Transformers documentation
Mistral
Mistral
Mistral is a 7B parameter language model, available as a pretrained and instruction-tuned variant, focused on balancing the scaling costs of large models with performance and efficient inference. This model uses sliding window attention (SWA) trained with a 8K context length and a fixed cache size to handle longer sequences more effectively. Grouped-query attention (GQA) speeds up inference and reduces memory requirements. Mistral also features a byte-fallback BPE tokenizer to improve token handling and efficiency by ensuring characters are never mapped to out-of-vocabulary tokens.
You can find all the original Mistral checkpoints under the Mistral AI_ organization.
Click on the Mistral models in the right sidebar for more examples of how to apply Mistral to different language tasks.
The example below demonstrates how to chat with Pipeline or the AutoModel, and from the command line.
>>> import torch
>>> from transformers import pipeline
>>> messages = [
... {"role": "user", "content": "What is your favourite condiment?"},
... {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
... {"role": "user", "content": "Do you have mayonnaise recipes?"}
... ]
>>> chatbot = pipeline("text-generation", model="mistralai/Mistral-7B-Instruct-v0.3", torch_dtype=torch.bfloat16, device=0)
>>> chatbot(messages)
Quantization reduces the memory burden of large models by representing the weights in a lower precision. Refer to the Quantization overview for more available quantization backends.
The example below uses bitsandbytes to only quantize the weights to 4-bits.
>>> import torch
>>> from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
>>> # specify how to quantize the model
>>> quantization_config = BitsAndBytesConfig(
... load_in_4bit=True,
... bnb_4bit_quant_type="nf4",
... bnb_4bit_compute_dtype="torch.float16",
... )
>>> model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.3", quantization_config=True, torch_dtype=torch.bfloat16, device_map="auto")
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.3")
>>> prompt = "My favourite condiment is"
>>> messages = [
... {"role": "user", "content": "What is your favourite condiment?"},
... {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
... {"role": "user", "content": "Do you have mayonnaise recipes?"}
... ]
>>> model_inputs = tokenizer.apply_chat_template(messages, return_tensors="pt").to("cuda")
>>> generated_ids = model.generate(model_inputs, max_new_tokens=100, do_sample=True)
>>> tokenizer.batch_decode(generated_ids)[0]
"The expected output"
Use the AttentionMaskVisualizer to better understand what tokens the model can and cannot attend to.
>>> from transformers.utils.attention_visualizer import AttentionMaskVisualizer
>>> visualizer = AttentionMaskVisualizer("mistralai/Mistral-7B-Instruct-v0.3")
>>> visualizer("Do you have mayonnaise recipes?")

MistralConfig
class transformers.MistralConfig
< source >( vocab_size = 32000 hidden_size = 4096 intermediate_size = 14336 num_hidden_layers = 32 num_attention_heads = 32 num_key_value_heads = 8 head_dim = None hidden_act = 'silu' max_position_embeddings = 131072 initializer_range = 0.02 rms_norm_eps = 1e-06 use_cache = True pad_token_id = None bos_token_id = 1 eos_token_id = 2 tie_word_embeddings = False rope_theta = 10000.0 sliding_window = 4096 attention_dropout = 0.0 **kwargs )
Parameters
- vocab_size (
int
, optional, defaults to 32000) — Vocabulary size of the Mistral model. Defines the number of different tokens that can be represented by theinputs_ids
passed when calling MistralModel - hidden_size (
int
, optional, defaults to 4096) — Dimension of the hidden representations. - intermediate_size (
int
, optional, defaults to 14336) — Dimension of the MLP representations. - num_hidden_layers (
int
, optional, defaults to 32) — Number of hidden layers in the Transformer encoder. - num_attention_heads (
int
, optional, defaults to 32) — Number of attention heads for each attention layer in the Transformer encoder. - num_key_value_heads (
int
, optional, defaults to 8) — This is the number of key_value heads that should be used to implement Grouped Query Attention. Ifnum_key_value_heads=num_attention_heads
, the model will use Multi Head Attention (MHA), ifnum_key_value_heads=1
the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out this paper. If it is not specified, will default to8
. - head_dim (
int
, optional, defaults tohidden_size // num_attention_heads
) — The attention head dimension. - hidden_act (
str
orfunction
, optional, defaults to"silu"
) — The non-linear activation function (function or string) in the decoder. - max_position_embeddings (
int
, optional, defaults to4096*32
) — The maximum sequence length that this model might ever be used with. Mistral’s sliding window attention allows sequence of up to 4096*32 tokens. - initializer_range (
float
, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices. - rms_norm_eps (
float
, optional, defaults to 1e-06) — The epsilon used by the rms normalization layers. - use_cache (
bool
, optional, defaults toTrue
) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant ifconfig.is_decoder=True
. - pad_token_id (
int
, optional) — The id of the padding token. - bos_token_id (
int
, optional, defaults to 1) — The id of the “beginning-of-sequence” token. - eos_token_id (
int
, optional, defaults to 2) — The id of the “end-of-sequence” token. - tie_word_embeddings (
bool
, optional, defaults toFalse
) — Whether the model’s input and output word embeddings should be tied. - rope_theta (
float
, optional, defaults to 10000.0) — The base period of the RoPE embeddings. - sliding_window (
int
, optional, defaults to 4096) — Sliding window attention window size. If not specified, will default to4096
. - attention_dropout (
float
, optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
This is the configuration class to store the configuration of a MistralModel. It is used to instantiate an Mistral 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 Mistral-7B-v0.1 or Mistral-7B-Instruct-v0.1.
