Utilities for Generation

This page lists all the utility functions used by , , , , , , , and .

Most of those are only useful if you are studying the code of the generate methods in the library.

Generate Outputs

The output of is an instance of a subclass of . This output is a data structure containing all the information returned by , but that can also be used as tuple or dictionary.

Here’s an example:

Copied

from transformers import GPT2Tokenizer, GPT2LMHeadModel

tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
model = GPT2LMHeadModel.from_pretrained("gpt2")

inputs = tokenizer("Hello, my dog is cute and ", return_tensors="pt")
generation_output = model.generate(**inputs, return_dict_in_generate=True, output_scores=True)

The generation_output object is a , as we can see in the documentation of that class below, it means it has the following attributes:

sequences: the generated sequences of tokens
scores (optional): the prediction scores of the language modelling head, for each generation step
hidden_states (optional): the hidden states of the model, for each generation step
attentions (optional): the attention weights of the model, for each generation step

Here we have the scores since we passed along output_scores=True, but we don’t have hidden_states and attentions because we didn’t pass output_hidden_states=True or output_attentions=True.

You can access each attribute as you would usually do, and if that attribute has not been returned by the model, you will get None. Here for instance generation_output.scores are all the generated prediction scores of the language modeling head, and generation_output.attentions is None.

When using our generation_output object as a tuple, it only keeps the attributes that don’t have None values. Here, for instance, it has two elements, loss then logits, so

Copied

generation_output[:2]

will return the tuple (generation_output.sequences, generation_output.scores) for instance.

When using our generation_output object as a dictionary, it only keeps the attributes that don’t have None values. Here, for instance, it has two keys that are sequences and scores.

We document here all output types.

PyTorch

class transformers.generation.GreedySearchEncoderDecoderOutput

( sequences: LongTensor = Nonescores: typing.Optional[typing.Tuple[torch.FloatTensor]] = Noneencoder_attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = Noneencoder_hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = Nonedecoder_attentions: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = Nonecross_attentions: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = Nonedecoder_hidden_states: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = None )

Parameters

sequences (torch.LongTensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
encoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of encoder-decoder generation models using greedy search. Hidden states and attention weights of the decoder (respectively the encoder) can be accessed via the encoder_attentions and the encoder_hidden_states attributes (respectively the decoder_attentions and the decoder_hidden_states attributes)

class transformers.generation.GreedySearchDecoderOnlyOutput

( sequences: LongTensor = Nonescores: typing.Optional[typing.Tuple[torch.FloatTensor]] = Noneattentions: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = Nonehidden_states: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = None )

Parameters

sequences (torch.LongTensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using greedy search.

class transformers.generation.SampleEncoderDecoderOutput

Parameters

sequences (torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_return_sequences, config.vocab_size).
encoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer of the decoder) of shape (batch_size*num_return_sequences, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size*num_return_sequences, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_return_sequences, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_return_sequences, generated_length, hidden_size).

Base class for outputs of encoder-decoder generation models using sampling. Hidden states and attention weights of the decoder (respectively the encoder) can be accessed via the encoder_attentions and the encoder_hidden_states attributes (respectively the decoder_attentions and the decoder_hidden_states attributes)

class transformers.generation.SampleDecoderOnlyOutput

Parameters

sequences (torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_return_sequences, config.vocab_size).
attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (num_return_sequences*batch_size, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (num_return_sequences*batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using sampling.

class transformers.generation.BeamSearchEncoderDecoderOutput

( sequences: LongTensor = Nonesequences_scores: typing.Optional[torch.FloatTensor] = Nonescores: typing.Optional[typing.Tuple[torch.FloatTensor]] = Nonebeam_indices: typing.Optional[torch.LongTensor] = Noneencoder_attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = Noneencoder_hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = Nonedecoder_attentions: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = Nonecross_attentions: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = Nonedecoder_hidden_states: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = None )

Parameters

sequences (torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (torch.FloatTensor of shape (batch_size*num_return_sequences), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam transition scores for each vocabulary token at each generation step. Beam transition scores consisting of log probabilities of tokens conditioned on log softmax of previously generated tokens in this beam. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams, config.vocab_size).
beam_indices (torch.LongTensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length).
encoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size*num_beams*num_return_sequences, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams*num_return_sequences, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams*num_return_sequences, generated_length, hidden_size).

Base class for outputs of encoder-decoder generation models using beam search. Hidden states and attention weights of the decoder (respectively the encoder) can be accessed via the encoder_attentions and the encoder_hidden_states attributes (respectively the decoder_attentions and the decoder_hidden_states attributes)

class transformers.generation.BeamSearchDecoderOnlyOutput

Parameters

sequences (torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (torch.FloatTensor of shape (batch_size*num_return_sequences), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam transition scores for each vocabulary token at each generation step. Beam transition scores consisting of log probabilities of tokens conditioned on log softmax of previously generated tokens in this beam. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams*num_return_sequences, config.vocab_size).
beam_indices (torch.LongTensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length).
attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams*num_return_sequences, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using beam search.

class transformers.generation.BeamSampleEncoderDecoderOutput

Parameters

sequences (torch.LongTensor of shape (batch_size*num_beams, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (torch.FloatTensor of shape (batch_size * num_return_sequence), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam transition scores for each vocabulary token at each generation step. Beam transition scores consisting of log probabilities of tokens conditioned on log softmax of previously generated tokens in this beam. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams, config.vocab_size)).
beam_indices (torch.LongTensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length).
encoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size*num_beams, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams, generated_length, hidden_size).

Base class for outputs of encoder-decoder generation models using beam sampling. Hidden states and attention weights of the decoder (respectively the encoder) can be accessed via the encoder_attentions and the encoder_hidden_states attributes (respectively the decoder_attentions and the decoder_hidden_states attributes)

class transformers.generation.BeamSampleDecoderOnlyOutput

Parameters

sequences (torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (torch.FloatTensor of shape (batch_size * num_return_sequence), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam transition scores for each vocabulary token at each generation step. Beam transition scores consisting of log probabilities of tokens conditioned on log softmax of previously generated tokens in this beam. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams*num_return_sequences, config.vocab_size).
beam_indices (torch.LongTensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. torch.LongTensor of shape (batch_size*num_return_sequences, sequence_length).
attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size*num_beams, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using beam sample.

class transformers.generation.ContrastiveSearchEncoderDecoderOutput

Parameters

sequences (torch.LongTensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
encoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using contrastive search.

class transformers.generation.ContrastiveSearchDecoderOnlyOutput

Parameters

sequences (torch.LongTensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(torch.FloatTensor) optional, returned when output_scores=True is passed or when — config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of torch.FloatTensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
attentions (tuple(tuple(torch.FloatTensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(torch.FloatTensor)), optional, returned when output_hidden_states=True is —
passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of torch.FloatTensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using contrastive search.

TensorFlow

class transformers.generation.TFGreedySearchEncoderDecoderOutput

( sequences: Tensor = Nonescores: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Noneencoder_attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Noneencoder_hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Nonedecoder_attentions: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = Nonecross_attentions: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = Nonedecoder_hidden_states: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = None )

Parameters

sequences (tf.Tensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
encoder_attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of tf.Tensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, generated_length, hidden_size).

class transformers.generation.TFGreedySearchDecoderOnlyOutput

( sequences: Tensor = Nonescores: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Noneattentions: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = Nonehidden_states: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = None )

Parameters

sequences (tf.Tensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using greedy search.

class transformers.generation.TFSampleEncoderDecoderOutput

Parameters

sequences (tf.Tensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_return_sequences, config.vocab_size).
encoder_attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of tf.Tensor (one for each layer of the decoder) of shape (batch_size*num_return_sequences, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size*num_return_sequences, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_return_sequences, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_return_sequences, generated_length, hidden_size).

class transformers.generation.TFSampleDecoderOnlyOutput

Parameters

sequences (tf.Tensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_return_sequences, config.vocab_size).
attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (num_return_sequences*batch_size, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (num_return_sequences*batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using sampling.

class transformers.generation.TFBeamSearchEncoderDecoderOutput

( sequences: Tensor = Nonesequences_scores: typing.Optional[tensorflow.python.framework.ops.Tensor] = Nonescores: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Nonebeam_indices: typing.Optional[tensorflow.python.framework.ops.Tensor] = Noneencoder_attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Noneencoder_hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = Nonedecoder_attentions: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = Nonecross_attentions: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = Nonedecoder_hidden_states: typing.Optional[typing.Tuple[typing.Tuple[tensorflow.python.framework.ops.Tensor]]] = None )

Parameters

sequences (tf.Tensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (tf.Tensor of shape (batch_size*num_return_sequences), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed beam scores for each vocabulary token at each generation step. Beam scores consisting of log softmax scores for each vocabulary token and sum of log softmax of previously generated tokens in this beam. Tuple of tf.Tensorwith up tomax_new_tokenselements (one element for each generated token), with each tensor of shape(batch_size*num_beams, config.vocab_size)`.
beam_indices (tf.Tensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. tf.Tensor of shape (batch_size*num_return_sequences, sequence_length).
encoder_attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of tf.Tensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size*num_beams*num_return_sequences, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams*num_return_sequences, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams*num_return_sequences, generated_length, hidden_size).

class transformers.generation.TFBeamSearchDecoderOnlyOutput

Parameters

sequences (tf.Tensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (tf.Tensor of shape (batch_size*num_return_sequences), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed beam scores for each vocabulary token at each generation step. Beam scores consisting of log softmax scores for each vocabulary token and sum of log softmax of previously generated tokens in this beam. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams*num_return_sequences, config.vocab_size).
beam_indices (tf.Tensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. tf.Tensor of shape (batch_size*num_return_sequences, sequence_length).
attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams*num_return_sequences, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using beam search.

class transformers.generation.TFBeamSampleEncoderDecoderOutput

Parameters

sequences (tf.Tensor of shape (batch_size*num_beams, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (tf.Tensor of shape (batch_size * num_return_sequence), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed beam scores for each vocabulary token at each generation step. Beam scores consisting of log softmax scores for each vocabulary token and sum of log softmax of previously generated tokens in this beam. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams, config.vocab_size).
beam_indices (tf.Tensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. tf.Tensor of shape (batch_size*num_return_sequences, sequence_length).
encoder_attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of tf.Tensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size*num_beams, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams, generated_length, hidden_size).

class transformers.generation.TFBeamSampleDecoderOnlyOutput

Parameters

sequences (tf.Tensor of shape (batch_size*num_return_sequences, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
sequences_scores (tf.Tensor of shape (batch_size * num_return_sequence), optional, returned when output_scores=True is passed or when config.output_scores=True) — Final beam scores of the generated sequences.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed beam scores for each vocabulary token at each generation step. Beam scores consisting of log softmax scores for each vocabulary token and sum of log softmax of previously generated tokens in this beam. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size*num_beams*num_return_sequences, config.vocab_size).
beam_indices (tf.Tensor, optional, returned when output_scores=True is passed or when config.output_scores=True) — Beam indices of generated token id at each generation step. tf.Tensor of shape (batch_size*num_return_sequences, sequence_length).
attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size*num_beams, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using beam sample.

class transformers.generation.TFContrastiveSearchEncoderDecoderOutput

Parameters

sequences (tf.Tensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
encoder_attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple of tf.Tensor (one for each layer of the decoder) of shape (batch_size, num_heads, sequence_length, sequence_length).
encoder_hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
decoder_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
cross_attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
decoder_hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of encoder-decoder generation models using contrastive search. Hidden states and attention weights of the decoder (respectively the encoder) can be accessed via the encoder_attentions and the encoder_hidden_states attributes (respectively the decoder_attentions and the decoder_hidden_states attributes)

class transformers.generation.TFContrastiveSearchDecoderOnlyOutput

Parameters

sequences (tf.Tensor of shape (batch_size, sequence_length)) — The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.
scores (tuple(tf.Tensor) optional, returned when output_scores=True is passed or when config.output_scores=True) — Processed prediction scores of the language modeling head (scores for each vocabulary token before SoftMax) at each generation step. Tuple of tf.Tensor with up to max_new_tokens elements (one element for each generated token), with each tensor of shape (batch_size, config.vocab_size).
attentions (tuple(tuple(tf.Tensor)), optional, returned when output_attentions=True is passed or config.output_attentions=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, num_heads, generated_length, sequence_length).
hidden_states (tuple(tuple(tf.Tensor)), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple (one element for each generated token) of tuples (one element for each layer of the decoder) of tf.Tensor of shape (batch_size, generated_length, hidden_size).

Base class for outputs of decoder-only generation models using contrastive search.

FLAX

class transformers.generation.FlaxSampleOutput

( sequences: Array = None )

Parameters

sequences (jnp.ndarray of shape (batch_size, max_length)) — The generated sequences.

Flax Base class for outputs of decoder-only generation models using sampling.

replace

( **updates )

“Returns a new object replacing the specified fields with new values.

class transformers.generation.FlaxGreedySearchOutput

( sequences: Array = None )

Parameters

sequences (jnp.ndarray of shape (batch_size, max_length)) — The generated sequences.

Flax Base class for outputs of decoder-only generation models using greedy search.

replace

( **updates )

“Returns a new object replacing the specified fields with new values.

class transformers.generation.FlaxBeamSearchOutput

( sequences: Array = Nonescores: Array = None )

Parameters

sequences (jnp.ndarray of shape (batch_size, max_length)) — The generated sequences.
scores (jnp.ndarray of shape (batch_size,)) — The scores (log probabilities) of the generated sequences.

Flax Base class for outputs of decoder-only generation models using greedy search.

replace

( **updates )

“Returns a new object replacing the specified fields with new values.

LogitsProcessor

A can be used to modify the prediction scores of a language model head for generation.

PyTorch

class transformers.AlternatingCodebooksLogitsProcessor

( input_start_len: intsemantic_vocab_size: intcodebook_size: int )

Parameters

input_start_len (int) — The length of the initial input sequence.
semantic_vocab_size (int) — Vocabulary size of the semantic part, i.e number of tokens associated to the semantic vocabulary.
codebook_size (int) — Number of tokens associated to the codebook.

enforcing alternated generation between the two codebooks of Bark’s fine submodel.

__call__

( input_ids: LongTensorscores: FloatTensor )

class transformers.ClassifierFreeGuidanceLogitsProcessor

( guidance_scale )

Parameters

guidance_scale (float) — The guidance scale for classifier free guidance (CFG). CFG is enabled by setting guidance_scale > 1. Higher guidance scale encourages the model to generate samples that are more closely linked to the input prompt, usually at the expense of poorer quality.

Logits processor for classifier free guidance (CFG). The scores are split over the batch dimension, where the first half correspond to the conditional logits (predicted from the input prompt) and the second half correspond to the unconditional logits (predicted from an empty or ‘null’ prompt). The processor computes a weighted average across the conditional and unconditional logits, parameterised by the guidance_scale.

See for more information.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.EncoderNoRepeatNGramLogitsProcessor

( encoder_ngram_size: intencoder_input_ids: LongTensor )

Parameters

encoder_ngram_size (int) — All ngrams of size ngram_size can only occur within the encoder input ids.
encoder_input_ids (int) — The encoder_input_ids that should not be repeated within the decoder ids.

that enforces no repetition of encoder input ids n-grams for the decoder ids. See .

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.EncoderRepetitionPenaltyLogitsProcessor

( penalty: floatencoder_input_ids: LongTensor )

Parameters

hallucination_penalty (float) — The parameter for hallucination penalty. 1.0 means no penalty.
encoder_input_ids (torch.LongTensor) — The encoder_input_ids that should be repeated within the decoder ids.

enforcing an exponential penalty on tokens that are not in the original input.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.EpsilonLogitsWarper

( epsilon: floatfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

epsilon (float) — If set to > 0, only the most tokens with probabilities epsilon or higher are kept for generation.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs epsilon-sampling, i.e. restricting to tokens with prob >= epsilon. Takes the largest min_tokens_to_keep tokens if no tokens satisfy this constraint. See for more information.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed

>>> set_seed(0)
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")

>>> inputs = tokenizer("A sequence: 1, 2", return_tensors="pt")

>>> # With sampling, the output is unexpected -- sometimes too unexpected.
>>> outputs = model.generate(**inputs, do_sample=True)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
A sequence: 1, 2, 0, 2, 2. 2, 2, 2, 2

>>> # With epsilon sampling, the output gets restricted to high-probability tokens. Note that this is similar to
>>> # Top P sampling, which restricts tokens based on their cumulative probability.
>>> # Pro tip: The paper recomends using `epsilon_cutoff` values between 3e-4 and 9e-4
>>> outputs = model.generate(**inputs, do_sample=True, epsilon_cutoff=0.1)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
A sequence: 1, 2, 3, 4, 5, 6, 7, 8, 9

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.EtaLogitsWarper

( epsilon: floatfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

epsilon (float) — A float value in the range (0, 1). Hyperparameter used to calculate the dynamic cutoff value, eta. The suggested values from the paper ranges from 3e-4 to 4e-3 depending on the size of the model.
filter_value (float, optional, defaults to -float("Inf")) — All values that are found to be below the dynamic cutoff value, eta, are set to this float value. This parameter is useful when logits need to be modified for very low probability tokens that should be excluded from generation entirely.
min_tokens_to_keep (int, optional, defaults to 1) — Specifies the minimum number of tokens that must be kept for generation, regardless of their probabilities. For example, if min_tokens_to_keep is set to 1, at least one token will always be kept for generation, even if all tokens have probabilities below the cutoff eta.

that performs eta-sampling, a technique to filter out tokens with probabilities below a dynamic cutoff value, eta, which is calculated based on a combination of the hyperparameter epsilon and the entropy of the token probabilities, i.e. eta := min(epsilon, sqrt(epsilon * e^-entropy(probabilities))). Takes the largest min_tokens_to_keep tokens if no tokens satisfy this constraint. It addresses the issue of poor quality in long samples of text generated by neural language models leading to more coherent and fluent text. See for more information. Note: do_sample must be set to True for this LogitsWarper to work.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed

>>> set_seed(0)
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")

>>> inputs = tokenizer("A sequence: 1, 2", return_tensors="pt")

>>> # With sampling, the output is unexpected -- sometimes too unexpected.
>>> outputs = model.generate(**inputs, do_sample=True)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
A sequence: 1, 2, 0, 2, 2. 2, 2, 2, 2

>>> # With eta sampling, the output gets restricted to high-probability tokens. You can see it as a dynamic form of
>>> # epsilon sampling that adapts its cutoff probability based on the entropy (high entropy = lower cutoff).
>>> # Pro tip: The paper recomends using `eta_cutoff` values between 3e-4 to 4e-3
>>> outputs = model.generate(**inputs, do_sample=True, eta_cutoff=0.1)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
A sequence: 1, 2, 3, 4, 5, 6, 7, 8, 9

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.ExponentialDecayLengthPenalty

( exponential_decay_length_penalty: typing.Tuple[int, float]eos_token_id: typing.Union[int, typing.List[int]]input_ids_seq_length: int )

Parameters

exponential_decay_length_penalty (tuple(int, float)) — This tuple shall consist of: (start_index, decay_factor) where start_index indicates where penalty starts and decay_factor represents the factor of exponential decay
eos_token_id (Union[int, List[int]]) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.
input_ids_seq_length (int) — The length of the input sequence.

that exponentially increases the score of the eos_token_id after start_index has been reached. This allows generating shorter sequences without having a hard cutoff, allowing the eos_token to be predicted in a meaningful position.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed

>>> set_seed(1)
>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")

>>> text = "Just wanted to let you know, I"
>>> inputs = tokenizer(text, return_tensors="pt")

>>> # Generate sequences without exponential penalty. We want short sentences, so we limit max_length=30
>>> # see that the answer tends to end abruptly
>>> outputs = model.generate(**inputs, do_sample=True, temperature=0.9, max_length=30, pad_token_id=50256)
>>> print(tokenizer.batch_decode(outputs)[0])
Just wanted to let you know, I'm not even a lawyer. I'm a man. I have no real knowledge of politics. I'm a

>>> # Generate sequences with exponential penalty, we add the exponential_decay_length_penalty=(start_index, decay_factor)
>>> # We see that instead of cutting at max_tokens, the output comes to an end before (at 25 tokens) and with more meaning
>>> # What happens is that starting from `start_index` the EOS token score will be increased by decay_factor exponentially
>>> outputs = model.generate(
...     **inputs,
...     do_sample=True,
...     temperature=0.9,
...     max_length=30,
...     pad_token_id=50256,
...     exponential_decay_length_penalty=(15, 1.6),
... )
>>> print(tokenizer.batch_decode(outputs)[0])
Just wanted to let you know, I've got a very cool t-shirt educating people on how to use the Internet<|endoftext|>

>>> # Generate sequences with smaller decay_factor, still improving the hard cutoff mid-sentence
>>> outputs = model.generate(
...     **inputs,
...     do_sample=True,
...     temperature=0.9,
...     max_length=30,
...     pad_token_id=50256,
...     exponential_decay_length_penalty=(15, 1.05),
... )
>>> print(tokenizer.batch_decode(outputs)[0])
Just wanted to let you know, I've been working on it for about 6 months and now it's in Alpha.<|endoftext|>

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.ForcedBOSTokenLogitsProcessor

( bos_token_id: int )

Parameters

bos_token_id (int) — The id of the token to force as the first generated token.

that enforces the specified token as the first generated token.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.ForcedEOSTokenLogitsProcessor

( max_length: inteos_token_id: typing.Union[int, typing.List[int]] )

Parameters

max_length (int) — The maximum length of the sequence to be generated.
eos_token_id (Union[int, List[int]]) — The id of the token to force as the last generated token when max_length is reached. Optionally, use a list to set multiple end-of-sequence tokens.

that enforces the specified token as the last generated token when max_length is reached.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.ForceTokensLogitsProcessor

( force_token_map: typing.List[typing.List[int]] )

This processor takes a list of pairs of integers which indicates a mapping from generation indices to token indices that will be forced before sampling. The processor will set their log probs to inf so that they are sampled at their corresponding index.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.HammingDiversityLogitsProcessor

( diversity_penalty: floatnum_beams: intnum_beam_groups: int )

Parameters

diversity_penalty (float) — This value is subtracted from a beam’s score if it generates a token same as any beam from other group at a particular time. Note that diversity_penalty is only effective if group beam search is enabled. The penalty applied to a beam’s score when it generates a token that has already been chosen by another beam within the same group during the same time step. A higher diversity_penalty will enforce greater diversity among the beams, making it less likely for multiple beams to choose the same token. Conversely, a lower penalty will allow beams to more freely choose similar tokens. Adjusting this value can help strike a balance between diversity and natural likelihood.
num_beams (int) — Number of beams used for group beam search. Beam search is a method used that maintains beams (or “multiple hypotheses”) at each step, expanding each one and keeping the top-scoring sequences. A higher num_beams will explore more potential sequences. This can increase chances of finding a high-quality output but also increases computational cost.
num_beam_groups (int) — Number of groups to divide num_beams into in order to ensure diversity among different groups of beams. Each group of beams will operate independently, selecting tokens without considering the choices of other groups. This division promotes diversity by ensuring that beams within different groups explore different paths. For instance, if num_beams is 6 and num_beam_groups is 2, there will be 2 groups each containing 3 beams. The choice of num_beam_groups should be made considering the desired level of output diversity and the total number of beams. See for more details.

that enforces diverse beam search.

Note that this logits processor is only effective for . See for more details.

Diverse beam search can be particularly useful in scenarios where a variety of different outputs is desired, rather than multiple similar sequences. It allows the model to explore different generation paths and provides a broader coverage of possible outputs.

This logits processor can be resource-intensive, especially when using large models or long sequences.

Traditional beam search often generates very similar sequences across different beams. HammingDiversityLogitsProcessor addresses this by penalizing beams that generate tokens already chosen by other beams in the same time step.

How It Works:

Grouping Beams: Beams are divided into groups. Each group selects tokens independently of the others.
Penalizing Repeated Tokens: If a beam in a group selects a token already chosen by another group in the same step, a penalty is applied to that token’s score.
Promoting Diversity: This penalty discourages beams within a group from selecting the same tokens as beams in other groups.

Benefits:

Diverse Outputs: Produces a variety of different sequences.
Exploration: Allows the model to explore different paths.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
>>> import torch

>>> # Initialize the model and tokenizer
>>> tokenizer = AutoTokenizer.from_pretrained("t5-base")
>>> model = AutoModelForSeq2SeqLM.from_pretrained("t5-base")

>>> # A long text about the solar system
>>> text = "The Solar System is a gravitationally bound system comprising the Sun and the objects that orbit it, either directly or indirectly. Of the objects that orbit the Sun directly, the largest are the eight planets, with the remainder being smaller objects, such as the five dwarf planets and small Solar System bodies. The Solar System formed 4.6 billion years ago from the gravitational collapse of a giant interstellar molecular cloud."
>>> inputs = tokenizer("summarize: " + text, return_tensors="pt")

>>> # Generate diverse summary
>>> outputs_diverse = model.generate(
...     **inputs,
...     num_beam_groups=2,
...     diversity_penalty=10.0,
...     max_length=100,
...     num_beams=4,
...     num_return_sequences=2,
... )
>>> summaries_diverse = tokenizer.batch_decode(outputs_diverse, skip_special_tokens=True)

>>> # Generate non-diverse summary
>>> outputs_non_diverse = model.generate(
...     **inputs,
...     max_length=100,
...     num_beams=4,
...     num_return_sequences=2,
... )
>>> summary_non_diverse = tokenizer.batch_decode(outputs_non_diverse, skip_special_tokens=True)

>>> # With `diversity_penalty`, the resulting beams are much more diverse
>>> print(summary_non_diverse)
['the solar system formed 4.6 billion years ago from the collapse of a giant interstellar molecular cloud. of the objects that orbit the Sun directly, the largest are the eight planets.',
'the Solar System formed 4.6 billion years ago from the collapse of a giant interstellar molecular cloud. of the objects that orbit the Sun directly, the largest are the eight planets.']

>>> print(summaries_diverse)
['the solar system formed 4.6 billion years ago from the collapse of a giant interstellar molecular cloud. of the objects that orbit the Sun directly, the largest are the eight planets.',
'the solar system formed 4.6 billion years ago from the collapse of a giant interstellar molecular cloud. of the objects that orbit the Sun directly, the largest are the eight planets. the rest of the objects are smaller objects, such as the five dwarf planets and small solar system bodies.']

__call__

( input_ids: LongTensorscores: FloatTensorcurrent_tokens: LongTensorbeam_group_idx: int ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search
current_tokens (torch.LongTensor of shape (batch_size)) — Indices of input sequence tokens in the vocabulary, corresponding to the tokens selected by the other beam groups in the current generation step.
beam_group_idx (int) — The index of the beam group currently being processed.

