o Zh@sjdZddlmZddlmZddlmZddlmZe e Z GdddeZ Gd d d eZ dd gZd S) zLongT5 model configuration)Mapping)PretrainedConfig)OnnxSeq2SeqConfigWithPast)loggingcsbeZdZdZdZdgZdddddZ               dfdd ZZS) LongT5Configa This is the configuration class to store the configuration of a [`LongT5Model`] or a [`FlaxLongT5Model`]. It is used to instantiate a LongT5 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 LongT5 [google/long-t5-local-base](https://huggingface.co/google/long-t5-local-base) 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 32128): Vocabulary size of the LongT5 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`LongT5Model`]. 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. `d_kv` has to be equal to `d_model // num_heads`. d_ff (`int`, *optional*, defaults to 2048): Size of the intermediate feed forward layer in each `LongT5Block`. num_layers (`int`, *optional*, defaults to 6): 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 8): Number of attention heads for each attention layer in the Transformer encoder. local_radius (`int`, *optional*, defaults to 127) Number of tokens to the left/right for each token to locally self-attend in a local attention mechanism. global_block_size (`int`, *optional*, defaults to 16) Length of blocks an input sequence is divided into for a global token representation. Used only for `encoder_attention_type = "transient-global"`. 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. 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 `"relu"`): Type of feed forward layer to be used. Should be one of `"relu"` or `"gated-gelu"`. LongT5v1.1 uses the `"gated-gelu"` feed forward projection. Original LongT5 implementation uses `"gated-gelu"`. encoder_attention_type (`string`, *optional*, defaults to `"local"`): Type of encoder attention to be used. Should be one of `"local"` or `"transient-global"`, which are supported by LongT5 implementation. use_cache (`bool`, *optional*, defaults to `True`): Whether or not the model should return the last key/values attentions (not used by all models). Zlongt5Zpast_key_valuesd_model num_heads num_layersd_kv)Z hidden_sizeZnum_attention_headsZnum_hidden_layersZhead_dim}@N 皙?ư>?reluTlocalrc s||_||_||_||_||_|dur|n|j|_||_||_| |_| |_ | |_ | |_ | |_ ||_ ||_||_||_|jd}|d|_|ddk|_t|dkrX|ddks^t|dkrftd|d|d krmd |_tjd |||d |dS) N-rZgatedrz`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'z gated-geluZgelu_new) pad_token_id eos_token_idis_encoder_decoder) vocab_sizerr d_ffr num_decoder_layersr local_radiusglobal_block_sizerelative_attention_num_bucketsrelative_attention_max_distance dropout_ratelayer_norm_epsiloninitializer_factorfeed_forward_projencoder_attention_type use_cachesplitZ dense_act_fnZ is_gated_actlen ValueErrorsuper__init__)selfr#rr r$r r%r r&r'r(r)r*r+r,r-r!r.r/rr kwargsZact_info __class__r"^/var/www/auris/lib/python3.10/site-packages/transformers/models/longt5/configuration_longt5.pyr4YsB  $  zLongT5Config.__init__)r r rrrNrrrrrrrrrTrTrr) __name__ __module__ __qualname____doc__Z model_typeZkeys_to_ignore_at_inferenceZ attribute_mapr4 __classcell__r"r"r7r9rs<4 rc@s@eZdZedeeeeefffddZedefddZdS)LongT5OnnxConfigreturncCsxddddddd}|jr"d|dd<ddi|d <dd d|d <ndd d|d <dd d|d <|jr:|j|d d|S)NbatchZencoder_sequence)rr)Z input_idsattention_maskz past_encoder_sequence + sequencerBrrZdecoder_input_idsz past_decoder_sequence + sequenceZdecoder_attention_maskZdecoder_sequenceinputs) direction)Zuse_pastZfill_with_past_key_values_)r5Z common_inputsr"r"r9rCs  zLongT5OnnxConfig.inputscCsdS)N r")r5r"r"r9default_onnx_opsetsz#LongT5OnnxConfig.default_onnx_opsetN) r:r;r<propertyrstrintrCrFr"r"r"r9r?s  r?N)r=typingrZconfiguration_utilsrZonnxrutilsrZ get_loggerr:loggerrr?__all__r"r"r"r9s