mistralai/Mistral-7B-v0.1 mistralai/Mistral-7B-Instruct-v0.1
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
>>> from transformers import MistralModel, MistralConfig
>>> # Initializing a Mistral 7B style configuration
>>> configuration = MistralConfig()
>>> # Initializing a model from the Mistral 7B style configuration
>>> model = MistralModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
MistralCommonTokenizer
class transformers.MistralCommonTokenizer
< source >( tokenizer_path: typing.Union[str, os.PathLike, pathlib.Path] mode: ValidationMode = <ValidationMode.test: 'test'> model_max_length: int = 1000000000000000019884624838656 padding_side: str = 'left' truncation_side: str = 'right' model_input_names: typing.Optional[list[str]] = None clean_up_tokenization_spaces: bool = False **kwargs )
Class to wrap mistral-common
tokenizers.
mistral-common
is the official tokenizer library for Mistral AI models. To use it, you need to install it with:
pip install transformers[mistral-common]
Otherwise the tokenizer falls back to the Transformers implementation of the tokenizer.
For more info on mistral-common
, see mistral-common.
This class is a wrapper around a mistral_common.tokens.tokenizers.mistral.MistralTokenizer
.
It provides a Hugging Face compatible interface to tokenize using the official mistral-common tokenizer.
Supports the following methods from the PreTrainedTokenizerBase
class:
- get_vocab(): Returns the vocabulary as a dictionary of token to index.
- encode(): Encode a string to a list of integers.
- decode(): Decode a list of integers to a string.
- batch_decode(): Decode a batch of list of integers to a list of strings.
- convert_tokens_to_ids(): Convert a list of tokens to a list of integers.
- convert_ids_to_tokens(): Convert a list of integers to a list of tokens.
- tokenize(): Tokenize a string.
- get_special_tokens_mask(): Get the special tokens mask for a list of tokens.
- prepare_for_model(): Prepare a list of inputs for the model.
- pad(): Pad a list of inputs to the same length.
- truncate_sequences(): Truncate a list of sequences to the same length.
- apply_chat_template(): Apply a chat template to a list of messages.
__call__()
: Tokenize a string or a list of strings.- from_pretrained(): Download and cache a pretrained tokenizer from the Hugging Face model hub or local directory.
- save_pretrained(): Save a tokenizer to a directory, so it can be reloaded using the
from_pretrained
class method. - push_to_hub(): Upload tokenizer to the Hugging Face model hub.
Here are the key differences with the PreTrainedTokenizerBase
class:
- Pair of sequences are not supported. The signature have been kept for compatibility but all arguments related to pair of sequences are ignored. The return values of pairs are returned as
None
. - The
is_split_into_words
argument is not supported. - The
return_token_type_ids
argument is not supported. - It is not possible to add new tokens to the tokenizer. Also the special tokens are handled differently from Transformers. In
mistral-common
, special tokens are never encoded directly. This means that:tokenizer.encode("<s>")
will not return the ID of the<s>
token. Instead, it will return a list of IDs corresponding to the tokenization of the string"<s>"
. For more information, see the mistral-common documentation.
If you have suggestions to improve this class, please open an issue on the mistral-common GitHub repository if it is related to the tokenizer or on the Transformers GitHub repository if it is related to the Hugging Face interface.
apply_chat_template
< source >( conversation: typing.Union[list[dict[str, str]], list[list[dict[str, str]]]] tools: typing.Optional[list[typing.Union[dict, typing.Callable]]] = None continue_final_message: bool = False tokenize: bool = True padding: typing.Union[bool, str, transformers.utils.generic.PaddingStrategy] = False truncation: bool = False max_length: typing.Optional[int] = None return_tensors: typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None return_dict: bool = False **kwargs ) → Union[str, List[int], List[str], List[List[int]], BatchEncoding]
Parameters
- conversation (Union[List[Dict[str, str]], List[List[Dict[str, str]]]]) — A list of dicts with “role” and “content” keys, representing the chat history so far.
- tools (
List[Union[Dict, Callable]]
, optional) — A list of tools (callable functions) that will be accessible to the model. If the template does not support function calling, this argument will have no effect. Each tool should be passed as a JSON Schema, giving the name, description and argument types for the tool. See our chat templating guide for more information. - continue_final_message (bool, optional) —
If this is set, the chat will be formatted so that the final
message in the chat is open-ended, without any EOS tokens. The model will continue this message
rather than starting a new one. This allows you to “prefill” part of
the model’s response for it. Cannot be used at the same time as
add_generation_prompt
. - tokenize (
bool
, defaults toTrue
) — Whether to tokenize the output. IfFalse
, the output will be a string. - padding (
bool
,str
or PaddingStrategy, optional, defaults toFalse
) — Select a strategy to pad the returned sequences (according to the model’s padding side and padding index) among:True
or'longest'
: Pad to the longest sequence in the batch (or no padding if only a single sequence if provided).'max_length'
: Pad to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.False
or'do_not_pad'
(default): No padding (i.e., can output a batch with sequences of different lengths).
- truncation (
bool
, defaults toFalse
) — Whether to truncate sequences at the maximum length. Has no effect if tokenize isFalse
. - max_length (
int
, optional) — Maximum length (in tokens) to use for padding or truncation. Has no effect if tokenize isFalse
. If not specified, the tokenizer’smax_length
attribute will be used as a default. - return_tensors (
str
or TensorType, optional) — If set, will return tensors of a particular framework. Has no effect if tokenize isFalse
. Acceptable values are:'pt'
: Return PyTorchtorch.Tensor
objects.
- return_dict (
bool
, defaults toFalse
) — Whether to return a dictionary with named outputs. Has no effect if tokenize isFalse
. If at least one conversation contains an image, its pixel values will be returned in thepixel_values
key. - kwargs (additional keyword arguments, optional) —
Not supported by
MistralCommonTokenizer.apply_chat_template
. Will raise an error if used.
Returns
Union[str, List[int], List[str], List[List[int]], BatchEncoding]
A list of token ids representing the tokenized chat so far, including control
tokens. This output is ready to pass to the model, either directly or via methods like generate()
.
Converts a list of dictionaries with "role"
and "content"
keys to a list of token
ids.
batch_decode
< source >( sequences: typing.Union[list[int], list[list[int]], ForwardRef('np.ndarray'), ForwardRef('torch.Tensor')] skip_special_tokens: bool = False clean_up_tokenization_spaces: typing.Optional[bool] = None **kwargs ) → List[str]
Parameters
- sequences (
Union[List[int], List[List[int]], np.ndarray, torch.Tensor]
) — List of tokenized input ids. Can be obtained using the__call__
method. - skip_special_tokens (
bool
, optional, defaults toFalse
) — Whether or not to remove special tokens in the decoding. - clean_up_tokenization_spaces (
bool
, optional) — Whether or not to clean up the tokenization spaces. IfNone
, will default toself.clean_up_tokenization_spaces
. - kwargs (additional keyword arguments, optional) —
Not supported by
MistralCommonTokenizer.batch_decode
. Will raise an error if used.
Returns
List[str]
The list of decoded sentences.
Convert a list of lists of token ids into a list of strings by calling decode.
convert_ids_to_tokens
< source >( ids: typing.Union[int, list[int]] skip_special_tokens: bool = False ) → str
or List[str]
Converts a single index or a sequence of indices in a token or a sequence of tokens, using the vocabulary and added tokens.
convert_tokens_to_ids
< source >( tokens: typing.Union[str, list[str]] ) → int
or List[int]
Converts a token string (or a sequence of tokens) in a single integer id (or a sequence of ids), using the vocabulary.
decode
< source >( token_ids: typing.Union[int, list[int], ForwardRef('np.ndarray'), ForwardRef('torch.Tensor')] skip_special_tokens: bool = False clean_up_tokenization_spaces: typing.Optional[bool] = None **kwargs ) → str
Parameters
- token_ids (
Union[int, List[int], np.ndarray, torch.Tensor]
) — List of tokenized input ids. Can be obtained using the__call__
method. - skip_special_tokens (
bool
, optional, defaults toFalse
) — Whether or not to remove special tokens in the decoding. - clean_up_tokenization_spaces (
bool
, optional) — Whether or not to clean up the tokenization spaces. IfNone
, will default toself.clean_up_tokenization_spaces
. - kwargs (additional keyword arguments, optional) —
Not supported by
MistralCommonTokenizer.decode
. Will raise an error if used.
Returns
str
The decoded sentence.
Converts a sequence of ids in a string, using the tokenizer and vocabulary with options to remove special tokens and clean up tokenization spaces.
encode
< source >( text: typing.Union[str, list[int]] text_pair: None = None add_special_tokens: bool = True padding: typing.Union[bool, str, transformers.utils.generic.PaddingStrategy] = False truncation: typing.Union[bool, str, transformers.tokenization_utils_base.TruncationStrategy, NoneType] = None max_length: typing.Optional[int] = None stride: int = 0 pad_to_multiple_of: typing.Optional[int] = None padding_side: typing.Optional[str] = None return_tensors: typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None verbose: bool = True **kwargs ) → List[int]
, torch.Tensor
Parameters
- text (
str
orList[int]
) — The first sequence to be encoded. This can be a string or a list of integers (tokenized string ids). - text_pair (
None
, optional) — Not supported byMistralCommonTokenizer.encode
. Kept to matchPreTrainedTokenizerBase.encode
signature. - add_special_tokens (
bool
, optional, defaults toTrue
) — Whether or not to add special tokens when encoding the sequences. This will use the underlyingPretrainedTokenizerBase.build_inputs_with_special_tokens
function, which defines which tokens are automatically added to the input ids. This is useful if you want to addbos
oreos
tokens automatically. - padding (
bool
,str
or PaddingStrategy, optional, defaults toFalse
) — Activates and controls padding. Accepts the following values:True
or'longest'
: Pad to the longest sequence in the batch (or no padding if only a single sequence is provided).'max_length'
: Pad to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.False
or'do_not_pad'
(default): No padding (i.e., can output a batch with sequences of different lengths).
- truncation (
bool
,str
or TruncationStrategy, optional, defaults toFalse
) — Activates and controls truncation. Accepts the following values:True
or'longest_first'
: Truncate to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.False
or'do_not_truncate'
(default): No truncation (i.e., can output batch with sequence lengths greater than the model maximum admissible input size).
- max_length (
int
, optional) — Controls the maximum length to use by one of the truncation/padding parameters.If left unset or set to
None
, this will use the predefined model maximum length if a maximum length is required by one of the truncation/padding parameters. If the model has no specific maximum input length (like XLNet) truncation/padding to a maximum length will be deactivated. - stride (
int
, optional, defaults to 0) — If set to a number along withmax_length
, the overflowing tokens returned whenreturn_overflowing_tokens=True
will contain some tokens from the end of the truncated sequence returned to provide some overlap between truncated and overflowing sequences. The value of this argument defines the number of overlapping tokens. - pad_to_multiple_of (
int
, optional) — If set will pad the sequence to a multiple of the provided value. Requirespadding
to be activated. This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability>= 7.5
(Volta). - padding_side (
str
, optional) — The side on which the model should have padding applied. Should be selected between [‘right’, ‘left’]. Default value is picked from the class attribute of the same name. - return_tensors (
str
or TensorType, optional) — If set, will return tensors instead of list of python integers. Acceptable values are:'pt'
: Return PyTorchtorch.Tensor
objects.
- **kwargs — Not supported by
MistralCommonTokenizer.encode
. Will raise an error if used.
Returns
List[int]
, torch.Tensor
The tokenized ids of the text.
Converts a string to a sequence of ids (integer), using the tokenizer and vocabulary.
from_pretrained
< source >( pretrained_model_name_or_path: typing.Union[str, os.PathLike] *init_inputs mode: ValidationMode = <ValidationMode.test: 'test'> cache_dir: typing.Union[str, os.PathLike, NoneType] = None force_download: bool = False local_files_only: bool = False token: typing.Union[bool, str, NoneType] = None revision: str = 'main' model_max_length: int = 1000000000000000019884624838656 padding_side: str = 'left' truncation_side: str = 'right' model_input_names: typing.Optional[list[str]] = None clean_up_tokenization_spaces: bool = False **kwargs )
Parameters
- pretrained_model_name_or_path (
str
oros.PathLike
) — Can be either:- A string, the model id of a predefined tokenizer hosted inside a model repo on huggingface.co.
- A path to a directory containing the tokenizer config, for instance saved
using the
MistralCommonTokenizer.tokenization_mistral_common.save_pretrained
method, e.g.,./my_model_directory/
.