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.InfNanRemoveLogitsProcessor

( )

that removes all nan and inf values to avoid the generation method to fail. Note that using the logits processor should only be used if necessary since it can slow down the generation method.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.LogitNormalization

( )

and for normalizing the scores using log-softmax. It’s important to normalize the scores during beam search, after applying the logits processors or warpers, since the search algorithm used in this library doesn’t do it (it only does it before, but they may need re-normalization) but it still supposes that the scores are normalized when comparing the hypotheses.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.LogitsProcessor

( )

Abstract base class for all logit processors that can be applied during generation.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.LogitsProcessorList

( iterable = () )

This class can be used to create a list of or to subsequently process a scores input tensor. This class inherits from list and adds a specific call method to apply each or to the inputs.

__call__

( input_ids: LongTensorscores: FloatTensor**kwargs ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search
kwargs (Dict[str, Any], optional) — Additional kwargs that are specific to a logits processor.

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.LogitsWarper

( )

Abstract base class for all logit warpers that can be applied during generation with multinomial sampling.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.MinLengthLogitsProcessor

( min_length: inteos_token_id: typing.Union[int, typing.List[int]] )

Parameters

min_length (int) — The minimum length below which the score of eos_token_id is set to -float("Inf").
eos_token_id (Union[int, List[int]]) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.

enforcing a min-length by setting EOS probability to 0.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.MinNewTokensLengthLogitsProcessor

( prompt_length_to_skip: intmin_new_tokens: inteos_token_id: typing.Union[int, typing.List[int]] )

Parameters

prompt_length_to_skip (int) — The input tokens length. Not a valid argument when used with generate as it will automatically assign the input length.
min_new_tokens (int) — The minimum new tokens length below which the score of eos_token_id is set to -float("Inf").
eos_token_id (Union[int, List[int]]) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.

enforcing a min-length of new tokens by setting EOS (End-Of-Sequence) token probability to 0. Note that for decoder-only models, such as Llama2, min_length will compute the length of prompt + newly generated tokens whereas for other models it will behave as min_new_tokens, that is, taking only into account the newly generated ones.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM

>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
>>> model.config.pad_token_id = model.config.eos_token_id
>>> inputs = tokenizer(["BOINC AI Company is"], return_tensors="pt")

>>> # If the maximum length (default = 20) is smaller than the minimum length constraint, the latter is ignored!
>>> outputs = model.generate(**inputs, min_new_tokens=30)
>>> print(tokenizer.decode(outputs[0], skip_special_tokens=True))
BOINC AI Company is a company that has been working on a new product for the past year.

>>> # For testing purposes, let's set `eos_token` to `"company"`, the first generated token. This will make
>>> # generation end there.
>>> outputs = model.generate(**inputs, eos_token_id=1664)
>>> print(tokenizer.decode(outputs[0], skip_special_tokens=True))
BOINC AI Company is a company

>>> # Increasing `min_new_tokens` will make generation ignore occurences `"company"` (eos token) before the
>>> # minimum length condition is honored.
>>> outputs = model.generate(**inputs, min_new_tokens=2, eos_token_id=1664)
>>> print(tokenizer.decode(outputs[0], skip_special_tokens=True))
BOINC AI Company is a new company

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.NoBadWordsLogitsProcessor

( bad_words_ids: typing.List[typing.List[int]]eos_token_id: typing.Union[int, typing.List[int]] )

Parameters

bad_words_ids (List[List[int]]) — List of list of token ids that are not allowed to be generated.
eos_token_id (Union[int, List[int]]) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.

that enforces that specified sequences will never be selected.

In order to get the token ids of the words that should not appear in the generated text, make sure to set add_prefix_space=True when initializing the tokenizer, and use tokenizer(bad_words, add_special_tokens=False).input_ids. The add_prefix_space argument is only supported for some slow tokenizers, as fast tokenizers’ prefixing behaviours come from pre tokenizers. Read more .

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM

>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
>>> inputs = tokenizer(["In a word, the cake is a"], return_tensors="pt")

>>> output_ids = model.generate(inputs["input_ids"], max_new_tokens=5, pad_token_id=tokenizer.eos_token_id)
>>> print(tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0])
In a word, the cake is a bit of a mess.

>>> # Now let's take the bad words out. Please note that the tokenizer is initialized differently
>>> tokenizer_with_prefix_space = AutoTokenizer.from_pretrained("gpt2", add_prefix_space=True)


>>> def get_tokens_as_list(word_list):
...     "Converts a sequence of words into a list of tokens"
...     tokens_list = []
...     for word in word_list:
...         tokenized_word = tokenizer_with_prefix_space([word], add_special_tokens=False).input_ids[0]
...         tokens_list.append(tokenized_word)
...     return tokens_list


>>> bad_words_ids = get_tokens_as_list(word_list=["mess"])
>>> output_ids = model.generate(
...     inputs["input_ids"], max_new_tokens=5, bad_words_ids=bad_words_ids, pad_token_id=tokenizer.eos_token_id
... )
>>> print(tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0])
In a word, the cake is a bit of a surprise.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.NoRepeatNGramLogitsProcessor

( ngram_size: int )

Parameters

ngram_size (int) — All ngrams of size ngram_size can only occur once.

N-grams are groups of “n” consecutive words, characters, or tokens taken from a sequence of text. Given the sentence: “She runs fast”, the bi-grams (n=2) would be (“she”, “runs”) and (“runs”, “fast”). In text generation, avoiding repetitions of word sequences provides a more diverse output. This enforces no repetition of n-grams by setting the scores of banned tokens to negative infinity which eliminates those tokens from consideration when further processing the scores. .

Use n-gram penalties with care. For instance, penalizing 2-grams (bigrams) in an article about the city of New York might lead to undesirable outcomes where the city’s name appears only once in the entire text.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM

>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")
>>> inputs = tokenizer(["Today I"], return_tensors="pt")

>>> output = model.generate(**inputs)
>>> print(tokenizer.decode(output[0], skip_special_tokens=True))
Today I’m not sure if I’m going to be able to do it.

>>> # Now let's add ngram size using `no_repeat_ngram_size`. This stops the repetitions ("I’m") in the output.
>>> output = model.generate(**inputs, no_repeat_ngram_size=2)
>>> print(tokenizer.decode(output[0], skip_special_tokens=True))
Today I’m not sure if I can get a better understanding of the nature of this issue

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.PrefixConstrainedLogitsProcessor

( prefix_allowed_tokens_fn: typing.Callable[[int, torch.Tensor], typing.List[int]]num_beams: int )

Parameters

prefix_allowed_tokens_fn (Callable[[int, torch.Tensor], List[int]]) — This function constraints the beam search to allowed tokens only at each step. This function takes 2 arguments inputs_ids and the batch ID batch_id. It has to return a list with the allowed tokens for the next generation step conditioned on the previously generated tokens inputs_ids and the batch ID batch_id.

that enforces constrained generation and is useful for prefix-conditioned constrained generation. See for more information.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.RepetitionPenaltyLogitsProcessor

( penalty: float )

Parameters

repetition_penalty (float) — The parameter for repetition penalty. 1.0 means no penalty. See for more details.

that prevents the repetition of previous tokens through an exponential penalty. This technique shares some similarities with coverage mechanisms and other aimed at reducing repetition. During the text generation process, the probability distribution for the next token is determined using a formula that incorporates token scores based on their occurrence in the generated sequence. Tokens with higher scores are more likely to be selected. The formula can be seen in the original . According to the paper a penalty of around 1.2 yields a good balance between truthful generation and lack of repetition.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM

>>> # Initializing the model and tokenizer for it
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")
>>> inputs = tokenizer(["I'm not going to"], return_tensors="pt")

>>> # This shows a normal generate without any specific parameters
>>> summary_ids = model.generate(**inputs)
>>> print(tokenizer.batch_decode(summary_ids, skip_special_tokens=True)[0])
I'm not going to be able to do that. I'm going to be able to do that

>>> # This generates a penalty for repeated tokens
>>> penalized_ids = model.generate(**inputs, repetition_penalty=1.1)
>>> print(tokenizer.batch_decode(penalized_ids, skip_special_tokens=True)[0])
I'm not going to be able to do that. I'll just have to go out and play

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.SequenceBiasLogitsProcessor

( sequence_bias: typing.Dict[typing.Tuple[int], float] )

Parameters

sequence_bias (Dict[Tuple[int], float]) — Dictionary that maps a sequence of tokens to its bias term. Positive biases increase the odds of the sequence being selected, while negative biases do the opposite. If a sequence has a length of 1, its bias will always be applied. Otherwise, the bias will only be applied if the sequence in question is about to be completed (in the token selection step after this processor is applied).

that applies an additive bias on sequences. The bias is applied to the last token of a sequence when the next generated token can complete it. Consequently, to take the most of biasing sequences with more than one token, consider using beam methods (to gracefully work around partially completed sequences that have a negative bias) and applying the bias to their prefixes (to ensure the bias is applied earlier).

In order to get the token ids of the sequences that you want to bias, make sure to set add_prefix_space=True when initializing the tokenizer, and use tokenizer(bad_words, add_special_tokens=False).input_ids. The add_prefix_space argument is only supported for some slow tokenizers, as fast tokenizers’ prefixing behaviours come from pre tokenizers. Read more .

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM

>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
>>> inputs = tokenizer(["The full name of Donald is Donald"], return_tensors="pt")

>>> summary_ids = model.generate(inputs["input_ids"], max_new_tokens=4)
>>> print(tokenizer.batch_decode(summary_ids, skip_special_tokens=True)[0])
The full name of Donald is Donald J. Trump Jr

>>> # Now let's control generation through a bias. Please note that the tokenizer is initialized differently!
>>> tokenizer_with_prefix_space = AutoTokenizer.from_pretrained("gpt2", add_prefix_space=True)


>>> def get_tokens_as_tuple(word):
...     return tuple(tokenizer_with_prefix_space([word], add_special_tokens=False).input_ids[0])


>>> # If we add a negative bias without beam search, it may become "stuck" in a prefix without good continuations
>>> sequence_bias = {get_tokens_as_tuple("Trump"): -10.0}
>>> biased_ids = model.generate(inputs["input_ids"], max_new_tokens=4, sequence_bias=sequence_bias)
>>> print(tokenizer.batch_decode(biased_ids, skip_special_tokens=True)[0])
The full name of Donald is Donald J. Donald,

>>> biased_ids = model.generate(inputs["input_ids"], max_new_tokens=4, num_beams=4, sequence_bias=sequence_bias)
>>> print(tokenizer.batch_decode(biased_ids, skip_special_tokens=True)[0])
The full name of Donald is Donald Rumsfeld,

>>> # We can also add a positive bias to nudge the model towards specific tokens or continuations
>>> sequence_bias = {get_tokens_as_tuple("Donald Duck"): 10.0}
>>> biased_ids = model.generate(inputs["input_ids"], max_new_tokens=4, num_beams=4, sequence_bias=sequence_bias)
>>> print(tokenizer.batch_decode(biased_ids, skip_special_tokens=True)[0])
The full name of Donald is Donald Duck.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.SuppressTokensAtBeginLogitsProcessor

( begin_suppress_tokensbegin_index )

supresses a list of tokens as soon as the generate function starts generating using begin_index tokens. This should ensure that the tokens defined by begin_suppress_tokens at not sampled at the begining of the generation.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.SuppressTokensLogitsProcessor

( suppress_tokens )

This processor can be used to suppress a list of tokens. The processor will set their log probs to -inf so that they are not sampled.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.TemperatureLogitsWarper

( temperature: float )

Parameters

temperature (float) — Strictly positive float value used to modulate the logits distribution. A value smaller than 1 decreases randomness (and vice versa), with 0 being equivalent to shifting all probability mass to the most likely token.

for temperature (exponential scaling output probability distribution), which effectively means that it can control the randomness of the predicted tokens.

Make sure that do_sample=True is included in the generate arguments otherwise the temperature value won’t have any effect.

Examples:

Copied

>>> import torch
>>> from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed

>>> set_seed(0)  # for reproducibility

>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> model.config.pad_token_id = model.config.eos_token_id
>>> inputs = tokenizer(["BOINC AI Company is"], return_tensors="pt")

>>> # With temperature=1.0, the default, we consistently get random outputs due to random sampling.
>>> generate_kwargs = {"max_new_tokens": 10, "do_sample": True, "temperature": 1.0, "num_return_sequences": 2}
>>> outputs = model.generate(**inputs, **generate_kwargs)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
['BOINC AI Company is a joint venture between GEO Group, one of',
'BOINC AI Company is not an exact science – but what we believe does']

>>> # However, with temperature close to 0, it approximates greedy decoding strategies (invariant)
>>> generate_kwargs["temperature"] = 0.0001
>>> outputs = model.generate(**inputs, **generate_kwargs)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
['BOINC AI Company is a company that has been around for over 20 years',
'BOINC AI Company is a company that has been around for over 20 years']

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.TopKLogitsWarper

( top_k: intfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_k (int) — The number of highest probability vocabulary tokens to keep for top-k-filtering.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs top-k, i.e. restricting to the k highest probability elements.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.TopPLogitsWarper

( top_p: floatfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_p (float) — If set to < 1, only the smallest set of most probable tokens with probabilities that add up to top_p or higher are kept for generation.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs top-p, i.e. restricting to top tokens summing to prob_cut_off <= prob_cut_off.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed

>>> set_seed(0)
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")

>>> inputs = tokenizer("A sequence: 1, 2", return_tensors="pt")

>>> # With sampling, the output is unexpected -- sometimes too unexpected.
>>> outputs = model.generate(**inputs, do_sample=True)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
A sequence: 1, 2, 0, 2, 2. 2, 2, 2, 2

>>> # With `top_p` sampling, the output gets restricted to high-probability tokens.
>>> # Pro tip: In practice, LLMs use `top_p` in the 0.9-0.95 range.
>>> outputs = model.generate(**inputs, do_sample=True, top_p=0.1)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
A sequence: 1, 2, 3, 4, 5, 6, 7, 8, 9

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.TypicalLogitsWarper

( mass: float = 0.9filter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

mass (float) — Value of typical_p between 0 and 1 inclusive, defaults to 0.9.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs typical decoding. See for more information.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.UnbatchedClassifierFreeGuidanceLogitsProcessor

( guidance_scale: floatmodelunconditional_ids: typing.Optional[torch.LongTensor] = Noneunconditional_attention_mask: typing.Optional[torch.LongTensor] = Noneuse_cache: typing.Optional[bool] = True )

Parameters

guidance_scale (float) — The guidance scale for classifier free guidance (CFG). CFG is enabled by setting guidance_scale != 1. Higher guidance scale encourages the model to generate samples that are more closely linked to the input prompt, usually at the expense of poorer quality. A value smaller than 1 has the opposite effect, while making the negative prompt provided with negative_prompt_ids (if any) act as a positive prompt.
unconditional_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary for the unconditional branch. If unset, will default to the last token of the prompt.
unconditional_attention_mask (torch.LongTensor of shape (batch_size, sequence_length), optional) — Attention mask for unconditional_ids.
model (PreTrainedModel) — The model computing the unconditional scores. Supposedly the same as the one computing the conditional scores. Both models must use the same tokenizer.
smooth_factor (float, optional) — The interpolation weight for CFG Rescale. 1 means no rescaling, 0 reduces to the conditional scores without CFG. Turn it lower if the output degenerates.
use_cache (bool, optional) — Whether to cache key/values during the negative prompt forward pass.

Logits processor for Classifier-Free Guidance (CFG). The processors computes a weighted average across scores from prompt conditional and prompt unconditional (or negative) logits, parameterized by the guidance_scale. The unconditional scores are computed internally by prompting model with the unconditional_ids branch.

See for more information.

Examples:

Copied

>>> from transformers import AutoTokenizer, AutoModelForCausalLM

>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
>>> inputs = tokenizer(["Today, a dragon flew over Paris, France,"], return_tensors="pt")
>>> out = model.generate(inputs["input_ids"], guidance_scale=1.5)
>>> tokenizer.batch_decode(out, skip_special_tokens=True)[0]
'Today, a dragon flew over Paris, France, killing at least 50 people and injuring more than 100'

>>> # with a negative prompt
>>> neg_inputs = tokenizer(["A very happy event happened,"], return_tensors="pt")
>>> out = model.generate(inputs["input_ids"], guidance_scale=2, negative_prompt_ids=neg_inputs["input_ids"])
>>> tokenizer.batch_decode(out, skip_special_tokens=True)[0]
'Today, a dragon flew over Paris, France, killing at least 130 people. French media reported that'

>>> # with a positive prompt
>>> neg_inputs = tokenizer(["A very happy event happened,"], return_tensors="pt")
>>> out = model.generate(inputs["input_ids"], guidance_scale=0, negative_prompt_ids=neg_inputs["input_ids"])
>>> tokenizer.batch_decode(out, skip_special_tokens=True)[0]
"Today, a dragon flew over Paris, France, and I'm very happy to be here. I"

__call__

( input_idsscores )

class transformers.WhisperTimeStampLogitsProcessor

( generate_config )

Parameters

generate_config (GenerateConfig) — The generate config used to generate the output. The following parameters are required: eos_token_id (int, optional, defaults to 50257): The id of the end-of-sequence token. no_timestamps_token_id (int, optional, defaults to 50363): The id of the "<|notimestamps|>" token. max_initial_timestamp_index (int, optional, defaults to 1): Used to set the maximum value of the initial timestamp. This is used to prevent the model from predicting timestamps that are too far in the future.

Whisper specific Processor. This processor can be used to force a list of tokens. The processor will set their log probs to inf so that they are sampled at their corresponding index.

See for more information.

__call__

( input_ids: LongTensorscores: FloatTensor ) → torch.FloatTensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search

Returns

torch.FloatTensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

TensorFlow

class transformers.TFForcedBOSTokenLogitsProcessor

( bos_token_id: int )

Parameters

bos_token_id (int) — The id of the token to force as the first generated token.

that enforces the specified token as the first generated token.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFForcedEOSTokenLogitsProcessor

( max_length: inteos_token_id: int )

Parameters

max_length (int) — The maximum length of the sequence to be generated.
eos_token_id (int) — The id of the token to force as the last generated token when max_length is reached.

that enforces the specified token as the last generated token when max_length is reached.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFForceTokensLogitsProcessor

( force_token_map: typing.List[typing.List[int]] )

This processor takes a list of pairs of integers which indicates a mapping from generation indices to token indices that will be forced before sampling. The processor will set their log probs to 0 and all other tokens to -inf so that they are sampled at their corresponding index.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFLogitsProcessor

( )

Abstract base class for all logit processors that can be applied during generation.

__call__

( input_ids: Tensorscores: Tensorcur_len: int ) → tf.Tensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (tf.Tensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (tf.Tensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search.
cur_len (int) — The current length of valid input sequence tokens. In the TF implementation, the input_ids’ sequence length is the maximum length generate can produce, and we need to know which of its tokens are valid.
kwargs (Dict[str, Any], optional) — Additional logits processor specific kwargs.

Returns

tf.Tensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

TF method for processing logits.

class transformers.TFLogitsProcessorList

( iterable = () )

This class can be used to create a list of to subsequently process a scores input tensor. This class inherits from list and adds a specific call method to apply each to the inputs.

__call__

( input_ids: Tensorscores: Tensorcur_len: int**kwargs ) → tf.Tensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (tf.Tensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (tf.Tensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search.
cur_len (int) — The current length of valid input sequence tokens. In the TF implementation, the input_ids’ sequence length is the maximum length generate can produce, and we need to know which of its tokens are valid.
kwargs (Dict[str, Any], optional) — Additional logits processor specific kwargs.

Returns

tf.Tensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.TFLogitsWarper

( )

Abstract base class for all logit warpers that can be applied during generation with multinomial sampling.

__call__

( input_ids: Tensorscores: Tensorcur_len: int ) → tf.Tensor of shape (batch_size, config.vocab_size)

Parameters

input_ids (tf.Tensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (tf.Tensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search.
cur_len (int) — The current length of valid input sequence tokens. In the TF implementation, the input_ids’ sequence length is the maximum length generate can produce, and we need to know which of its tokens are valid.
kwargs (Dict[str, Any], optional) — Additional logits processor specific kwargs.

Returns

tf.Tensor of shape (batch_size, config.vocab_size)

The processed prediction scores.

TF method for warping logits.

class transformers.TFMinLengthLogitsProcessor

( min_length: inteos_token_id: int )

Parameters

min_length (int) — The minimum length below which the score of eos_token_id is set to -float("Inf").
eos_token_id (int) — The id of the end-of-sequence token.

enforcing a min-length by setting EOS probability to 0.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFNoBadWordsLogitsProcessor

( bad_words_ids: typing.List[typing.List[int]]eos_token_id: int )

Parameters

bad_words_ids (List[List[int]]) — List of list of token ids that are not allowed to be generated. In order to get the tokens of the words that should not appear in the generated text, make sure to set add_prefix_space=True when initializing the tokenizer, and use tokenizer(bad_words, add_special_tokens=False).input_ids. The add_prefix_space argument is only supported for some slow tokenizers, as fast tokenizers’ prefixing behaviours come from pre tokenizers. Read more .
eos_token_id (int) — The id of the end-of-sequence token.

that enforces that specified sequences will never be sampled.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFNoRepeatNGramLogitsProcessor

( ngram_size: int )

Parameters

ngram_size (int) — All ngrams of size ngram_size can only occur once.

that enforces no repetition of n-grams. See .

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFRepetitionPenaltyLogitsProcessor

( penalty: float )

Parameters

repetition_penalty (float) — The parameter for repetition penalty. 1.0 means no penalty. See for more details.

enforcing an exponential penalty on repeated sequences.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFSuppressTokensAtBeginLogitsProcessor

( begin_suppress_tokensbegin_index )

suppresses a list of tokens as soon as the generate function starts generating using begin_index tokens. This should ensure that the tokens defined by begin_suppress_tokens at not sampled at the begining of the generation.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFSuppressTokensLogitsProcessor

( suppress_tokens )

This processor can be used to suppress a list of tokens. The processor will set their log probs to -inf so that they are not sampled.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFTemperatureLogitsWarper

( temperature: float )

Parameters

temperature (float) — The value used to module the logits distribution.

for temperature (exponential scaling output probability distribution).

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFTopKLogitsWarper

( top_k: intfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_k (int) — The number of highest probability vocabulary tokens to keep for top-k-filtering.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs top-k, i.e. restricting to the k highest probability elements.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

class transformers.TFTopPLogitsWarper

( top_p: floatfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_p (float) — If set to < 1, only the smallest set of most probable tokens with probabilities that add up to top_p or higher are kept for generation.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs top-p, i.e. restricting to top tokens summing to <= prob_cut_off.

__call__

( input_ids: Tensorscores: Tensorcur_len: int )

FLAX

class transformers.FlaxForcedBOSTokenLogitsProcessor

( bos_token_id: int )

Parameters

bos_token_id (int) — The id of the token to force as the first generated token.

that enforces the specified token as the first generated token.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxForcedEOSTokenLogitsProcessor

( max_length: inteos_token_id: int )

Parameters

max_length (int) — The maximum length of the sequence to be generated.
eos_token_id (int) — The id of the token to force as the last generated token when max_length is reached.

that enforces the specified token as the last generated token when max_length is reached.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxForceTokensLogitsProcessor

( force_token_map )

Parameters

force_token_map (list) — Map giving token ids and indices where they will be forced to be sampled.

that takes a list of pairs of integers which indicates a mapping from generation indices to token indices that will be forced before sampling. The processor will set their log probs to 0 and all other tokens to -inf so that they are sampled at their corresponding index.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxLogitsProcessor

( )

Abstract base class for all logit processors that can be applied during generation.

__call__

( input_ids: Arrayscores: Array ) → jnp.ndarray of shape (batch_size, config.vocab_size)

Parameters

input_ids (jnp.ndarray of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (jnp.ndarray of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search
kwargs (Dict[str, Any], optional) — Additional logits processor specific kwargs.

Returns

jnp.ndarray of shape (batch_size, config.vocab_size)

The processed prediction scores.

Flax method for processing logits.

class transformers.FlaxLogitsProcessorList

( iterable = () )

This class can be used to create a list of or to subsequently process a scores input tensor. This class inherits from list and adds a specific call method to apply each or to the inputs.

__call__

( input_ids: Arrayscores: Arraycur_len: int**kwargs ) → jnp.ndarray of shape (batch_size, config.vocab_size)

Parameters

input_ids (jnp.ndarray of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (jnp.ndarray of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search
kwargs (Dict[str, Any], optional) — Additional logits processor specific kwargs.

Returns

jnp.ndarray of shape (batch_size, config.vocab_size)

The processed prediction scores.

class transformers.FlaxLogitsWarper

( )

Abstract base class for all logit warpers that can be applied during generation with multinomial sampling.

__call__

( input_ids: Arrayscores: Array ) → jnp.ndarray of shape (batch_size, config.vocab_size)

Parameters

input_ids (jnp.ndarray of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (jnp.ndarray of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be logits for each vocabulary when not using beam search or log softmax for each vocabulary token when using beam search
kwargs (Dict[str, Any], optional) — Additional logits processor specific kwargs.

Returns

jnp.ndarray of shape (batch_size, config.vocab_size)

The processed prediction scores.

Flax method for warping logits.

class transformers.FlaxMinLengthLogitsProcessor

( min_length: inteos_token_id: int )

Parameters

min_length (int) — The minimum length below which the score of eos_token_id is set to -float("Inf").
eos_token_id (int) — The id of the end-of-sequence token.

enforcing a min-length by setting EOS probability to 0.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxSuppressTokensAtBeginLogitsProcessor

( begin_suppress_tokensbegin_index )

Parameters

begin_suppress_tokens (List[int]) — Tokens to not sample.
begin_index (int) — Index where the tokens are suppressed.

supressing a list of tokens as soon as the generate function starts generating using begin_index tokens. This should ensure that the tokens defined by begin_suppress_tokens are not sampled at the begining of the generation.

__call__

( input_idsscorescur_len: int )

class transformers.FlaxSuppressTokensLogitsProcessor

( suppress_tokens: list )

Parameters

suppress_tokens (list) — Tokens to not sample.

suppressing a list of tokens at each decoding step. The processor will set their log probs to be -inf so they are not sampled.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxTemperatureLogitsWarper

( temperature: float )

Parameters

temperature (float) — The value used to module the logits distribution.

for temperature (exponential scaling output probability distribution).

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxTopKLogitsWarper

( top_k: intfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_k (int) — The number of highest probability vocabulary tokens to keep for top-k-filtering.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs top-k, i.e. restricting to the k highest probability elements.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxTopPLogitsWarper

( top_p: floatfilter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_p (float) — If set to < 1, only the smallest set of most probable tokens with probabilities that add up to top_p or higher are kept for generation.
filter_value (float, optional, defaults to -float("Inf")) — All filtered values will be set to this float value.
min_tokens_to_keep (int, optional, defaults to 1) — Minimum number of tokens that cannot be filtered.

that performs top-p, i.e. restricting to top tokens summing to prob_cut_off <= prob_cut_off.

__call__

( input_ids: Arrayscores: Arraycur_len: int )

class transformers.FlaxWhisperTimeStampLogitsProcessor

( generate_configmodel_configdecoder_input_length )

Parameters

generate_config (GenerateConfig) — The generate config used to generate the output. The following parameters are required: eos_token_id (int, optional, defaults to 50257): The id of the end-of-sequence token. no_timestamps_token_id (int, optional, defaults to 50363): The id of the "<|notimestamps|>" token. max_initial_timestamp_index (int, optional, defaults to 1): Used to set the maximum value of the initial timestamp. This is used to prevent the model from predicting timestamps that are too far in the future.

Whisper specific Processor. This processor can be used to force a list of tokens. The processor will set their log probs to inf so that they are sampled at their corresponding index.

__call__

( input_idsscorescur_len )

StoppingCriteria

A can be used to change when to stop generation (other than EOS token). Please note that this is exclusivelly available to our PyTorch implementations.

class transformers.StoppingCriteria

( )

Abstract base class for all stopping criteria that can be applied during generation.

__call__

( input_ids: LongTensorscores: FloatTensor**kwargs )

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be scores for each vocabulary token before SoftMax or scores for each vocabulary token after SoftMax.
kwargs (Dict[str, Any], optional) — Additional stopping criteria specific kwargs.

class transformers.StoppingCriteriaList

( iterable = () )

__call__

( input_ids: LongTensorscores: FloatTensor**kwargs )

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
Indices can be obtained using . See and for details.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be scores for each vocabulary token before SoftMax or scores for each vocabulary token after SoftMax.
kwargs (Dict[str, Any], optional) — Additional stopping criteria specific kwargs.

class transformers.MaxLengthCriteria

( max_length: intmax_position_embeddings: typing.Optional[int] = None )

Parameters

max_length (int) — The maximum length that the output sequence can have in number of tokens.
max_position_embeddings (int, optional) — The maximum model length, as defined by the model’s config.max_position_embeddings attribute.

This class can be used to stop generation whenever the full generated number of tokens exceeds max_length. Keep in mind for decoder-only type of transformers, this will include the initial prompted tokens.

__call__

( input_ids: LongTensorscores: FloatTensor**kwargs )

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be scores for each vocabulary token before SoftMax or scores for each vocabulary token after SoftMax.
kwargs (Dict[str, Any], optional) — Additional stopping criteria specific kwargs.

class transformers.MaxTimeCriteria

( max_time: floatinitial_timestamp: typing.Optional[float] = None )

Parameters

max_time (float) — The maximum allowed time in seconds for the generation.
initial_time (float, optional, defaults to time.time()) — The start of the generation allowed time.

This class can be used to stop generation whenever the full generation exceeds some amount of time. By default, the time will start being counted when you initialize this function. You can override this by passing an initial_time.

__call__

( input_ids: LongTensorscores: FloatTensor**kwargs )

Parameters

input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.
scores (torch.FloatTensor of shape (batch_size, config.vocab_size)) — Prediction scores of a language modeling head. These can be scores for each vocabulary token before SoftMax or scores for each vocabulary token after SoftMax.
kwargs (Dict[str, Any], optional) — Additional stopping criteria specific kwargs.

Constraints

class transformers.Constraint

( )

Abstract base class for all constraints that can be applied during generation. It must define how the constraint can be satisfied.

All classes that inherit Constraint must follow the requirement that

Copied

completed = False
while not completed:
    _, completed = constraint.update(constraint.advance())

will always terminate (halt).

advance

( ) → token_ids(torch.tensor)

Returns

token_ids(torch.tensor)

Must be a tensor of a list of indexable tokens, not some integer.

When called, returns the token that would take this constraint one step closer to being fulfilled.

copy

( stateful = False ) → constraint(Constraint)

Returns

constraint(Constraint)

The same constraint as the one being called from.

Creates a new instance of this constraint.

does_advance

( token_id: int )

Reads in a token and returns whether it creates progress.

remaining

( )

Returns the number of remaining steps of advance() in order to complete this constraint.

reset

( )

Resets the state of this constraint to its initialization. We would call this in cases where the fulfillment of a constraint is abrupted by an unwanted token.

test

( )

Tests whether this constraint has been properly defined.

update

( token_id: int ) → stepped(bool)

Returns

stepped(bool)

Whether this constraint has become one step closer to being fulfuilled. completed(bool): Whether this constraint has been completely fulfilled by this token being generated. reset (bool): Whether this constraint has reset its progress by this token being generated.

Reads in a token and returns booleans that indicate the progress made by it. This function will update the state of this object unlikes does_advance(self, token_id: int).

This isn’t to test whether a certain token will advance the progress; it’s to update its state as if it has been generated. This becomes important if token_id != desired token (refer to else statement in PhrasalConstraint)

class transformers.PhrasalConstraint

( token_ids: typing.List[int] )

Parameters

token_ids (List[int]) — The id of the token that must be generated by the output.

class transformers.DisjunctiveConstraint

( nested_token_ids: typing.List[typing.List[int]] )

Parameters

nested_token_ids (List[List[int]]) — a list of words, where each word is a list of ids. This constraint
is fulfilled by generating just one from the list of words. —

class transformers.ConstraintListState

( constraints: typing.List[transformers.generation.beam_constraints.Constraint] )

Parameters

A class for beam scorers to track its progress through a list of constraints.

advance

( )

The list of tokens to generate such that we can make progress. By “list” we don’t mean the list of token that will fully fulfill a constraint.

Given constraints c_i = {t_ij | j == # of tokens}, If we’re not in the middle of progressing through a specific constraint c_i, we return:

[t_k1 for k in indices of unfulfilled constraints]

If we are in the middle of a constraint, then we return: [t_ij], where i is the index of the inprogress constraint, j is the next step for the constraint.

Though we don’t care which constraint is fulfilled first, if we are in the progress of fulfilling a constraint, that’s the only one we’ll return.

reset

( token_ids: typing.Optional[typing.List[int]] )

token_ids: the tokens generated thus far to reset the state of the progress through constraints.

BeamSearch

class transformers.BeamScorer

( )

process

( input_ids: LongTensornext_scores: FloatTensornext_tokens: LongTensornext_indices: LongTensor**kwargs ) → UserDict

Parameters

input_ids (torch.LongTensor of shape (batch_size * num_beams, sequence_length)) — Indices of input sequence tokens in the vocabulary.
next_scores (torch.FloatTensor of shape (batch_size, 2 * num_beams)) — Current scores of the top 2 * num_beams non-finished beam hypotheses.
next_tokens (torch.LongTensor of shape (batch_size, 2 * num_beams)) — input_ids of the tokens corresponding to the top 2 * num_beams non-finished beam hypotheses.
next_indices (torch.LongTensor of shape (batch_size, 2 * num_beams)) — Beam indices indicating to which beam hypothesis the next_tokens correspond.
pad_token_id (int, optional) — The id of the padding token.
eos_token_id (Union[int, List[int]], optional) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.
beam_indices (torch.LongTensor, optional) — Beam indices indicating to which beam hypothesis each token correspond.

Returns

UserDict

A dictionary composed of the fields as defined above:

next_beam_scores (torch.FloatTensor of shape (batch_size * num_beams)) — Updated scores of all non-finished beams.
next_beam_tokens (torch.FloatTensor of shape (batch_size * num_beams)) — Next tokens to be added to the non-finished beam_hypotheses.
next_beam_indices (torch.FloatTensor of shape (batch_size * num_beams)) — Beam indices indicating to which beam the next tokens shall be added.

finalize

( input_ids: LongTensornext_scores: FloatTensornext_tokens: LongTensornext_indices: LongTensormax_length: int**kwargs ) → torch.LongTensor of shape (batch_size * num_return_sequences, sequence_length)

Parameters

input_ids (torch.LongTensor of shape (batch_size * num_beams, sequence_length)) — Indices of input sequence tokens in the vocabulary.
final_beam_scores (torch.FloatTensor of shape (batch_size * num_beams)) — The final scores of all non-finished beams.
final_beam_tokens (torch.FloatTensor of shape (batch_size * num_beams)) — The last tokens to be added to the non-finished beam_hypotheses.
final_beam_indices (torch.FloatTensor of shape (batch_size * num_beams)) — The beam indices indicating to which beam the final_beam_tokens shall be added.
pad_token_id (int, optional) — The id of the padding token.
eos_token_id (Union[int, List[int]], optional) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.

Returns

torch.LongTensor of shape (batch_size * num_return_sequences, sequence_length)

The generated sequences. The second dimension (sequence_length) is either equal to max_length or shorter if all batches finished early due to the eos_token_id.

class transformers.BeamSearchScorer

( batch_size: intnum_beams: intdevice: devicelength_penalty: typing.Optional[float] = 1.0do_early_stopping: typing.Union[str, bool, NoneType] = Falsenum_beam_hyps_to_keep: typing.Optional[int] = 1num_beam_groups: typing.Optional[int] = 1max_length: typing.Optional[int] = None )

Parameters

batch_size (int) — Batch Size of input_ids for which standard beam search decoding is run in parallel.
num_beams (int) — Number of beams for beam search.
device (torch.device) — Defines the device type (e.g., "cpu" or "cuda") on which this instance of BeamSearchScorer will be allocated.
length_penalty (float, optional, defaults to 1.0) — Exponential penalty to the length that is used with beam-based generation. It is applied as an exponent to the sequence length, which in turn is used to divide the score of the sequence. Since the score is the log likelihood of the sequence (i.e. negative), length_penalty > 0.0 promotes longer sequences, while length_penalty < 0.0 encourages shorter sequences.
do_early_stopping (bool or str, optional, defaults to False) — Controls the stopping condition for beam-based methods, like beam-search. It accepts the following values: True, where the generation stops as soon as there are num_beams complete candidates; False, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; "never", where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm).
num_beam_hyps_to_keep (int, optional, defaults to 1) — The number of beam hypotheses that shall be returned upon calling ~transformer.BeamSearchScorer.finalize.
max_length (int, optional) — The maximum length of the sequence to be generated.

process

( input_ids: LongTensornext_scores: FloatTensornext_tokens: LongTensornext_indices: LongTensorpad_token_id: typing.Optional[int] = Noneeos_token_id: typing.Union[int, typing.List[int], NoneType] = Nonebeam_indices: typing.Optional[torch.LongTensor] = Nonegroup_index: typing.Optional[int] = 0 )

finalize

( input_ids: LongTensorfinal_beam_scores: FloatTensorfinal_beam_tokens: LongTensorfinal_beam_indices: LongTensormax_length: intpad_token_id: typing.Optional[int] = Noneeos_token_id: typing.Union[int, typing.List[int], NoneType] = Nonebeam_indices: typing.Optional[torch.LongTensor] = None )

class transformers.ConstrainedBeamSearchScorer

( batch_size: intnum_beams: intconstraints: typing.List[transformers.generation.beam_constraints.Constraint]device: devicelength_penalty: typing.Optional[float] = 1.0do_early_stopping: typing.Union[str, bool, NoneType] = Falsenum_beam_hyps_to_keep: typing.Optional[int] = 1num_beam_groups: typing.Optional[int] = 1max_length: typing.Optional[int] = None )

Parameters

batch_size (int) — Batch Size of input_ids for which standard beam search decoding is run in parallel.
num_beams (int) — Number of beams for beam search.
device (torch.device) — Defines the device type (e.g., "cpu" or "cuda") on which this instance of BeamSearchScorer will be allocated.
length_penalty (float, optional, defaults to 1.0) — Exponential penalty to the length that is used with beam-based generation. It is applied as an exponent to the sequence length, which in turn is used to divide the score of the sequence. Since the score is the log likelihood of the sequence (i.e. negative), length_penalty > 0.0 promotes longer sequences, while length_penalty < 0.0 encourages shorter sequences.
do_early_stopping (bool or str, optional, defaults to False) — Controls the stopping condition for beam-based methods, like beam-search. It accepts the following values: True, where the generation stops as soon as there are num_beams complete candidates; False, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; "never", where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm).
num_beam_hyps_to_keep (int, optional, defaults to 1) — The number of beam hypotheses that shall be returned upon calling ~transformer.BeamSearchScorer.finalize.
max_length (int, optional) — The maximum length of the sequence to be generated.

process

( input_ids: LongTensornext_scores: FloatTensornext_tokens: LongTensornext_indices: LongTensorscores_for_all_vocab: FloatTensorpad_token_id: typing.Optional[int] = Noneeos_token_id: typing.Union[int, typing.List[int], NoneType] = Nonebeam_indices: typing.Optional[torch.LongTensor] = None ) → UserDict

Parameters

input_ids (torch.LongTensor of shape (batch_size * num_beams, sequence_length)) — Indices of input sequence tokens in the vocabulary.
next_scores (torch.FloatTensor of shape (batch_size, 2 * num_beams)) — Current scores of the top 2 * num_beams non-finished beam hypotheses.
next_tokens (torch.LongTensor of shape (batch_size, 2 * num_beams)) — input_ids of the tokens corresponding to the top 2 * num_beams non-finished beam hypotheses.
next_indices (torch.LongTensor of shape (batch_size, 2 * num_beams)) — Beam indices indicating to which beam hypothesis the next_tokens correspond.
scores_for_all_vocab (torch.FloatTensor of shape (batch_size * num_beams, sequence_length)) — The scores of all tokens in the vocabulary for each of the beam hypotheses.
pad_token_id (int, optional) — The id of the padding token.
eos_token_id (Union[int, List[int]], optional) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.
beam_indices (torch.LongTensor, optional) — Beam indices indicating to which beam hypothesis each token correspond.

Returns

UserDict

A dictionary composed of the fields as defined above:

next_beam_scores (torch.FloatTensor of shape (batch_size * num_beams)) — Updated scores of all non-finished beams.
next_beam_tokens (torch.FloatTensor of shape (batch_size * num_beams)) — Next tokens to be added to the non-finished beam_hypotheses.
next_beam_indices (torch.FloatTensor of shape (batch_size * num_beams)) — Beam indices indicating to which beam the next tokens shall be added.

finalize

Utilities

transformers.top_k_top_p_filtering

( logits: FloatTensortop_k: int = 0top_p: float = 1.0filter_value: float = -infmin_tokens_to_keep: int = 1 )

Parameters

top_k (int, optional, defaults to 0) — If > 0, only keep the top k tokens with highest probability (top-k filtering)
min_tokens_to_keep (int, optional, defaults to 1) — Minimumber of tokens we keep per batch example in the output.

Filter a distribution of logits using top-k and/or nucleus (top-p) filtering

transformers.tf_top_k_top_p_filtering

( logitstop_k = 0top_p = 1.0filter_value = -infmin_tokens_to_keep = 1 )

Parameters

top_k (int, optional, defaults to 0) — If > 0, only keep the top k tokens with highest probability (top-k filtering)
min_tokens_to_keep (int, optional, defaults to 1) — Minimumber of tokens we keep per batch example in the output.

Filter a distribution of logits using top-k and/or nucleus (top-p) filtering

Streamers

class transformers.TextStreamer

( tokenizer: AutoTokenizerskip_prompt: bool = False**decode_kwargs )

Parameters

tokenizer (AutoTokenizer) — The tokenized used to decode the tokens.
skip_prompt (bool, optional, defaults to False) — Whether to skip the prompt to .generate() or not. Useful e.g. for chatbots.
decode_kwargs (dict, optional) — Additional keyword arguments to pass to the tokenizer’s decode method.

Simple text streamer that prints the token(s) to stdout as soon as entire words are formed.

The API for the streamer classes is still under development and may change in the future.

Examples:

Copied

>>> from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

>>> tok = AutoTokenizer.from_pretrained("gpt2")
>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> inputs = tok(["An increasing sequence: one,"], return_tensors="pt")
>>> streamer = TextStreamer(tok)

>>> # Despite returning the usual output, the streamer will also print the generated text to stdout.
>>> _ = model.generate(**inputs, streamer=streamer, max_new_tokens=20)
An increasing sequence: one, two, three, four, five, six, seven, eight, nine, ten, eleven,

end

( )

Flushes any remaining cache and prints a newline to stdout.

on_finalized_text

( text: strstream_end: bool = False )

Prints the new text to stdout. If the stream is ending, also prints a newline.

put

( value )

Receives tokens, decodes them, and prints them to stdout as soon as they form entire words.

class transformers.TextIteratorStreamer

( tokenizer: AutoTokenizerskip_prompt: bool = Falsetimeout: typing.Optional[float] = None**decode_kwargs )

Parameters

tokenizer (AutoTokenizer) — The tokenized used to decode the tokens.
skip_prompt (bool, optional, defaults to False) — Whether to skip the prompt to .generate() or not. Useful e.g. for chatbots.
timeout (float, optional) — The timeout for the text queue. If None, the queue will block indefinitely. Useful to handle exceptions in .generate(), when it is called in a separate thread.
decode_kwargs (dict, optional) — Additional keyword arguments to pass to the tokenizer’s decode method.

Streamer that stores print-ready text in a queue, to be used by a downstream application as an iterator. This is useful for applications that benefit from acessing the generated text in a non-blocking way (e.g. in an interactive Gradio demo).

The API for the streamer classes is still under development and may change in the future.

Examples:

Copied

>>> from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
>>> from threading import Thread

>>> tok = AutoTokenizer.from_pretrained("gpt2")
>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
>>> inputs = tok(["An increasing sequence: one,"], return_tensors="pt")
>>> streamer = TextIteratorStreamer(tok)

>>> # Run the generation in a separate thread, so that we can fetch the generated text in a non-blocking way.
>>> generation_kwargs = dict(inputs, streamer=streamer, max_new_tokens=20)
>>> thread = Thread(target=model.generate, kwargs=generation_kwargs)
>>> thread.start()
>>> generated_text = ""
>>> for new_text in streamer:
...     generated_text += new_text
>>> generated_text
'An increasing sequence: one, two, three, four, five, six, seven, eight, nine, ten, eleven,'

on_finalized_text

( text: strstream_end: bool = False )

Put the new text in the queue. If the stream is ending, also put a stop signal in the queue.