Pegasus
Overview
The Pegasus model was proposed in PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu on Dec 18, 2019.
According to the abstract,
- Pegasusβ pretraining task is intentionally similar to summarization: important sentences are removed/masked from an input document and are generated together as one output sequence from the remaining sentences, similar to an extractive summary.
- Pegasus achieves SOTA summarization performance on all 12 downstream tasks, as measured by ROUGE and human eval.
This model was contributed by sshleifer. The Authorsβ code can be found here.
Usage tips
Sequence-to-sequence model with the same encoder-decoder model architecture as BART. Pegasus is pre-trained jointly on two self-supervised objective functions: Masked Language Modeling (MLM) and a novel summarization specific pretraining objective, called Gap Sentence Generation (GSG).
- MLM: encoder input tokens are randomly replaced by a mask tokens and have to be predicted by the encoder (like in BERT)
- GSG: whole encoder input sentences are replaced by a second mask token and fed to the decoder, but which has a causal mask to hide the future words like a regular auto-regressive transformer decoder.
FP16 is not supported (help/ideas on this appreciated!).
The adafactor optimizer is recommended for pegasus fine-tuning.
Checkpoints
All the checkpoints are fine-tuned for summarization, besides pegasus-large, whence the other checkpoints are fine-tuned:
- Each checkpoint is 2.2 GB on disk and 568M parameters.
- FP16 is not supported (help/ideas on this appreciated!).
- Summarizing xsum in fp32 takes about 400ms/sample, with default parameters on a v100 GPU.
- Full replication results and correctly pre-processed data can be found in this Issue.
- Distilled checkpoints are described in this paper.
Implementation Notes
- All models are transformer encoder-decoders with 16 layers in each component.
- The implementation is completely inherited from BartForConditionalGeneration
- Some key configuration differences:
- static, sinusoidal position embeddings
- the model starts generating with pad_token_id (which has 0 token_embedding) as the prefix.
- more beams are used (
num_beams=8
)
- All pretrained pegasus checkpoints are the same besides three attributes:
tokenizer.model_max_length
(maximum input size),max_length
(the maximum number of tokens to generate) andlength_penalty
. - The code to convert checkpoints trained in the authorβs repo can be
found in
convert_pegasus_tf_to_pytorch.py
.
Usage Example
>>> from transformers import PegasusForConditionalGeneration, PegasusTokenizer
>>> import torch
>>> src_text = [
... """ PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow."""
... ]
... model_name = "google/pegasus-xsum"
... device = "cuda" if torch.cuda.is_available() else "cpu"
... tokenizer = PegasusTokenizer.from_pretrained(model_name)
... model = PegasusForConditionalGeneration.from_pretrained(model_name).to(device)
... batch = tokenizer(src_text, truncation=True, padding="longest", return_tensors="pt").to(device)
... translated = model.generate(**batch)
... tgt_text = tokenizer.batch_decode(translated, skip_special_tokens=True)
... assert (
... tgt_text[0]
... == "California's largest electricity provider has turned off power to hundreds of thousands of customers."
... )
Resources
- Script to fine-tune pegasus on the XSUM dataset. Data download instructions at examples/pytorch/summarization/.
- Causal language modeling task guide
- Translation task guide
- Summarization task guide
PegasusConfig
class transformers.PegasusConfig
< source >( vocab_size = 50265 max_position_embeddings = 1024 encoder_layers = 12 encoder_ffn_dim = 4096 encoder_attention_heads = 16 decoder_layers = 12 decoder_ffn_dim = 4096 decoder_attention_heads = 16 encoder_layerdrop = 0.0 decoder_layerdrop = 0.0 use_cache = True is_encoder_decoder = True activation_function = 'gelu' d_model = 1024 dropout = 0.1 attention_dropout = 0.0 activation_dropout = 0.0 init_std = 0.02 decoder_start_token_id = 0 scale_embedding = False pad_token_id = 0 eos_token_id = 1 forced_eos_token_id = 1 **kwargs )
Parameters
- vocab_size (
int
, optional, defaults to 50265) — Vocabulary size of the PEGASUS model. Defines the number of different tokens that can be represented by theinputs_ids
passed when calling PegasusModel or TFPegasusModel. - d_model (
int
, optional, defaults to 1024) — Dimensionality of the layers and the pooler layer. - encoder_layers (
int
, optional, defaults to 12) — Number of encoder layers. - decoder_layers (
int
, optional, defaults to 12) — Number of decoder layers. - encoder_attention_heads (
int
, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer encoder. - decoder_attention_heads (
int
, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer decoder. - decoder_ffn_dim (
int
, optional, defaults to 4096) — Dimensionality of the “intermediate” (often named feed-forward) layer in decoder. - encoder_ffn_dim (
int
, optional, defaults to 4096) — Dimensionality of the “intermediate” (often named feed-forward) layer in decoder. - activation_function (
str
orfunction
, optional, defaults to"gelu"
) — The non-linear activation function (function or string) in the encoder and pooler. If string,"gelu"
,"relu"
,"silu"
and"gelu_new"
are supported. - dropout (
float
, optional, defaults to 0.1) — The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. - attention_dropout (
float
, optional, defaults to 0.0) — The dropout ratio for the attention probabilities. - activation_dropout (
float
, optional, defaults to 0.0) — The dropout ratio for activations inside the fully connected layer. - max_position_embeddings (
int
, optional, defaults to 1024) — The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048). - init_std (
float
, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices. - encoder_layerdrop (
float
, optional, defaults to 0.0) — The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more details. - decoder_layerdrop (
float
, optional, defaults to 0.0) — The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more details. - scale_embedding (
bool
, optional, defaults toFalse
) — Scale embeddings by diving by sqrt(d_model). - use_cache (
bool
, optional, defaults toTrue
) — Whether or not the model should return the last key/values attentions (not used by all models) - forced_eos_token_id (
int
, optional, defaults to 1) — The id of the token to force as the last generated token whenmax_length
is reached. Usually set toeos_token_id
.
This is the configuration class to store the configuration of a PegasusModel. It is used to instantiate an PEGASUS 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 PEGASUS google/pegasus-large architecture.
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
Example:
>>> from transformers import PegasusConfig, PegasusModel
>>> # Initializing a PEGASUS google/pegasus-large style configuration
>>> configuration = PegasusConfig()
>>> # Initializing a model (with random weights) from the google/pegasus-large style configuration
>>> model = PegasusModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
PegasusTokenizer
warning: add_tokens
does not work at the moment.
class transformers.PegasusTokenizer
< source >( vocab_file pad_token = '<pad>' eos_token = '</s>' unk_token = '<unk>' mask_token = '<mask_2>' mask_token_sent = '<mask_1>' additional_special_tokens = None offset = 103 sp_model_kwargs: typing.Optional[typing.Dict[str, typing.Any]] = None **kwargs )
Parameters
- vocab_file (
str
) — SentencePiece file (generally has a .spm extension) that contains the vocabulary necessary to instantiate a tokenizer. - pad_token (
str
, optional, defaults to"<pad>"
) — The token used for padding, for example when batching sequences of different lengths. - eos_token (
str
, optional, defaults to"</s>"
) — The end of sequence token.When building a sequence using special tokens, this is not the token that is used for the end of sequence. The token used is the
sep_token
. - unk_token (
str
, optional, defaults to"<unk>"
) — The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead. - mask_token (
str
, optional, defaults to"<mask_2>"
) — The token used for masking single token values. This is the token used when training this model with masked language modeling (MLM). This is the token that the PEGASUS encoder will try to predict during pretraining. It corresponds to [MASK2] in PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization. - mask_token_sent (
str
, optional, defaults to"<mask_1>"
) — The token used for masking whole target sentences. This is the token used when training this model with gap sentences generation (GSG). This is the sentence that the PEGASUS decoder will try to predict during pretraining. It corresponds to [MASK1] in PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization. - additional_special_tokens (
List[str]
, optional) — Additional special tokens used by the tokenizer. If no additional_special_tokens are providedand are used as additional special tokens corresponding to the original PEGASUS tokenizer that uses the tokens 2 - 104 only for pretraining - sp_model_kwargs (
dict
, optional) — Will be passed to theSentencePieceProcessor.__init__()
method. The Python wrapper for SentencePiece can be used, among other things, to set:-
enable_sampling
: Enable subword regularization. -
nbest_size
: Sampling parameters for unigram. Invalid for BPE-Dropout.nbest_size = {0,1}
: No sampling is performed.nbest_size > 1
: samples from the nbest_size results.nbest_size < 0
: assuming that nbest_size is infinite and samples from the all hypothesis (lattice) using forward-filtering-and-backward-sampling algorithm.
-
alpha
: Smoothing parameter for unigram sampling, and dropout probability of merge operations for BPE-dropout.
-
Construct a PEGASUS tokenizer. Based on SentencePiece.
This tokenizer inherits from PreTrainedTokenizer which contains most of the main methods. Users should refer to this superclass for more information regarding those methods.
build_inputs_with_special_tokens
< source >( token_ids_0 token_ids_1 = None ) β List[int]
Parameters
- token_ids_0 (
List[int]
) — List of IDs to which the special tokens will be added. - token_ids_1 (
List[int]
, optional) — Optional second list of IDs for sequence pairs.
Returns
List[int]
List of input IDs with the appropriate special tokens.
Build model inputs from a sequence or a pair of sequences for sequence classification tasks by concatenating
and adding special tokens. A PEGASUS sequence has the following format, where X
represents the sequence:
- single sequence:
X </s>
- pair of sequences:
A B </s>
(not intended use)
BOS is never used. Pairs of sequences are not the expected use case, but they will be handled without a separator.
Converts a sequence of tokens (string) in a single string.
get_special_tokens_mask
< source >( token_ids_0: typing.List token_ids_1: typing.Optional[typing.List] = None already_has_special_tokens: bool = False )
Get list where entries are [1] if a token is [eos] or [pad] else 0.
Just EOS
PegasusTokenizerFast
class transformers.PegasusTokenizerFast
< source >( vocab_file = None tokenizer_file = None pad_token = '<pad>' eos_token = '</s>' unk_token = '<unk>' mask_token = '<mask_2>' mask_token_sent = '<mask_1>' additional_special_tokens = None offset = 103 **kwargs )
Parameters
- vocab_file (
str
) — SentencePiece file (generally has a .spm extension) that contains the vocabulary necessary to instantiate a tokenizer. - pad_token (
str
, optional, defaults to"<pad>"
) — The token used for padding, for example when batching sequences of different lengths. - eos_token (
str
, optional, defaults to"</s>"
) — The end of sequence token.When building a sequence using special tokens, this is not the token that is used for the end of sequence. The token used is the
sep_token
. - unk_token (
str
, optional, defaults to"<unk>"
) — The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead. - mask_token (
str
, optional, defaults to"<mask_2>"
) — The token used for masking single token values. This is the token used when training this model with masked language modeling (MLM). This is the token that the PEGASUS encoder will try to predict during pretraining. It corresponds to [MASK2] in PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization. - mask_token_sent (
str
, optional, defaults to"<mask_1>"
) — The token used for masking whole target sentences. This is the token used when training this model with gap sentences generation (GSG). This is the sentence that the PEGASUS decoder will try to predict during pretraining. It corresponds to [MASK1] in PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization. - additional_special_tokens (
List[str]
, optional) — Additional special tokens used by the tokenizer. If no additional_special_tokens are providedand are used as additional special tokens corresponding to the original PEGASUS tokenizer that uses the tokens 2 - 104 only for pretraining
Construct a βfastβ PEGASUS tokenizer (backed by HuggingFaceβs tokenizers library). Based on Unigram.
This tokenizer inherits from PreTrainedTokenizerFast which contains most of the main methods. Users should refer to this superclass for more information regarding those methods.
build_inputs_with_special_tokens
< source >( token_ids_0 token_ids_1 = None ) β List[int]
Parameters
- token_ids_0 (
List[int]
) — List of IDs to which the special tokens will be added - token_ids_1 (
List[int]
, optional) — Optional second list of IDs for sequence pairs.
Returns
List[int]
list of input IDs with the appropriate special tokens.
Build model inputs from a sequence by adding eos to the end. no bos token is added to the front.
- single sequence:
X </s>
- pair of sequences:
A B </s>
(not intended use)
get_special_tokens_mask
< source >( token_ids_0: typing.List token_ids_1: typing.Optional[typing.List] = None already_has_special_tokens: bool = False )
Get list where entries are [1] if a token is [eos] or [pad] else 0.
PegasusModel
class transformers.PegasusModel
< source >( config: PegasusConfig )
Parameters
- config (PegasusConfig) — 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 from_pretrained() method to load the model weights.
The bare PEGASUS Model outputting raw hidden-states without any specific head on top. 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 subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None decoder_input_ids: typing.Optional[torch.Tensor] = None decoder_attention_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None decoder_head_mask: typing.Optional[torch.Tensor] = None cross_attn_head_mask: typing.Optional[torch.Tensor] = None encoder_outputs: typing.Optional[typing.Tuple[torch.FloatTensor]] = None past_key_values: typing.Optional[typing.Tuple[torch.FloatTensor]] = None inputs_embeds: typing.Optional[torch.Tensor] = None decoder_inputs_embeds: typing.Optional[torch.Tensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) β transformers.modeling_outputs.Seq2SeqModelOutput or tuple(torch.FloatTensor)
Parameters
- 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.
- 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.
- decoder_input_ids (
torch.LongTensor
of shape(batch_size, target_sequence_length)
, optional) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
Pegasus uses the
pad_token_id
as the starting token fordecoder_input_ids
generation. Ifpast_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
). - decoder_attention_mask (
torch.LongTensor
of shape(batch_size, target_sequence_length)
, optional) — Default behavior: generate a tensor that ignores pad tokens indecoder_input_ids
. Causal mask will also be used by default. - head_mask (
torch.Tensor
of shape(encoder_layers, encoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- decoder_head_mask (
torch.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the decoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- cross_attn_head_mask (
torch.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the cross-attention modules in the decoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- encoder_outputs (
tuple(tuple(torch.FloatTensor)
, optional) — Tuple consists of (last_hidden_state
, optional:hidden_states
, optional:attentions
)last_hidden_state
of shape(batch_size, sequence_length, hidden_size)
, optional) is a sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of the decoder. - past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.If
past_key_values
are used, the user can optionally input only the lastdecoder_input_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of alldecoder_input_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - decoder_inputs_embeds (
torch.FloatTensor
of shape(batch_size, target_sequence_length, hidden_size)
, optional) — Optionally, instead of passingdecoder_input_ids
you can choose to directly pass an embedded representation. Ifpast_key_values
is used, optionally only the lastdecoder_inputs_embeds
have to be input (seepast_key_values
). This is useful if you want more control over how to convertdecoder_input_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
decoder_input_ids
anddecoder_inputs_embeds
are both unset,decoder_inputs_embeds
takes the value ofinputs_embeds
. - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_outputs.Seq2SeqModelOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.Seq2SeqModelOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (PegasusConfig) and inputs.
-
last_hidden_state (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
) β Sequence of hidden-states at the output of the last layer of the decoder of the model.If
past_key_values
is used only the last hidden-state of the sequences of shape(batch_size, 1, hidden_size)
is output. -
past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding. -
decoder_hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the decoder at the output of each layer plus the optional initial embedding outputs.
-
decoder_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
encoder_last_hidden_state (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) β Sequence of hidden-states at the output of the last layer of the encoder of the model. -
encoder_hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the encoder at the output of each layer plus the optional initial embedding outputs.
-
encoder_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The PegasusModel forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoTokenizer, PegasusModel
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> model = PegasusModel.from_pretrained("google/pegasus-large")
>>> inputs = tokenizer("Studies have been shown that owning a dog is good for you", return_tensors="pt")
>>> decoder_inputs = tokenizer("Studies show that", return_tensors="pt")
>>> outputs = model(input_ids=inputs.input_ids, decoder_input_ids=decoder_inputs.input_ids)
>>> last_hidden_states = outputs.last_hidden_state
>>> list(last_hidden_states.shape)
[1, 4, 1024]
PegasusForConditionalGeneration
class transformers.PegasusForConditionalGeneration
< source >( config: PegasusConfig )
Parameters
- config (PegasusConfig) — 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 from_pretrained() method to load the model weights.
The PEGASUS Model with a language modeling head. Can be used for summarization. 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 subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None decoder_input_ids: typing.Optional[torch.Tensor] = None decoder_attention_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None decoder_head_mask: typing.Optional[torch.Tensor] = None cross_attn_head_mask: typing.Optional[torch.Tensor] = None encoder_outputs: typing.Optional[typing.Tuple[torch.FloatTensor]] = None past_key_values: typing.Optional[typing.Tuple[torch.FloatTensor]] = None inputs_embeds: typing.Optional[torch.Tensor] = None decoder_inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) β transformers.modeling_outputs.Seq2SeqLMOutput or tuple(torch.FloatTensor)
Parameters
- 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.
- 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.
- decoder_input_ids (
torch.LongTensor
of shape(batch_size, target_sequence_length)
, optional) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
Pegasus uses the
pad_token_id
as the starting token fordecoder_input_ids
generation. Ifpast_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
). - decoder_attention_mask (
torch.LongTensor
of shape(batch_size, target_sequence_length)
, optional) — Default behavior: generate a tensor that ignores pad tokens indecoder_input_ids
. Causal mask will also be used by default. - head_mask (
torch.Tensor
of shape(encoder_layers, encoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- decoder_head_mask (
torch.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the decoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- cross_attn_head_mask (
torch.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the cross-attention modules in the decoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- encoder_outputs (
tuple(tuple(torch.FloatTensor)
, optional) — Tuple consists of (last_hidden_state
, optional:hidden_states
, optional:attentions
)last_hidden_state
of shape(batch_size, sequence_length, hidden_size)
, optional) is a sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of the decoder. - past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.If
past_key_values
are used, the user can optionally input only the lastdecoder_input_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of alldecoder_input_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - decoder_inputs_embeds (
torch.FloatTensor
of shape(batch_size, target_sequence_length, hidden_size)
, optional) — Optionally, instead of passingdecoder_input_ids
you can choose to directly pass an embedded representation. Ifpast_key_values
is used, optionally only the lastdecoder_inputs_embeds
have to be input (seepast_key_values
). This is useful if you want more control over how to convertdecoder_input_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
decoder_input_ids
anddecoder_inputs_embeds
are both unset,decoder_inputs_embeds
takes the value ofinputs_embeds
. - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - labels (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Labels for computing the masked language modeling loss. Indices should either be in[0, ..., config.vocab_size]
or -100 (seeinput_ids
docstring). Tokens with indices set to-100
are ignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size]
.
Returns
transformers.modeling_outputs.Seq2SeqLMOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.Seq2SeqLMOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (PegasusConfig) and inputs.
-
loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenlabels
is provided) β Language modeling loss. -
logits (
torch.FloatTensor
of shape(batch_size, sequence_length, config.vocab_size)
) β Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). -
past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding. -
decoder_hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.
-
decoder_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
encoder_last_hidden_state (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) β Sequence of hidden-states at the output of the last layer of the encoder of the model. -
encoder_hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.
-
encoder_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The PegasusForConditionalGeneration forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Summarization example:
>>> from transformers import AutoTokenizer, PegasusForConditionalGeneration
>>> model = PegasusForConditionalGeneration.from_pretrained("google/pegasus-xsum")
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-xsum")
>>> ARTICLE_TO_SUMMARIZE = (
... "PG&E stated it scheduled the blackouts in response to forecasts for high winds "
... "amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were "
... "scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow."
... )
>>> inputs = tokenizer(ARTICLE_TO_SUMMARIZE, max_length=1024, return_tensors="pt")
>>> # Generate Summary
>>> summary_ids = model.generate(inputs["input_ids"])
>>> tokenizer.batch_decode(summary_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"California's largest electricity provider has turned off power to hundreds of thousands of customers."
PegasusForCausalLM
forward
< source >( input_ids: LongTensor = None attention_mask: typing.Optional[torch.Tensor] = None encoder_hidden_states: typing.Optional[torch.FloatTensor] = None encoder_attention_mask: typing.Optional[torch.FloatTensor] = None head_mask: typing.Optional[torch.Tensor] = None cross_attn_head_mask: typing.Optional[torch.Tensor] = None past_key_values: typing.Optional[typing.List[torch.FloatTensor]] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) β transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
Parameters
- 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.
- 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.
- encoder_hidden_states (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder. - encoder_attention_mask (
torch.FloatTensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in[0, 1]
: - head_mask (
torch.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- cross_attn_head_mask (
torch.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the cross-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
. The two additional tensors are only required when the model is used as a decoder in a Sequence to Sequence model.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.If
past_key_values
are used, the user can optionally input only the lastdecoder_input_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of alldecoder_input_ids
of shape(batch_size, sequence_length)
. - labels (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Labels for computing the masked language modeling loss. Indices should either be in[0, ..., config.vocab_size]
or -100 (seeinput_ids
docstring). Tokens with indices set to-100
are ignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size]
. - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
).- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (PegasusConfig) and inputs.
-
loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenlabels
is provided) β Language modeling loss (for next-token prediction). -
logits (
torch.FloatTensor
of shape(batch_size, sequence_length, config.vocab_size)
) β Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). -
hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
-
attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple oftorch.FloatTensor
tuples of lengthconfig.n_layers
, with each tuple containing the cached key, value states of the self-attention and the cross-attention layers if model is used in encoder-decoder setting. Only relevant ifconfig.is_decoder = True
.Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.
Example:
>>> from transformers import AutoTokenizer, PegasusForCausalLM
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> model = PegasusForCausalLM.from_pretrained("google/pegasus-large", add_cross_attention=False)
>>> assert model.config.is_decoder, f"{model.__class__} has to be configured as a decoder."
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs)
>>> logits = outputs.logits
>>> expected_shape = [1, inputs.input_ids.shape[-1], model.config.vocab_size]
>>> list(logits.shape) == expected_shape
True
TFPegasusModel
class transformers.TFPegasusModel
< source >( config: PegasusConfig *inputs **kwargs )
Parameters
- config (PegasusConfig) — 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 from_pretrained() method to load the model weights.
The bare PEGASUS Model outputting raw hidden-states without any specific head on top. This model inherits from TFPreTrainedModel. 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 keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.
TensorFlow models and layers in transformers
accept two formats as input:
- having all inputs as keyword arguments (like PyTorch models), or
- having all inputs as a list, tuple or dict in the first positional argument.
The reason the second format is supported is that Keras methods prefer this format when passing inputs to models
and layers. Because of this support, when using methods like model.fit()
things should βjust workβ for you - just
pass your inputs and labels in any format that model.fit()
supports! If, however, you want to use the second
format outside of Keras methods like fit()
and predict()
, such as when creating your own layers or models with
the Keras Functional
API, there are three possibilities you can use to gather all the input Tensors in the first
positional argument:
- a single Tensor with
input_ids
only and nothing else:model(input_ids)
- a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
model([input_ids, attention_mask])
ormodel([input_ids, attention_mask, token_type_ids])
- a dictionary with one or several input Tensors associated to the input names given in the docstring:
model({"input_ids": input_ids, "token_type_ids": token_type_ids})
Note that when creating models and layers with subclassing then you donβt need to worry about any of this, as you can just pass inputs like you would to any other Python function!
call
< source >( input_ids: TFModelInputType | None = None attention_mask: np.ndarray | tf.Tensor | None = None decoder_input_ids: np.ndarray | tf.Tensor | None = None decoder_attention_mask: np.ndarray | tf.Tensor | None = None decoder_position_ids: np.ndarray | tf.Tensor | None = None head_mask: np.ndarray | tf.Tensor | None = None decoder_head_mask: np.ndarray | tf.Tensor | None = None cross_attn_head_mask: np.ndarray | tf.Tensor | None = None encoder_outputs: Optional[Union[Tuple, TFBaseModelOutput]] = None past_key_values: Optional[Tuple[Tuple[Union[np.ndarray, tf.Tensor]]]] = None inputs_embeds: np.ndarray | tf.Tensor | None = None decoder_inputs_embeds: np.ndarray | tf.Tensor | None = None use_cache: Optional[bool] = None output_attentions: Optional[bool] = None output_hidden_states: Optional[bool] = None return_dict: Optional[bool] = None training: bool = False **kwargs ) β transformers.modeling_tf_outputs.TFSeq2SeqModelOutput or tuple(tf.Tensor)
Parameters
- input_ids (
tf.Tensor
of shape(batch_size, sequence_length)
) — Indices of input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
tf.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.
- decoder_input_ids (
tf.Tensor
of shape(batch_size, target_sequence_length)
, optional) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
Pegasus uses the
pad_token_id
as the starting token fordecoder_input_ids
generation. Ifpast_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
). - decoder_attention_mask (
tf.Tensor
of shape(batch_size, target_sequence_length)
, optional) — will be made by default and ignore pad tokens. It is not recommended to set this for most use cases. - decoder_position_ids (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each decoder input sequence tokens in the position embeddings. Selected in the range[0, config.max_position_embeddings - 1]
. - head_mask (
tf.Tensor
of shape(encoder_layers, encoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- decoder_head_mask (
tf.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the decoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- cross_attn_head_mask (
tf.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the cross-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- encoder_outputs (
tf.FloatTensor
, optional) — hidden states at the output of the last layer of the encoder. Used in the cross-attention of the decoder. of shape(batch_size, sequence_length, hidden_size)
is a sequence of - past_key_values (
Tuple[Tuple[tf.Tensor]]
of lengthconfig.n_layers
) — contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding. Ifpast_key_values
are used, the user can optionally input only the lastdecoder_input_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of alldecoder_input_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - use_cache (
bool
, optional, defaults toTrue
) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). Set toFalse
during training,True
during generation output_attentions (bool
, optional): Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True. - training (
bool
, optional, defaults toFalse
) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
Returns
transformers.modeling_tf_outputs.TFSeq2SeqModelOutput or tuple(tf.Tensor)
A transformers.modeling_tf_outputs.TFSeq2SeqModelOutput or a tuple of tf.Tensor
(if
return_dict=False
is passed or when config.return_dict=False
) comprising various elements depending on the
configuration (PegasusConfig) and inputs.
-
last_hidden_state (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
) β Sequence of hidden-states at the output of the last layer of the decoder of the model.If
past_key_values
is used only the last hidden-state of the sequences of shape(batch_size, 1, hidden_size)
is output. -
past_key_values (
List[tf.Tensor]
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β List oftf.Tensor
of lengthconfig.n_layers
, with each tensor of shape(2, batch_size, num_heads, sequence_length, embed_size_per_head)
).Contains pre-computed hidden-states (key and values in the attention blocks) of the decoder that can be used (see
past_key_values
input) to speed up sequential decoding. -
decoder_hidden_states (
tuple(tf.Tensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftf.Tensor
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.
-
decoder_attentions (
tuple(tf.Tensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftf.Tensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(tf.Tensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftf.Tensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
encoder_last_hidden_state (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) β Sequence of hidden-states at the output of the last layer of the encoder of the model. -
encoder_hidden_states (
tuple(tf.Tensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftf.Tensor
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.
-
encoder_attentions (
tuple(tf.Tensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftf.Tensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The TFPegasusModel forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoTokenizer, TFPegasusModel
>>> import tensorflow as tf
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> model = TFPegasusModel.from_pretrained("google/pegasus-large")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf")
>>> outputs = model(inputs)
>>> last_hidden_states = outputs.last_hidden_state
TFPegasusForConditionalGeneration
class transformers.TFPegasusForConditionalGeneration
< source >( config *inputs **kwargs )
Parameters
- config (PegasusConfig) — 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 from_pretrained() method to load the model weights.
The PEGASUS Model with a language modeling head. Can be used for summarization. This model inherits from TFPreTrainedModel. 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 keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.
TensorFlow models and layers in transformers
accept two formats as input:
- having all inputs as keyword arguments (like PyTorch models), or
- having all inputs as a list, tuple or dict in the first positional argument.
The reason the second format is supported is that Keras methods prefer this format when passing inputs to models
and layers. Because of this support, when using methods like model.fit()
things should βjust workβ for you - just
pass your inputs and labels in any format that model.fit()
supports! If, however, you want to use the second
format outside of Keras methods like fit()
and predict()
, such as when creating your own layers or models with
the Keras Functional
API, there are three possibilities you can use to gather all the input Tensors in the first
positional argument:
- a single Tensor with
input_ids
only and nothing else:model(input_ids)
- a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
model([input_ids, attention_mask])
ormodel([input_ids, attention_mask, token_type_ids])
- a dictionary with one or several input Tensors associated to the input names given in the docstring:
model({"input_ids": input_ids, "token_type_ids": token_type_ids})
Note that when creating models and layers with subclassing then you donβt need to worry about any of this, as you can just pass inputs like you would to any other Python function!
call
< source >( input_ids: TFModelInputType | None = None attention_mask: np.ndarray | tf.Tensor | None = None decoder_input_ids: np.ndarray | tf.Tensor | None = None decoder_attention_mask: np.ndarray | tf.Tensor | None = None decoder_position_ids: np.ndarray | tf.Tensor | None = None head_mask: np.ndarray | tf.Tensor | None = None decoder_head_mask: np.ndarray | tf.Tensor | None = None cross_attn_head_mask: np.ndarray | tf.Tensor | None = None encoder_outputs: Optional[TFBaseModelOutput] = None past_key_values: Optional[Tuple[Tuple[Union[np.ndarray, tf.Tensor]]]] = None inputs_embeds: np.ndarray | tf.Tensor | None = None decoder_inputs_embeds: np.ndarray | tf.Tensor | None = None use_cache: Optional[bool] = None output_attentions: Optional[bool] = None output_hidden_states: Optional[bool] = None return_dict: Optional[bool] = None labels: np.ndarray | tf.Tensor | None = None training: bool = False ) β transformers.modeling_tf_outputs.TFSeq2SeqLMOutput or tuple(tf.Tensor)
Parameters
- input_ids (
tf.Tensor
of shape({0})
) — Indices of input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
tf.Tensor
of shape({0})
, 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.
- decoder_input_ids (
tf.Tensor
of shape(batch_size, target_sequence_length)
, optional) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
Pegasus uses the
pad_token_id
as the starting token fordecoder_input_ids
generation. Ifpast_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
). - decoder_attention_mask (
tf.Tensor
of shape(batch_size, target_sequence_length)
, optional) — will be made by default and ignore pad tokens. It is not recommended to set this for most use cases. - decoder_position_ids (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each decoder input sequence tokens in the position embeddings. Selected in the range[0, config.max_position_embeddings - 1]
. - head_mask (
tf.Tensor
of shape(encoder_layers, encoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- decoder_head_mask (
tf.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the attention modules in the decoder. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- cross_attn_head_mask (
tf.Tensor
of shape(decoder_layers, decoder_attention_heads)
, optional) — Mask to nullify selected heads of the cross-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- encoder_outputs (
tf.FloatTensor
, optional) — hidden states at the output of the last layer of the encoder. Used in the cross-attention of the decoder. of shape(batch_size, sequence_length, hidden_size)
is a sequence of - past_key_values (
Tuple[Tuple[tf.Tensor]]
of lengthconfig.n_layers
) — contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding. Ifpast_key_values
are used, the user can optionally input only the lastdecoder_input_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of alldecoder_input_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - use_cache (
bool
, optional, defaults toTrue
) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). Set toFalse
during training,True
during generation output_attentions (bool
, optional): Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True. - training (
bool
, optional, defaults toFalse
) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation). - labels (
tf.tensor
of shape(batch_size, sequence_length)
, optional) — Labels for computing the masked language modeling loss. Indices should either be in[0, ..., config.vocab_size]
or -100 (seeinput_ids
docstring). Tokens with indices set to-100
are ignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size]
.
Returns
transformers.modeling_tf_outputs.TFSeq2SeqLMOutput or tuple(tf.Tensor)
A transformers.modeling_tf_outputs.TFSeq2SeqLMOutput or a tuple of tf.Tensor
(if
return_dict=False
is passed or when config.return_dict=False
) comprising various elements depending on the
configuration (PegasusConfig) and inputs.
-
loss (
tf.Tensor
of shape(n,)
, optional, where n is the number of non-masked labels, returned whenlabels
is provided) β Language modeling loss. -
logits (
tf.Tensor
of shape(batch_size, sequence_length, config.vocab_size)
) β Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). -
past_key_values (
List[tf.Tensor]
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β List oftf.Tensor
of lengthconfig.n_layers
, with each tensor of shape(2, batch_size, num_heads, sequence_length, embed_size_per_head)
).Contains pre-computed hidden-states (key and values in the attention blocks) of the decoder that can be used (see
past_key_values
input) to speed up sequential decoding. -
decoder_hidden_states (
tuple(tf.Tensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftf.Tensor
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.
-
decoder_attentions (
tuple(tf.Tensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftf.Tensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(tf.Tensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftf.Tensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
encoder_last_hidden_state (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) β Sequence of hidden-states at the output of the last layer of the encoder of the model. -
encoder_hidden_states (
tuple(tf.Tensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple oftf.Tensor
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.
-
encoder_attentions (
tuple(tf.Tensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple oftf.Tensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The TFPegasusForConditionalGeneration forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Summarization example:
>>> from transformers import AutoTokenizer, TFPegasusForConditionalGeneration
>>> model = TFPegasusForConditionalGeneration.from_pretrained("google/pegasus-xsum")
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-xsum")
>>> ARTICLE_TO_SUMMARIZE = (
... "PG&E stated it scheduled the blackouts in response to forecasts for high winds "
... "amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were "
... "scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow."
... )
>>> inputs = tokenizer(ARTICLE_TO_SUMMARIZE, max_length=1024, return_tensors="tf")
>>> # Generate Summary
>>> summary_ids = model.generate(input_ids)
>>> print(tokenizer.batch_decode(summary_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False))
FlaxPegasusModel
class transformers.FlaxPegasusModel
< source >( config: PegasusConfig input_shape: typing.Tuple[int] = (1, 1) seed: int = 0 dtype: dtype = <class 'jax.numpy.float32'> _do_init: bool = True **kwargs )
Parameters
- config (PegasusConfig) — 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 from_pretrained() method to load the model weights.
- dtype (
jax.numpy.dtype
, optional, defaults tojax.numpy.float32
) — The data type of the computation. Can be one ofjax.numpy.float32
,jax.numpy.float16
(on GPUs) andjax.numpy.bfloat16
(on TPUs).This can be used to enable mixed-precision training or half-precision inference on GPUs or TPUs. If specified all the computation will be performed with the given
dtype
.Note that this only specifies the dtype of the computation and does not influence the dtype of model parameters.
If you wish to change the dtype of the model parameters, see to_fp16() and to_bf16().
The bare Pegasus Model transformer outputting raw hidden-states without any specific head on top. This model inherits from FlaxPreTrainedModel. 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 Flax Linen flax.nn.Module subclass. Use it as a regular Flax Module and refer to the Flax documentation for all matter related to general usage and behavior.
Finally, this model supports inherent JAX features such as:
__call__
< source >( input_ids: Array attention_mask: typing.Optional[jax.Array] = None decoder_input_ids: typing.Optional[jax.Array] = None decoder_attention_mask: typing.Optional[jax.Array] = None position_ids: typing.Optional[jax.Array] = None decoder_position_ids: typing.Optional[jax.Array] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None train: bool = False params: dict = None dropout_rng: <function PRNGKey at 0x7f3f3ee6fd90> = None ) β transformers.modeling_flax_outputs.FlaxSeq2SeqModelOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
jnp.ndarray
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.
- attention_mask (
jnp.ndarray
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.
- decoder_input_ids (
jnp.ndarray
of shape(batch_size, target_sequence_length)
, optional) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- decoder_attention_mask (
jnp.ndarray
of shape(batch_size, target_sequence_length)
, optional) — Default behavior: generate a tensor that ignores pad tokens indecoder_input_ids
. Causal mask will also be used by default.If you want to change padding behavior, you should modify to your needs. See diagram 1 in the paper for more information on the default strategy.
- position_ids (
numpy.ndarray
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.max_position_embeddings - 1]
. - decoder_position_ids (
numpy.ndarray
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each decoder input sequence tokens in the position embeddings. Selected in the range[0, config.max_position_embeddings - 1]
. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxSeq2SeqModelOutput or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxSeq2SeqModelOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (PegasusConfig) and inputs.
-
last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
) β Sequence of hidden-states at the output of the last layer of the decoder of the model.If
past_key_values
is used only the last hidden-state of the sequences of shape(batch_size, 1, hidden_size)
is output. -
past_key_values (
tuple(tuple(jnp.ndarray))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple oftuple(jnp.ndarray)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding. -
decoder_hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.
-
decoder_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
encoder_last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
, optional) β Sequence of hidden-states at the output of the last layer of the encoder of the model. -
encoder_hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.
-
encoder_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The FlaxPegasusPreTrainedModel
forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoTokenizer, FlaxPegasusModel
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> model = FlaxPegasusModel.from_pretrained("google/pegasus-large")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="jax")
>>> outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_state
encode
< source >( input_ids: Array attention_mask: typing.Optional[jax.Array] = None position_ids: typing.Optional[jax.Array] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None train: bool = False params: dict = None dropout_rng: <function PRNGKey at 0x7f3f3ee6fd90> = None ) β transformers.modeling_flax_outputs.FlaxBaseModelOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
jnp.ndarray
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.
- attention_mask (
jnp.ndarray
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.
- position_ids (
numpy.ndarray
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.max_position_embeddings - 1]
. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxBaseModelOutput or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxBaseModelOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (<class 'transformers.models.pegasus.configuration_pegasus.PegasusConfig'>
) and inputs.
-
last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
) β Sequence of hidden-states at the output of the last layer of the model. -
hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
Example:
>>> from transformers import AutoTokenizer, FlaxPegasusForConditionalGeneration
>>> model = FlaxPegasusForConditionalGeneration.from_pretrained("google/pegasus-large")
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> text = "My friends are cool but they eat too many carbs."
>>> inputs = tokenizer(text, max_length=1024, return_tensors="np")
>>> encoder_outputs = model.encode(**inputs)
decode
< source >( decoder_input_ids encoder_outputs encoder_attention_mask: typing.Optional[jax.Array] = None decoder_attention_mask: typing.Optional[jax.Array] = None decoder_position_ids: typing.Optional[jax.Array] = None past_key_values: dict = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None train: bool = False params: dict = None dropout_rng: <function PRNGKey at 0x7f3f3ee6fd90> = None ) β transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPastAndCrossAttentions or tuple(torch.FloatTensor)
Parameters
- decoder_input_ids (
jnp.ndarray
of shape(batch_size, target_sequence_length)
) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- encoder_outputs (
tuple(tuple(jnp.ndarray)
) — Tuple consists of (last_hidden_state
, optional:hidden_states
, optional:attentions
)last_hidden_state
of shape(batch_size, sequence_length, hidden_size)
, optional) is a sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of the decoder. - encoder_attention_mask (
jnp.ndarray
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.
- decoder_attention_mask (
jnp.ndarray
of shape(batch_size, target_sequence_length)
, optional) — Default behavior: generate a tensor that ignores pad tokens indecoder_input_ids
. Causal mask will also be used by default.If you want to change padding behavior, you should modify to your needs. See diagram 1 in the paper for more information on the default strategy.
- decoder_position_ids (
numpy.ndarray
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each decoder input sequence tokens in the position embeddings. Selected in the range[0, config.max_position_embeddings - 1]
. - past_key_values (
Dict[str, np.ndarray]
, optional, returned byinit_cache
or when passing previouspast_key_values
) — Dictionary of pre-computed hidden-states (key and values in the attention blocks) that can be used for fast auto-regressive decoding. Pre-computed key and value hidden-states are of shape [batch_size, max_length]. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPastAndCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPastAndCrossAttentions or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (<class 'transformers.models.pegasus.configuration_pegasus.PegasusConfig'>
) and inputs.
-
last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
) β Sequence of hidden-states at the output of the last layer of the model.If
past_key_values
is used only the last hidden-state of the sequences of shape(batch_size, 1, hidden_size)
is output. -
past_key_values (
tuple(tuple(jnp.ndarray))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple oftuple(jnp.ndarray)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and optionally ifconfig.is_encoder_decoder=True
2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if
config.is_encoder_decoder=True
in the cross-attention blocks) that can be used (seepast_key_values
input) to speed up sequential decoding. -
hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
andconfig.add_cross_attention=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
Example:
>>> import jax.numpy as jnp
>>> from transformers import AutoTokenizer, FlaxPegasusForConditionalGeneration
>>> model = FlaxPegasusForConditionalGeneration.from_pretrained("google/pegasus-large")
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> text = "My friends are cool but they eat too many carbs."
>>> inputs = tokenizer(text, max_length=1024, return_tensors="np")
>>> encoder_outputs = model.encode(**inputs)
>>> decoder_start_token_id = model.config.decoder_start_token_id
>>> decoder_input_ids = jnp.ones((inputs.input_ids.shape[0], 1), dtype="i4") * decoder_start_token_id
>>> outputs = model.decode(decoder_input_ids, encoder_outputs)
>>> last_decoder_hidden_states = outputs.last_hidden_state
FlaxPegasusForConditionalGeneration
class transformers.FlaxPegasusForConditionalGeneration
< source >( config: PegasusConfig input_shape: typing.Tuple[int] = (1, 1) seed: int = 0 dtype: dtype = <class 'jax.numpy.float32'> _do_init: bool = True **kwargs )
Parameters
- config (PegasusConfig) — 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 from_pretrained() method to load the model weights.
- dtype (
jax.numpy.dtype
, optional, defaults tojax.numpy.float32
) — The data type of the computation. Can be one ofjax.numpy.float32
,jax.numpy.float16
(on GPUs) andjax.numpy.bfloat16
(on TPUs).This can be used to enable mixed-precision training or half-precision inference on GPUs or TPUs. If specified all the computation will be performed with the given
dtype
.Note that this only specifies the dtype of the computation and does not influence the dtype of model parameters.
If you wish to change the dtype of the model parameters, see to_fp16() and to_bf16().
The PEGASUS Model with a language modeling head. Can be used for summarization. This model inherits from FlaxPreTrainedModel. 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 Flax Linen flax.nn.Module subclass. Use it as a regular Flax Module and refer to the Flax documentation for all matter related to general usage and behavior.
Finally, this model supports inherent JAX features such as:
__call__
< source >( input_ids: Array attention_mask: typing.Optional[jax.Array] = None decoder_input_ids: typing.Optional[jax.Array] = None decoder_attention_mask: typing.Optional[jax.Array] = None position_ids: typing.Optional[jax.Array] = None decoder_position_ids: typing.Optional[jax.Array] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None train: bool = False params: dict = None dropout_rng: <function PRNGKey at 0x7f3f3ee6fd90> = None ) β transformers.modeling_flax_outputs.FlaxSeq2SeqLMOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
jnp.ndarray
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.
- attention_mask (
jnp.ndarray
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.
- decoder_input_ids (
jnp.ndarray
of shape(batch_size, target_sequence_length)
, optional) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- decoder_attention_mask (
jnp.ndarray
of shape(batch_size, target_sequence_length)
, optional) — Default behavior: generate a tensor that ignores pad tokens indecoder_input_ids
. Causal mask will also be used by default.If you want to change padding behavior, you should modify to your needs. See diagram 1 in the paper for more information on the default strategy.
- position_ids (
numpy.ndarray
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.max_position_embeddings - 1]
. - decoder_position_ids (
numpy.ndarray
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each decoder input sequence tokens in the position embeddings. Selected in the range[0, config.max_position_embeddings - 1]
. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxSeq2SeqLMOutput or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxSeq2SeqLMOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (PegasusConfig) and inputs.
-
logits (
jnp.ndarray
of shape(batch_size, sequence_length, config.vocab_size)
) β Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). -
past_key_values (
tuple(tuple(jnp.ndarray))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple oftuple(jnp.ndarray)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and 2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding. -
decoder_hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.
-
decoder_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoderβs cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
encoder_last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
, optional) β Sequence of hidden-states at the output of the last layer of the encoder of the model. -
encoder_hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.
-
encoder_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The FlaxPegasusPreTrainedModel
forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Summarization example:
>>> from transformers import AutoTokenizer, FlaxPegasusForConditionalGeneration
>>> model = FlaxPegasusForConditionalGeneration.from_pretrained('google/pegasus-large')
>>> tokenizer = AutoTokenizer.from_pretrained('google/pegasus-large')
>>> ARTICLE_TO_SUMMARIZE = "My friends are cool but they eat too many carbs."
>>> inputs = tokenizer([ARTICLE_TO_SUMMARIZE], max_length=1024, return_tensors='np')
>>> # Generate Summary
>>> summary_ids = model.generate(inputs['input_ids']).sequences
>>> print(tokenizer.batch_decode(summary_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False))
Mask filling example:
>>> from transformers import AutoTokenizer, FlaxPegasusForConditionalGeneration
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> TXT = "My friends are <mask> but they eat too many carbs."
>>> model = FlaxPegasusForConditionalGeneration.from_pretrained("google/pegasus-large")
>>> input_ids = tokenizer([TXT], return_tensors="np")["input_ids"]
>>> logits = model(input_ids).logits
>>> masked_index = (input_ids[0] == tokenizer.mask_token_id).nonzero().item()
>>> probs = jax.nn.softmax(logits[0, masked_index], axis=0)
>>> values, predictions = jax.lax.top_k(probs)
>>> tokenizer.decode(predictions).split()
encode
< source >( input_ids: Array attention_mask: typing.Optional[jax.Array] = None position_ids: typing.Optional[jax.Array] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None train: bool = False params: dict = None dropout_rng: <function PRNGKey at 0x7f3f3ee6fd90> = None ) β transformers.modeling_flax_outputs.FlaxBaseModelOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
jnp.ndarray
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.
- attention_mask (
jnp.ndarray
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.
- position_ids (
numpy.ndarray
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.max_position_embeddings - 1]
. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxBaseModelOutput or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxBaseModelOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (<class 'transformers.models.pegasus.configuration_pegasus.PegasusConfig'>
) and inputs.
-
last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
) β Sequence of hidden-states at the output of the last layer of the model. -
hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
Example:
>>> from transformers import AutoTokenizer, FlaxPegasusForConditionalGeneration
>>> model = FlaxPegasusForConditionalGeneration.from_pretrained("google/pegasus-large")
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> text = "My friends are cool but they eat too many carbs."
>>> inputs = tokenizer(text, max_length=1024, return_tensors="np")
>>> encoder_outputs = model.encode(**inputs)
decode
< source >( decoder_input_ids encoder_outputs encoder_attention_mask: typing.Optional[jax.Array] = None decoder_attention_mask: typing.Optional[jax.Array] = None decoder_position_ids: typing.Optional[jax.Array] = None past_key_values: dict = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None deterministic: bool = True params: dict = None dropout_rng: <function PRNGKey at 0x7f3f3ee6fd90> = None ) β transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
Parameters
- decoder_input_ids (
jnp.ndarray
of shape(batch_size, target_sequence_length)
) — Indices of decoder input sequence tokens in the vocabulary.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- encoder_outputs (
tuple(tuple(jnp.ndarray)
) — Tuple consists of (last_hidden_state
, optional:hidden_states
, optional:attentions
)last_hidden_state
of shape(batch_size, sequence_length, hidden_size)
, optional) is a sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention of the decoder. - encoder_attention_mask (
jnp.ndarray
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.
- decoder_attention_mask (
jnp.ndarray
of shape(batch_size, target_sequence_length)
, optional) — Default behavior: generate a tensor that ignores pad tokens indecoder_input_ids
. Causal mask will also be used by default.If you want to change padding behavior, you should modify to your needs. See diagram 1 in the paper for more information on the default strategy.
- decoder_position_ids (
numpy.ndarray
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each decoder input sequence tokens in the position embeddings. Selected in the range[0, config.max_position_embeddings - 1]
. - past_key_values (
Dict[str, np.ndarray]
, optional, returned byinit_cache
or when passing previouspast_key_values
) — Dictionary of pre-computed hidden-states (key and values in the attention blocks) that can be used for fast auto-regressive decoding. Pre-computed key and value hidden-states are of shape [batch_size, max_length]. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (<class 'transformers.models.pegasus.configuration_pegasus.PegasusConfig'>
) and inputs.
-
logits (
jnp.ndarray
of shape(batch_size, sequence_length, config.vocab_size)
) β Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). -
hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) β Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) β Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
past_key_values (
tuple(tuple(jnp.ndarray))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) β Tuple ofjnp.ndarray
tuples of lengthconfig.n_layers
, with each tuple containing the cached key, value states of the self-attention and the cross-attention layers if model is used in encoder-decoder setting. Only relevant ifconfig.is_decoder = True
.Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.
Example:
>>> import jax.numpy as jnp
>>> from transformers import AutoTokenizer, FlaxPegasusForConditionalGeneration
>>> model = FlaxPegasusForConditionalGeneration.from_pretrained("google/pegasus-large")
>>> tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large")
>>> text = "My friends are cool but they eat too many carbs."
>>> inputs = tokenizer(text, max_length=1024, return_tensors="np")
>>> encoder_outputs = model.encode(**inputs)
>>> decoder_start_token_id = model.config.decoder_start_token_id
>>> decoder_input_ids = jnp.ones((inputs.input_ids.shape[0], 1), dtype="i4") * decoder_start_token_id
>>> outputs = model.decode(decoder_input_ids, encoder_outputs)
>>> logits = outputs.logits