- mode (
ValidationMode
, optional, defaults toValidationMode.test
) — Validation mode for theMistralTokenizer
tokenizer. - cache_dir (
str
oros.PathLike
, optional) — Path to a directory in which a downloaded predefined tokenizer vocabulary files should be cached if the standard cache should not be used. - force_download (
bool
, optional, defaults toFalse
) — Whether or not to force the (re-)download the vocabulary files and override the cached versions if they exist. - token (
str
or bool, optional) — The token to use as HTTP bearer authorization for remote files. IfTrue
, will use the token generated when runninghuggingface-cli login
(stored in~/.huggingface
). - local_files_only (
bool
, optional, defaults toFalse
) — Whether or not to only rely on local files and not to attempt to download any files. - revision (
str
, optional, defaults to"main"
) — The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a git-based system for storing models and other artifacts on huggingface.co, sorevision
can be any identifier allowed by git. - max_length (
int
, optional) — Controls the maximum length to use by one of the truncation/padding parameters.If left unset or set to
None
, this will use the predefined model maximum length if a maximum length is required by one of the truncation/padding parameters. If the model has no specific maximum input length (like XLNet) truncation/padding to a maximum length will be deactivated. - padding_side (
str
, optional, defaults to"left"
) — The side on which the model should have padding applied. Should be selected between [‘right’, ‘left’]. Default value is picked from the class attribute of the same name. - truncation_side (
str
, optional, defaults to"right"
) — The side on which the model should have truncation applied. Should be selected between [‘right’, ‘left’]. - model_input_names (
List[string]
, optional) — The list of inputs accepted by the forward pass of the model (like"token_type_ids"
or"attention_mask"
). Default value is picked from the class attribute of the same name. - clean_up_tokenization_spaces (
bool
, optional, defaults toFalse
) — Whether or not the model should cleanup the spaces that were added when splitting the input text during the tokenization process. - kwargs (additional keyword arguments, optional) —
Not supported by
MistralCommonTokenizer.from_pretrained
. Will raise an error if used.
Instantiate a MistralCommonTokenizer
from a predefined
tokenizer.
get_special_tokens_mask
< source >( token_ids_0: list token_ids_1: None = None already_has_special_tokens: bool = False ) → A list of integers in the range [0, 1]
Parameters
- token_ids_0 (
List[int]
) — List of ids of the sequence. - token_ids_1 (
List[int]
, optional) — Not supported byMistralCommonTokenizer
. Kept to match the interface ofPreTrainedTokenizerBase
. - already_has_special_tokens (
bool
, optional, defaults toFalse
) — Whether or not the token list is already formatted with special tokens for the model.
Returns
A list of integers in the range [0, 1]
1 for a special token, 0 for a sequence token.
Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding
special tokens using the tokenizer prepare_for_model
or encode_plus
methods.
Returns the vocabulary as a dictionary of token to index.
This is a lossy conversion. There may be multiple token ids that decode to the same string due to partial UTF-8 byte sequences being converted to �.
pad
< source >( encoded_inputs: typing.Union[transformers.tokenization_utils_base.BatchEncoding, list[transformers.tokenization_utils_base.BatchEncoding], dict[str, list[int]], dict[str, list[list[int]]], list[dict[str, list[int]]]] padding: typing.Union[bool, str, transformers.utils.generic.PaddingStrategy] = True max_length: typing.Optional[int] = None pad_to_multiple_of: typing.Optional[int] = None padding_side: typing.Optional[str] = None return_attention_mask: typing.Optional[bool] = None return_tensors: typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None verbose: bool = True )
Parameters
- encoded_inputs (BatchEncoding, list of BatchEncoding,
Dict[str, List[int]]
,Dict[str, List[List[int]]
orList[Dict[str, List[int]]]
) — Tokenized inputs. Can represent one input (BatchEncoding orDict[str, List[int]]
) or a batch of tokenized inputs (list of BatchEncoding, Dict[str, List[List[int]]] or List[Dict[str, List[int]]]) so you can use this method during preprocessing as well as in a PyTorch Dataloader collate function.Instead of
List[int]
you can have tensors (numpy arrays, PyTorch tensors), see the note above for the return type. - padding (
bool
,str
or PaddingStrategy, optional, defaults toTrue
) — Select a strategy to pad the returned sequences (according to the model’s padding side and padding index) among:True
or'longest'
(default): Pad to the longest sequence in the batch (or no padding if only a single sequence if provided).'max_length'
: Pad to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.False
or'do_not_pad'
: No padding (i.e., can output a batch with sequences of different lengths).
