlmZddlmZddlmZddlmZeje Z GddeZ Gd d eZ dd gZy ) zmT5 model configuration)Mapping)PretrainedConfig)OnnxSeq2SeqConfigWithPast)loggingcfeZdZdZdZdgZdddddZ d fd ZxZS) MT5Configa7 This is the configuration class to store the configuration of a [`MT5Model`] or a [`TFMT5Model`]. It is used to instantiate a mT5 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 mT5 [google/mt5-small](https://huggingface.co/google/mt5-small) architecture. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Arguments: vocab_size (`int`, *optional*, defaults to 250112): Vocabulary size of the T5 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`T5Model`] or [`TFT5Model`]. d_model (`int`, *optional*, defaults to 512): Size of the encoder layers and the pooler layer. d_kv (`int`, *optional*, defaults to 64): Size of the key, query, value projections per attention head. In the conventional context, it is typically expected that `d_kv` has to be equal to `d_model // num_heads`. But in the architecture of mt5-small, `d_kv` is not equal to `d_model //num_heads`. The `inner_dim` of the projection layer will be defined as `num_heads * d_kv`. d_ff (`int`, *optional*, defaults to 1024): Size of the intermediate feed forward layer in each `T5Block`. num_layers (`int`, *optional*, defaults to 8): Number of hidden layers in the Transformer encoder. num_decoder_layers (`int`, *optional*): Number of hidden layers in the Transformer decoder. Will use the same value as `num_layers` if not set. num_heads (`int`, *optional*, defaults to 6): Number of attention heads for each attention layer in the Transformer encoder. relative_attention_num_buckets (`int`, *optional*, defaults to 32): The number of buckets to use for each attention layer. relative_attention_max_distance (`int`, *optional*, defaults to 128): The maximum distance of the longer sequences for the bucket separation. dropout_rate (`float`, *optional*, defaults to 0.1): The ratio for all dropout layers. classifier_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for classifier. layer_norm_eps (`float`, *optional*, defaults to 1e-6): The epsilon used by the layer normalization layers. initializer_factor (`float`, *optional*, defaults to 1): A factor for initializing all weight matrices (should be kept to 1, used internally for initialization testing). feed_forward_proj (`string`, *optional*, defaults to `"gated-gelu"`): Type of feed forward layer to be used. Should be one of `"relu"` or `"gated-gelu"`. use_cache (`bool`, *optional*, defaults to `True`): Whether or not the model should return the last key/values attentions (not used by all models). mt5past_key_valuesd_model num_heads num_layersd_kv) hidden_sizenum_attention_headsnum_hidden_layershead_dimc ||_||_||_||_||_||n |j|_||_||_| |_| |_ ||_ | |_ | |_ | |_ ||_|jjd}|d|_|ddk(|_t%|dkDr|ddk7st%|dkDrt'd| d| d k(rd |_t)|Td ||||||d |y) N-rgatedz`feed_forward_proj`: z is not a valid activation function of the dense layer. Please make sure `feed_forward_proj` is of the format `gated-{ACT_FN}` or `{ACT_FN}`, e.g. 'gated-gelu' or 'relu' gated-gelugelu_new)is_encoder_decodertokenizer_classtie_word_embeddings pad_token_id eos_token_iddecoder_start_token_id) vocab_sizer rd_ffrnum_decoder_layersr relative_attention_num_bucketsrelative_attention_max_distance dropout_rateclassifier_dropoutlayer_norm_epsiloninitializer_factorfeed_forward_proj use_cachesplit dense_act_fn is_gated_actlen ValueErrorsuper__init__)selfr#r rr$rr%r r&r'r(r*r+r,rr-rrrr r!r)kwargsact_info __class__s g/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/models/mt5/configuration_mt5.pyr4zMT5Config.__init__Rs<2%   $"4"@ doo #.L+/N,("4"4"4!2"))//4$RL$QK72 x=1 !!73x=1;L'(9':;))   , *D   1+ 3%%#9   )ii@iN g?gư>g?rTT T5TokenizerFrrrg) __name__ __module__ __qualname____doc__ model_typekeys_to_ignore_at_inference attribute_mapr4 __classcell__)r8s@r9r r sy+ZJ#4"5 *) M  ')(+&%! -B B r:r cbeZdZedeeeeefffdZedefdZede fdZ y) MT5OnnxConfigreturncddddddd}|jrd|dd<ddi|d <dd d|d <ndd d|d <dd d|d <|jr|j|d |S)Nbatchencoder_sequence)rr) input_idsattention_maskz past_encoder_sequence + sequencerPrrdecoder_input_idsz past_decoder_sequence + sequencedecoder_attention_maskdecoder_sequenceinputs) direction)use_pastfill_with_past_key_values_)r5 common_inputss r9rTzMT5OnnxConfig.inputss%);<").@A  ==1SM* +A .23WM- .:AFh6iM2 35rjsS3-   H %y  y x->  (r: