Falcon
Overview
Falcon is a class of causal decoder-only models built by TII. The largest Falcon checkpoints have been trained on >=1T tokens of text, with a particular emphasis on the RefinedWeb corpus. They are made available under the Apache 2.0 license.
Falcon’s architecture is modern and optimized for inference, with multi-query attention and support for efficient
attention variants like FlashAttention
. Both ‘base’ models trained only as causal language models as well as
‘instruct’ models that have received further fine-tuning are available.
Falcon models are (as of 2023) some of the largest and most powerful open-source language models, and consistently rank highly in the OpenLLM leaderboard.
Converting custom checkpoints
Falcon models were initially added to the Hugging Face Hub as custom code checkpoints. However, Falcon is now fully
supported in the Transformers library. If you fine-tuned a model from a custom code checkpoint, we recommend converting
your checkpoint to the new in-library format, as this should give significant improvements to stability and
performance, especially for generation, as well as removing the need to use trust_remote_code=True
!
You can convert custom code checkpoints to full Transformers checkpoints using the convert_custom_code_checkpoint.py
script located in the
Falcon model directory
of the Transformers library. To use this script, simply call it with
python convert_custom_code_checkpoint.py --checkpoint_dir my_model
. This will convert your checkpoint in-place, and
you can immediately load it from the directory afterwards with e.g. from_pretrained()
. If your model hasn’t been
uploaded to the Hub, we recommend making a backup before attempting the conversion, just in case!
FalconConfig
class transformers.FalconConfig
< source >( vocab_size = 65024 hidden_size = 4544 num_hidden_layers = 32 num_attention_heads = 71 num_ln_in_parallel_attn = None layer_norm_epsilon = 1e-05 initializer_range = 0.02 use_cache = True hidden_dropout = 0.0 attention_dropout = 0.0 num_kv_heads = None alibi = False new_decoder_architecture = False multi_query = True parallel_attn = True bias = False max_position_embeddings = 2048 rope_theta = 10000.0 rope_scaling = None bos_token_id = 11 eos_token_id = 11 ffn_hidden_size = None activation = 'gelu' **kwargs )
Parameters
- vocab_size (
int
, optional, defaults to 65024) — Vocabulary size of the Falcon model. Defines the number of different tokens that can be represented by theinputs_ids
passed when calling FalconModel - hidden_size (
int
, optional, defaults to 4544) — Dimension of the hidden representations. - num_hidden_layers (
int
, optional, defaults to 32) — Number of hidden layers in the Transformer decoder. - num_attention_heads (
int
, optional, defaults to 71) — Number of attention heads for each attention layer in the Transformer encoder. - num_ln_in_parallel_attn (
int
, optional) — Set to 2 if separate layer norms are to be used for the MLP and the attention output when using parallel attention, otherwise, 1. - layer_norm_epsilon (
float
, optional, defaults to 1e-05) — The epsilon used by the layer normalization layers. - initializer_range (
float
, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices. - use_cache (
bool
, optional, defaults toTrue
) — Whether the model should return the last key/values attentions (not used by all models). Only relevant ifconfig.is_decoder=True
. - hidden_dropout (
float
, optional, defaults to 0.0) — The dropout probability for MLP layers. - attention_dropout (
float
, optional, defaults to 0.0) — The dropout probability for attention layers. - num_kv_heads (
int
, optional) — Number of key-value heads to use per attention layer. If unset, defaults to the same value asnum_attention_heads
. - alibi (
bool
, optional, defaults toFalse
) — Whether to use ALiBi positional biases during self-attention. - new_decoder_architecture (
bool
, optional, defaults toFalse
) — Whether to use the new (Falcon-40B) decoder architecture. IfTrue
, themulti_query
andparallel_attn
arguments are ignored, as the new decoder always uses parallel attention. - multi_query (
bool
, optional, defaults toTrue
) — Whether to use multi-query attention in the decoder. Ignored whennew_decoder_architecture
isTrue
. - parallel_attn (
bool
, optional, defaults toTrue
) — Whether to compute attention in parallel with the feedforward layer. If False, they are consecutive instead, as in the original Transformer architecture. Ignored whennew_decoder_architecture
isTrue
. - bias (
bool
, optional, defaults toFalse
) — Whether to use bias on Linear layers. - max_position_embeddings (
int
, optional, defaults to 2048) — The maximum sequence length that this model might ever be used with, whenalibi
isFalse
. Pretrained Falcon models with RoPE support up to 2048 tokens. - rope_theta (
float
, optional, defaults to 10000.0) — The base period of the RoPE embeddings. - rope_scaling (
Dict
, optional) — Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longermax_position_embeddings
, we recommend you to update this value accordingly. Expected contents:rope_type
(str