- max_length (
int
, optional) — Maximum length of the returned list and optionally padding length (see above). - pad_to_multiple_of (
int
, optional) — If set will pad the sequence to a multiple of the provided value.This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability
>= 7.5
(Volta). - padding_side (
str
, optional) — The side on which the model should have padding applied. Should be selected between [‘right’, ‘left’]. Default value is picked from the class attribute of the same name. - return_attention_mask (
bool
, optional) — Whether to return the attention mask. If left to the default, will return the attention mask according to the specific tokenizer’s default, defined by thereturn_outputs
attribute. - return_tensors (
str
or TensorType, optional) — If set, will return tensors instead of list of python integers. Acceptable values are:'pt'
: Return PyTorchtorch.Tensor
objects.'np'
: Return Numpynp.ndarray
objects.
- verbose (
bool
, optional, defaults toTrue
) — Whether or not to print more information and warnings.
Pad a single encoded input or a batch of encoded inputs up to predefined length or to the max sequence length in the batch.
Padding side (left/right) padding token ids are defined at the tokenizer level (with self.padding_side
,
self.pad_token_id
).
If the encoded_inputs
passed are dictionary of numpy arrays, PyTorch tensors, the
result will use the same type unless you provide a different tensor type with return_tensors
. In the case of
PyTorch tensors, you will lose the specific device of your tensors however.
prepare_for_model
< source >( ids: list pair_ids: None = None add_special_tokens: bool = True padding: typing.Union[bool, str, transformers.utils.generic.PaddingStrategy] = False truncation: typing.Union[bool, str, transformers.tokenization_utils_base.TruncationStrategy, NoneType] = None max_length: typing.Optional[int] = None stride: int = 0 pad_to_multiple_of: typing.Optional[int] = None padding_side: typing.Optional[str] = None return_tensors: typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None return_attention_mask: typing.Optional[bool] = None return_overflowing_tokens: bool = False return_special_tokens_mask: bool = False return_length: bool = False verbose: bool = True prepend_batch_axis: bool = False **kwargs ) → BatchEncoding
Parameters
- ids (
List[int]
) — Tokenized input ids of the first sequence. - pair_ids (
None
, optional) — Not supported byMistralCommonTokenizer
. Kept to match the interface ofPreTrainedTokenizerBase
. - add_special_tokens (
bool
, optional, defaults toTrue
) — Whether or not to add special tokens when encoding the sequences. This will use the underlyingPretrainedTokenizerBase.build_inputs_with_special_tokens
function, which defines which tokens are automatically added to the input ids. This is useful if you want to addbos
oreos
tokens automatically. - padding (
bool
,str
or PaddingStrategy, optional, defaults toFalse
) — Activates and controls padding. Accepts the following values:True
or'longest'
: Pad to the longest sequence in the batch (or no padding if only a single sequence is provided).'max_length'
: Pad to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.False
or'do_not_pad'
(default): No padding (i.e., can output a batch with sequences of different lengths).
- truncation (
bool
,str
or TruncationStrategy, optional, defaults toFalse
) — Activates and controls truncation. Accepts the following values:True
or'longest_first'
: Truncate to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.False
or'do_not_truncate'
(default): No truncation (i.e., can output batch with sequence lengths greater than the model maximum admissible input size).
- max_length (
int
, optional) — Controls the maximum length to use by one of the truncation/padding parameters.If left unset or set to
None
, this will use the predefined model maximum length if a maximum length is required by one of the truncation/padding parameters. If the model has no specific maximum input length (like XLNet) truncation/padding to a maximum length will be deactivated. - stride (
int
, optional, defaults to 0) — If set to a number along withmax_length
, the overflowing tokens returned whenreturn_overflowing_tokens=True
will contain some tokens from the end of the truncated sequence returned to provide some overlap between truncated and overflowing sequences. The value of this argument defines the number of overlapping tokens. - pad_to_multiple_of (
int
, optional) — If set will pad the sequence to a multiple of the provided value. Requirespadding
to be activated. This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability>= 7.5
(Volta). - padding_side (
str
, optional) — The side on which the model should have padding applied. Should be selected between [‘right’, ‘left’]. Default value is picked from the class attribute of the same name. - return_tensors (
str
or TensorType, optional) — If set, will return tensors instead of list of python integers. Acceptable values are:'pt'
: Return PyTorchtorch.Tensor
objects.
- return_attention_mask (
bool
, optional) — Whether to return the attention mask. If left to the default, will return the attention mask according to the specific tokenizer’s default, defined by thereturn_outputs
attribute. - return_overflowing_tokens (
bool
, optional, defaults toFalse
) — Whether or not to return overflowing token sequences. If a pair of sequences of input ids (or a batch of pairs) is provided withtruncation_strategy = longest_first
orTrue
, an error is raised instead of returning overflowing tokens. - return_special_tokens_mask (
bool
, optional, defaults toFalse
) — Whether or not to return special tokens mask information. - return_offsets_mapping (
bool
, optional, defaults toFalse
) — Whether or not to return(char_start, char_end)
for each token.This is only available on fast tokenizers inheriting from PreTrainedTokenizerFast, if using Python’s tokenizer, this method will raise
NotImplementedError
. - return_length (
bool
, optional, defaults toFalse
) — Whether or not to return the lengths of the encoded inputs. - verbose (
bool
, optional, defaults toTrue
) — Whether or not to print more information and warnings. - **kwargs — passed to the
self.tokenize()
method
Returns
A BatchEncoding with the following fields:
-
input_ids — List of token ids to be fed to a model.
-
attention_mask — List of indices specifying which tokens should be attended to by the model (when
return_attention_mask=True
or if “attention_mask” is inself.model_input_names
). -
overflowing_tokens — List of overflowing tokens sequences (when a
max_length
is specified andreturn_overflowing_tokens=True
). -
num_truncated_tokens — Number of tokens truncated (when a
max_length
is specified andreturn_overflowing_tokens=True
). -
special_tokens_mask — List of 0s and 1s, with 1 specifying added special tokens and 0 specifying regular sequence tokens (when
add_special_tokens=True
andreturn_special_tokens_mask=True
). -
length — The length of the inputs (when
return_length=True
)
Prepares a sequence of input id so that it can be used by the model. It adds special tokens, truncates sequences if overflowing while taking into account the special tokens and manages a moving window (with user defined stride) for overflowing tokens.
save_pretrained
< source >( save_directory: typing.Union[str, os.PathLike, pathlib.Path] push_to_hub: bool = False token: typing.Union[bool, str, NoneType] = None commit_message: typing.Optional[str] = None repo_id: typing.Optional[str] = None private: typing.Optional[bool] = None repo_url: typing.Optional[str] = None organization: typing.Optional[str] = None **kwargs ) → A tuple of str
Parameters
- save_directory (
str
oros.PathLike
) — The path to a directory where the tokenizer will be saved. - push_to_hub (
bool
, optional, defaults toFalse
) — Whether or not to push your model to the Hugging Face model hub after saving it. You can specify the repository you want to push to withrepo_id
(will default to the name ofsave_directory
in your namespace). - token (
str
or bool, optional, defaults toNone
) — The token to use to push to the model hub. IfTrue
, will use the token in theHF_TOKEN
environment variable. - commit_message (
str
, optional) — The commit message to use when pushing to the hub. - repo_id (
str
, optional) — The name of the repository to which push to the Hub. - private (
bool
, optional) — Whether the model repository is private or not. - repo_url (
str
, optional) — The URL to the Git repository to which push to the Hub. - organization (
str
, optional) — The name of the organization in which you would like to push your model. - kwargs (
Dict[str, Any]
, optional) — Not supported byMistralCommonTokenizer.save_pretrained
. Will raise an error if used.
Returns
A tuple of str
The files saved.
Save the full tokenizer state.
This method make sure the full tokenizer can then be re-loaded using the
~MistralCommonTokenizer.tokenization_mistral_common.from_pretrained
class method.
tokenize
< source >( text: str **kwargs ) → List[str]
Converts a string into a sequence of tokens, using the tokenizer.
Split in words for word-based vocabulary or sub-words for sub-word-based vocabularies.
truncate_sequences
< source >( ids: list pair_ids: None = None num_tokens_to_remove: int = 0 truncation_strategy: typing.Union[str, transformers.tokenization_utils_base.TruncationStrategy] = 'longest_first' stride: int = 0 **kwargs ) → Tuple[List[int], None, List[int]]
Parameters
- ids (
List[int]
) — Tokenized input ids. Can be obtained from a string by chaining thetokenize
andconvert_tokens_to_ids
methods. - pair_ids (
None
, optional) — Not supported byMistralCommonTokenizer
. Kept to match the signature ofPreTrainedTokenizerBase.truncate_sequences
. - num_tokens_to_remove (
int
, optional, defaults to 0) — Number of tokens to remove using the truncation strategy. - truncation_strategy (
str
or TruncationStrategy, optional, defaults to'longest_first'
) — The strategy to follow for truncation. Can be:'longest_first'
: Truncate to a maximum length specified with the argumentmax_length
or to the maximum acceptable input length for the model if that argument is not provided.'do_not_truncate'
(default): No truncation (i.e., can output batch with sequence lengths greater than the model maximum admissible input size).
