% &h3fddlmZmZddlZddlZddlmZddlmZddlm Z ddl m Z m Z dd l mZmZmZmZd d lmZe j(eZd ZGd dej0ZGddeZdZe deGddeZdZe deGddeZGddeZgdZ y))OptionalTupleN)nn)ACT2FN)Cache)add_start_docstringslogging)GraniteMoeDecoderLayerGraniteMoeForCausalLMGraniteMoeModelGraniteMoePreTrainedModel)GraniteMoeSharedConfigrc`eZdZdZdeffd ZdejdejfdZxZ S)GraniteMoeSharedMLPz~ MLP layer for shared experts Args: config: Configuration object with model hyperparameters. configchtt| |j|_|j |_t |j|_tj|j|jdzd|_ tj|j|jd|_ y)Nr F)bias) superr__init__ hidden_size input_sizeshared_intermediate_sizer hidden_act activationrLinear input_linear output_linearselfr __class__s /var/www/pru.catia.catastroantioquia-mas.com/valormas/lib/python3.12/site-packages/transformers/models/granitemoeshared/modular_granitemoeshared.pyrzGraniteMoeSharedMLP.__init__1s !413 ,,!:: !2!23IIdoot7G7G!7KRWXYYt'7'7uU hidden_statesreturnc|j|}|jdd}|j|d|dz}|j|}|S)Nr )dimrr)rchunkrr )r"r&chunked_hidden_statess r$forwardzGraniteMoeSharedMLP.forward:s^))-8 - 3 3A2 3 >(=a(@ADYZ[D\\ **=9 r%) __name__ __module__ __qualname____doc__rrtorchTensorr- __classcell__r#s@r$rr(s2V5VU\\ellr%rcreZdZdedeffd Z ddejdeejdeejdee dee d ee d eejd ee d ee ejejfd e ejee ejejfffdZxZS)GraniteMoeSharedDecoderLayerr layer_idxctt||||jdk(rd|_yt||_y)Nr)rrrr shared_mlpr"rr8r#s r$rz%GraniteMoeSharedDecoderLayer.__init__Cs3 +"("A"AQ"F$L_`fLgr%r&attention_mask position_idspast_key_valueoutput_attentions use_cachecache_positionoutput_router_logitsposition_embeddingsr'c ~|} |j|}|jd|||||||| d| \}} } | ||jzz}|} |j|}|j |\}}|j |}n||j |z}~| ||jzz}|f}|r|| fz }|r|| fz }|r||fz }|S)a Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1, query_sequence_length, key_sequence_length)` if default attention is used. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. use_cache (`bool`, *optional*): If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see `past_key_values`). past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence output_router_logits (`bool`, *optional*): Whether or not to return the logits of all the routers. They are useful for computing the router loss, and should not be returned during inference. position_embeddings (`Tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model )r&r<r=r>r?r@rArC)input_layernorm self_attnresidual_multiplierpost_attention_layernormblock_sparse_moer:)r"r&r<r=r>r?r@rArBrCkwargsresidualself_attn_weightspresent_key_valuemoe_hidden_states router_logitsoutputss r$r-z$GraniteMoeSharedDecoderLayer.forwardGsL!,,]; ?Mdnn ? ')%)/) 3 ?  ? ; (*;!=43K3K#KK !55mD +/+@+@+O(= ?? "-M- 0NNM  =43K3K#KK  "  )+ +G  )+ +G   ' 'Gr%)NNNFFNFN)r.r/r0rintrr2r3r LongTensorrboolr FloatTensorr-r4r5s@r$r7r7Bsh5h#h2637*.,1$)59/4KOR||R!.Ru//0 R ! R $D> RD>R!!1!12R'tnR&eELL%,,,F&GHR u  (51B1BEDUDU1U+V"WW XRr%r7aU This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. Parameters: config ([`GraniteMoeSharedConfig`]): Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights. z^The bare GraniteMoeShared Model outputting raw hidden-states without any specific head on top.ceZdZeZdgZy)GraniteMoeSharedPreTrainedModelr7N)r.r/r0r config_class_no_split_modulesrEr%r$rWrWs *L78r%rWa Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it. Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and [`PreTrainedTokenizer.__call__`] for details. [What are input IDs?](../glossary#input-ids) attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*): Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and [`PreTrainedTokenizer.__call__`] for details. If `past_key_values` is used, optionally only the last `input_ids` have to be input (see `past_key_values`). If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`] and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more information on the default strategy. - 1 indicates the head is **not masked**, - 0 indicates the head is **masked**. position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) past_key_values (`Cache` or `tuple(tuple(torch.FloatTensor))`, *optional*): Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the `past_key_values` returned by the model at a previous stage of decoding, when `use_cache=True` or `config.use_cache=True`. Two formats are allowed: - a [`~cache_utils.Cache`] instance; - Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape `(batch_size, num_heads, sequence_length, embed_size_per_head)`). This is also known as the legacy cache format. The model will output the same cache format that is fed as input. If no `past_key_values` are passed, the legacy cache format will be returned. If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids` of shape `(batch_size, sequence_length)`. inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*): Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. use_cache (`bool`, *optional*): If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see `past_key_values`). output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. output_hidden_states (`bool`, *optional*): Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. return_dict (`bool`, *optional*): Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. Contrarily to `position_ids`, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. c(eZdZdZdeffd ZxZS)GraniteMoeSharedModelz Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`GraniteMoeDecoderLayer`] Args: config: GraniteMoeSharedConfig rc t||tjt |j Dcgc]}t ||c}|_ycc}wN)rrr ModuleListrangenum_hidden_layersr7layersr;s r$rzGraniteMoeSharedModel.__init__ sE  mmNSTZTlTlNm n )&) < n  nsA)r.r/r0r1rrr4r5s@r$r[r[s  5  r%r[c*eZdZdgZdeffd ZxZS)GraniteMoeSharedForCausalLMzlm_head.weightrcdt||t||_|j yr])rrr[model post_initr!s r$rz$GraniteMoeSharedForCausalLM.__init__s&  *62  r%)r.r/r0_tied_weights_keysrrr4r5s@r$rcrcs*+5r%rc)rcr[rW)!typingrrr2torch.utils.checkpointr activationsr cache_utilsrutilsr r granitemoe.modeling_granitemoer r rrconfiguration_granitemoesharedr get_loggerr.logger_CONFIG_FOR_DOCModulerr7 GRANITEMOESHARED_START_DOCSTRINGrW!GRANITEMOESHARED_INPUTS_DOCSTRINGr[rc__all__rEr%r$rvs # ! 2 C   H %+"))4W#9Wt$ "d$9&?9 9 G%!Td$  O    "7 fr%