): The sub-variant of RoPE to use. Can be one of [‘default’, ‘linear’, ‘dynamic’, ‘yarn’, ‘longrope’, ‘llama3’], with ‘default’ being the original RoPE implementation.factor
(float
, optional): Used with all rope types except ‘default’. The scaling factor to apply to the RoPE embeddings. In most scaling types, afactor
of x will enable the model to handle sequences of length x original maximum pre-trained length.original_max_position_embeddings
(int
, optional): Used with ‘dynamic’, ‘longrope’ and ‘llama3’. The original max position embeddings used during pretraining.attention_factor
(float
, optional): Used with ‘yarn’ and ‘longrope’. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using thefactor
field to infer the suggested value.beta_fast
(float
, optional): Only used with ‘yarn’. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32.beta_slow
(float
, optional): Only used with ‘yarn’. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1.short_factor
(List[float]
, optional): Only used with ‘longrope’. The scaling factor to be applied to short contexts (<original_max_position_embeddings
). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2long_factor
(List[float]
, optional): Only used with ‘longrope’. The scaling factor to be applied to long contexts (<original_max_position_embeddings
). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2low_freq_factor
(float
, optional): Only used with ‘llama3’. Scaling factor applied to low frequency components of the RoPEhigh_freq_factor
(float
, optional*): Only used with ‘llama3’. Scaling factor applied to high frequency components of the RoPE - bos_token_id (
int
, optional, defaults to 11) — The id of the “beginning-of-sequence” token. - eos_token_id (
int
, optional, defaults to 11) — The id of the “end-of-sequence” token. - ffn_hidden_size (
int
, optional) — The hidden size of the feedforward layer in the Transformer decoder. defaults to 4x hidden dim - activation (
str
, optional, defaults to"gelu"
) — The activation function used in the feedforward layer.
This is the configuration class to store the configuration of a FalconModel. It is used to instantiate a Falcon model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the tiiuae/falcon-7b architecture.
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
Example:
>>> from transformers import FalconModel, FalconConfig
>>> # Initializing a small (2-layer) Falcon configuration
>>> configuration = FalconConfig(num_hidden_layers=2)
>>> # Initializing a model from the small configuration
>>> model = FalconModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
FalconModel
class transformers.FalconModel
< source >( config: FalconConfig )
Parameters
- config (FalconConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
The bare Falcon Model transformer outputting raw hidden-states without any specific head on top.
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Union[transformers.cache_utils.Cache, typing.Tuple[typing.Tuple[torch.Tensor, torch.Tensor], ...], NoneType] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None head_mask: typing.Optional[torch.LongTensor] = None inputs_embeds: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None ) → transformers.modeling_outputs.BaseModelOutputWithPastAndCrossAttentions or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, input_ids_length)
) —input_ids_length
=sequence_length
ifpast_key_values
isNone
elsepast_key_values[0][0].shape[2]
(sequence_length
of input past key value states). Indices of input sequence tokens in the vocabulary.If
past_key_values
is used, onlyinput_ids
that do not have their past calculated should be passed asinput_ids
.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- past_key_values (
Cache
ortuple(tuple(torch.FloatTensor))
, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_values
returned by the model at a previous stage of decoding, whenuse_cache=True
orconfig.use_cache=True
.Two formats are allowed:
- a Cache instance, see our kv cache guide;
- Tuple of
tuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
). This is also known as the legacy cache format.
The model will output the same cache format that is fed as input. If no
past_key_values
are passed, the legacy cache format will be returned.If
past_key_values
are used, the user can optionally input only the lastinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - attention_mask (
torch.FloatTensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - head_mask (
torch.FloatTensor
of shape(num_heads,)
or(num_layers, num_heads)
, optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
past_key_values
is used, optionally only the lastinputs_embeds
have to be input (seepast_key_values
). - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - cache_position (
torch.LongTensor
of shape(sequence_length)
, optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids
, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.
Returns
transformers.modeling_outputs.BaseModelOutputWithPastAndCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_outputs.BaseModelOutputWithPastAndCrossAttentions or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (FalconConfig) and inputs.
-
last_hidden_state (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
) — Sequence of hidden-states at the output of the last layer of the model.If
past_key_values
is used only the last hidden-state of the sequences of shape(batch_size, 1, hidden_size)
is output. -
past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
) and optionally ifconfig.is_encoder_decoder=True
2 additional tensors of shape(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)
.Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if
config.is_encoder_decoder=True
in the cross-attention blocks) that can be used (seepast_key_values
input) to speed up sequential decoding. -
hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
-
attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
andconfig.add_cross_attention=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights of the decoder’s cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
The FalconModel forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoTokenizer, FalconModel
>>> import torch
>>> tokenizer = AutoTokenizer.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> model = FalconModel.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_state
FalconForCausalLM
class transformers.FalconForCausalLM
< source >( config: FalconConfig )
Parameters
- config (FalconConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
The Falcon Model transformer with a language modeling head on top (linear layer with weights tied to the input embeddings).
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Union[transformers.cache_utils.Cache, typing.Tuple[typing.Tuple[torch.Tensor, torch.Tensor], ...], NoneType] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None num_logits_to_keep: int = 0 ) → transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, input_ids_length)
) —input_ids_length
=sequence_length
ifpast_key_values
isNone
elsepast_key_values[0][0].shape[2]
(sequence_length
of input past key value states). Indices of input sequence tokens in the vocabulary.If
past_key_values
is used, onlyinput_ids
that do not have their past calculated should be passed asinput_ids
.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- past_key_values (
Cache
ortuple(tuple(torch.FloatTensor))
, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_values
returned by the model at a previous stage of decoding, whenuse_cache=True
orconfig.use_cache=True
.Two formats are allowed:
- a Cache instance, see our kv cache guide;
- Tuple of
tuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
). This is also known as the legacy cache format.
The model will output the same cache format that is fed as input. If no
past_key_values
are passed, the legacy cache format will be returned.If
past_key_values
are used, the user can optionally input only the lastinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - attention_mask (
torch.FloatTensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - head_mask (
torch.FloatTensor
of shape(num_heads,)
or(num_layers, num_heads)
, optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
past_key_values
is used, optionally only the lastinputs_embeds
have to be input (seepast_key_values
). - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - cache_position (
torch.LongTensor
of shape(sequence_length)
, optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids
, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. - labels (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Labels for language modeling. Note that the labels are shifted inside the model, i.e. you can setlabels = input_ids
Indices are selected in[-100, 0, ..., config.vocab_size]
All labels set to-100
are ignored (masked), the loss is only computed for labels in[0, ..., config.vocab_size]
- num_logits_to_keep (
int
, optional) — Calculate logits for the lastnum_logits_to_keep
tokens. If0
, calculate logits for allinput_ids
(special case). Only last token logits are needed for generation, and calculating them only for that token can save memory, which becomes pretty significant for long sequences or large vocabulary size.
Returns
transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (FalconConfig) and inputs.
-
loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenlabels
is provided) — Language modeling loss (for next-token prediction). -
logits (
torch.FloatTensor
of shape(batch_size, sequence_length, config.vocab_size)
) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). -
hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
-
attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple oftorch.FloatTensor
tuples of lengthconfig.n_layers
, with each tuple containing the cached key, value states of the self-attention and the cross-attention layers if model is used in encoder-decoder setting. Only relevant ifconfig.is_decoder = True
.Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.
The FalconForCausalLM forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> import torch
>>> from transformers import AutoTokenizer, FalconForCausalLM
>>> tokenizer = AutoTokenizer.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> model = FalconForCausalLM.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs, labels=inputs["input_ids"])
>>> loss = outputs.loss
>>> logits = outputs.logits
FalconForSequenceClassification
class transformers.FalconForSequenceClassification
< source >( config: FalconConfig )
Parameters
- config (FalconConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
The Falcon Model transformer with a sequence classification head on top (linear layer).
FalconForSequenceClassification uses the last token in order to do the classification, as other causal models (e.g. GPT-1) do.
Since it does classification on the last token, it requires to know the position of the last token. If a
pad_token_id
is defined in the configuration, it finds the last token that is not a padding token in each row. If
no pad_token_id
is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
padding tokens when inputs_embeds
are passed instead of input_ids
, it does the same (take the last value in
each row of the batch).
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor, torch.Tensor], ...]] = None attention_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) → transformers.modeling_outputs.SequenceClassifierOutputWithPast
or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, input_ids_length)
) —input_ids_length
=sequence_length
ifpast_key_values
isNone
elsepast_key_values[0][0].shape[2]
(sequence_length
of input past key value states). Indices of input sequence tokens in the vocabulary.If
past_key_values
is used, onlyinput_ids
that do not have their past calculated should be passed asinput_ids
.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- past_key_values (
Cache
ortuple(tuple(torch.FloatTensor))
, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_values
returned by the model at a previous stage of decoding, whenuse_cache=True
orconfig.use_cache=True
.Two formats are allowed:
- a Cache instance, see our kv cache guide;
- Tuple of
tuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
). This is also known as the legacy cache format.
The model will output the same cache format that is fed as input. If no
past_key_values
are passed, the legacy cache format will be returned.If
past_key_values
are used, the user can optionally input only the lastinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - attention_mask (
torch.FloatTensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - head_mask (
torch.FloatTensor
of shape(num_heads,)
or(num_layers, num_heads)
, optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
past_key_values
is used, optionally only the lastinputs_embeds
have to be input (seepast_key_values
). - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - cache_position (
torch.LongTensor
of shape(sequence_length)
, optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids
, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. - labels (
torch.LongTensor
of shape(batch_size,)
, optional) — Labels for computing the sequence classification/regression loss. Indices should be in[0, ..., config.num_labels - 1]
. Ifconfig.num_labels == 1
a regression loss is computed (Mean-Square loss), Ifconfig.num_labels > 1
a classification loss is computed (Cross-Entropy).
Returns
transformers.modeling_outputs.SequenceClassifierOutputWithPast
or tuple(torch.FloatTensor)
A transformers.modeling_outputs.SequenceClassifierOutputWithPast
or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (FalconConfig) and inputs.
-
loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenlabels
is provided) — Classification (or regression if config.num_labels==1) loss. -
logits (
torch.FloatTensor
of shape(batch_size, config.num_labels)
) — Classification (or regression if config.num_labels==1) scores (before SoftMax). -
past_key_values (
tuple(tuple(torch.FloatTensor))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple oftuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
)Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding. -
hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
-
attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
The FalconForSequenceClassification forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example of single-label classification:
>>> import torch
>>> from transformers import AutoTokenizer, FalconForSequenceClassification
>>> tokenizer = AutoTokenizer.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> model = FalconForSequenceClassification.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_class_id = logits.argmax().item()
>>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)`
>>> num_labels = len(model.config.id2label)
>>> model = FalconForSequenceClassification.from_pretrained("Rocketknight1/falcon-rw-1b", num_labels=num_labels)
>>> labels = torch.tensor([1])
>>> loss = model(**inputs, labels=labels).loss
Example of multi-label classification:
>>> import torch
>>> from transformers import AutoTokenizer, FalconForSequenceClassification
>>> tokenizer = AutoTokenizer.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> model = FalconForSequenceClassification.from_pretrained("Rocketknight1/falcon-rw-1b", problem_type="multi_label_classification")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_class_ids = torch.arange(0, logits.shape[-1])[torch.sigmoid(logits).squeeze(dim=0) > 0.5]
>>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)`
>>> num_labels = len(model.config.id2label)
>>> model = FalconForSequenceClassification.from_pretrained(
... "Rocketknight1/falcon-rw-1b", num_labels=num_labels, problem_type="multi_label_classification"
... )
>>> labels = torch.sum(
... torch.nn.functional.one_hot(predicted_class_ids[None, :].clone(), num_classes=num_labels), dim=1
... ).to(torch.float)
>>> loss = model(**inputs, labels=labels).loss
FalconForTokenClassification
class transformers.FalconForTokenClassification
< source >( config: FalconConfig )
Parameters
- config (FalconConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
Falcon Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks.
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor, torch.Tensor], ...]] = None attention_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) → transformers.modeling_outputs.TokenClassifierOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, input_ids_length)
) —input_ids_length
=sequence_length
ifpast_key_values
isNone
elsepast_key_values[0][0].shape[2]
(sequence_length
of input past key value states). Indices of input sequence tokens in the vocabulary.If
past_key_values
is used, onlyinput_ids
that do not have their past calculated should be passed asinput_ids
.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- past_key_values (
Cache
ortuple(tuple(torch.FloatTensor))
, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_values
returned by the model at a previous stage of decoding, whenuse_cache=True
orconfig.use_cache=True
.Two formats are allowed:
- a Cache instance, see our kv cache guide;
- Tuple of
tuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
). This is also known as the legacy cache format.
The model will output the same cache format that is fed as input. If no
past_key_values
are passed, the legacy cache format will be returned.If
past_key_values
are used, the user can optionally input only the lastinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - attention_mask (
torch.FloatTensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - head_mask (
torch.FloatTensor
of shape(num_heads,)
or(num_layers, num_heads)
, optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
past_key_values
is used, optionally only the lastinputs_embeds
have to be input (seepast_key_values
). - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - cache_position (
torch.LongTensor
of shape(sequence_length)
, optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids
, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. - labels (
torch.LongTensor
of shape(batch_size,)
, optional) — Labels for computing the sequence classification/regression loss. Indices should be in[0, ..., config.num_labels - 1]
. Ifconfig.num_labels == 1
a regression loss is computed (Mean-Square loss), Ifconfig.num_labels > 1
a classification loss is computed (Cross-Entropy).
Returns
transformers.modeling_outputs.TokenClassifierOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.TokenClassifierOutput or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (FalconConfig) and inputs.
-
loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenlabels
is provided) — Classification loss. -
logits (
torch.FloatTensor
of shape(batch_size, sequence_length, config.num_labels)
) — Classification scores (before SoftMax). -
hidden_states (
tuple(torch.FloatTensor)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple oftorch.FloatTensor
(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
-
attentions (
tuple(torch.FloatTensor)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
The FalconForTokenClassification forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoTokenizer, FalconForTokenClassification
>>> import torch
>>> tokenizer = AutoTokenizer.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> model = FalconForTokenClassification.from_pretrained("Rocketknight1/falcon-rw-1b")
>>> inputs = tokenizer(
... "HuggingFace is a company based in Paris and New York", add_special_tokens=False, return_tensors="pt"
... )
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_token_class_ids = logits.argmax(-1)
>>> # Note that tokens are classified rather then input words which means that
>>> # there might be more predicted token classes than words.
>>> # Multiple token classes might account for the same word
>>> predicted_tokens_classes = [model.config.id2label[t.item()] for t in predicted_token_class_ids[0]]
>>> labels = predicted_token_class_ids
>>> loss = model(**inputs, labels=labels).loss
FalconForQuestionAnswering
class transformers.FalconForQuestionAnswering
< source >( config )
Parameters
- config (FalconConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
The Falcon Model transformer with a span classification head on top for extractive question-answering tasks like
SQuAD (a linear layers on top of the hidden-states output to compute span start logits
and span end logits
).
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.FloatTensor] = None head_mask: typing.Optional[torch.FloatTensor] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None start_positions: typing.Optional[torch.LongTensor] = None end_positions: typing.Optional[torch.LongTensor] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None )
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, input_ids_length)
) —input_ids_length
=sequence_length
ifpast_key_values
isNone
elsepast_key_values[0][0].shape[2]
(sequence_length
of input past key value states). Indices of input sequence tokens in the vocabulary.If
past_key_values
is used, onlyinput_ids
that do not have their past calculated should be passed asinput_ids
.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- past_key_values (
Cache
ortuple(tuple(torch.FloatTensor))
, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_values
returned by the model at a previous stage of decoding, whenuse_cache=True
orconfig.use_cache=True
.Two formats are allowed:
- a Cache instance, see our kv cache guide;
- Tuple of
tuple(torch.FloatTensor)
of lengthconfig.n_layers
, with each tuple having 2 tensors of shape(batch_size, num_heads, sequence_length, embed_size_per_head)
). This is also known as the legacy cache format.
The model will output the same cache format that is fed as input. If no
past_key_values
are passed, the legacy cache format will be returned.If
past_key_values
are used, the user can optionally input only the lastinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, 1)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - attention_mask (
torch.FloatTensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - head_mask (
torch.FloatTensor
of shape(num_heads,)
or(num_layers, num_heads)
, optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in[0, 1]
:- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix.If
past_key_values
is used, optionally only the lastinputs_embeds
have to be input (seepast_key_values
). - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - cache_position (
torch.LongTensor
of shape(sequence_length)
, optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids
, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. - start_positions (
torch.LongTensor
of shape(batch_size,)
, optional) — Labels for position (index) of the start of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (sequence_length
). Position outside of the sequence are not taken into account for computing the loss. - end_positions (
torch.LongTensor
of shape(batch_size,)
, optional) — Labels for position (index) of the end of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (sequence_length
). Position outside of the sequence are not taken into account for computing the loss.
The FalconForQuestionAnswering forward method, overrides the __call__
special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.