- stride (
int
, optional, defaults to 0) — If set to a positive number, the overflowing tokens returned will contain some tokens from the main sequence returned. The value of this argument defines the number of additional tokens.
Returns
Tuple[List[int], None, List[int]]
The truncated ids
and the list of
overflowing tokens. None
is returned to match Transformers signature.
Truncates a sequence pair in-place following the strategy.
MistralModel
class transformers.MistralModel
< source >( config: MistralConfig )
Parameters
- config (MistralConfig) — 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 Mistral Model outputting raw hidden-states without any specific head on top.
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[transformers.cache_utils.Cache] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None use_cache: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None **kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs] ) → transformers.modeling_outputs.BaseModelOutputWithPast or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
torch.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- 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]
. - past_key_values (
~cache_utils.Cache
, 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
.Only Cache instance is allowed as input, see our kv cache guide. If no
past_key_values
are passed, DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_values
are used, the user is expected to input only unprocessedinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, unprocessed_length)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - 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
). - 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.BaseModelOutputWithPast or tuple(torch.FloatTensor)
A transformers.modeling_outputs.BaseModelOutputWithPast 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 (MistralConfig) 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 (
Cache
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — It is a Cache instance. For more details, see our kv cache guide.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.
The MistralModel 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.
MistralForCausalLM
class transformers.MistralForCausalLM
< source >( config )
Parameters
- config (MistralForCausalLM) — 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 Mistral Model for causal language modeling.
This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[transformers.cache_utils.Cache] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None logits_to_keep: typing.Union[int, torch.Tensor] = 0 **kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs] ) → transformers.modeling_outputs.CausalLMOutputWithPast or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
torch.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- 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]
. - past_key_values (
~cache_utils.Cache
, 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
.Only Cache instance is allowed as input, see our kv cache guide. If no
past_key_values
are passed, DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_values
are used, the user is expected to input only unprocessedinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, unprocessed_length)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - labels (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Labels for computing the masked language modeling loss. Indices should either be in[0, ..., config.vocab_size]
or -100 (seeinput_ids
docstring). Tokens with indices set to-100
are ignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size]
. - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - 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. - logits_to_keep (
Union[int, torch.Tensor]
, defaults to0
) — If anint
, compute logits for the lastlogits_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. If atorch.Tensor
, must be 1D corresponding to the indices to keep in the sequence length dimension. This is useful when using packed tensor format (single dimension for batch and sequence length).
Returns
transformers.modeling_outputs.CausalLMOutputWithPast or tuple(torch.FloatTensor)
A transformers.modeling_outputs.CausalLMOutputWithPast 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 (MistralConfig) 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). -
past_key_values (
Cache
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — It is a Cache instance. For more details, see our kv cache guide.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 MistralForCausalLM 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, MistralForCausalLM
>>> model = MistralForCausalLM.from_pretrained("meta-mistral/Mistral-2-7b-hf")
>>> tokenizer = AutoTokenizer.from_pretrained("meta-mistral/Mistral-2-7b-hf")
>>> prompt = "Hey, are you conscious? Can you talk to me?"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
MistralForSequenceClassification
class transformers.MistralForSequenceClassification
< source >( config )
Parameters
- config (MistralForSequenceClassification) — 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 Mistral Model transformer with a sequence classification head on top (linear layer).
MistralForSequenceClassification uses the last token in order to do the classification, as other causal models (e.g. GPT-2) 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, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[transformers.cache_utils.Cache] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None **kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs] ) → transformers.modeling_outputs.SequenceClassifierOutputWithPast
or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
torch.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- 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]
. - past_key_values (
~cache_utils.Cache
, 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
.Only Cache instance is allowed as input, see our kv cache guide. If no
past_key_values
are passed, DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_values
are used, the user is expected to input only unprocessedinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, unprocessed_length)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - 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). - 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
).
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 (MistralConfig) 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 (
Cache
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — It is a Cache instance. For more details, see our kv cache guide.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 MistralForSequenceClassification 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, MistralForSequenceClassification
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> model = MistralForSequenceClassification.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_class_id = logits.argmax().item()
>>> model.config.id2label[predicted_class_id]
...
>>> # 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 = MistralForSequenceClassification.from_pretrained("mistralai/Mistral-7B-v0.1", num_labels=num_labels)
>>> labels = torch.tensor([1])
>>> loss = model(**inputs, labels=labels).loss
>>> round(loss.item(), 2)
...
Example of multi-label classification:
>>> import torch
>>> from transformers import AutoTokenizer, MistralForSequenceClassification
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> model = MistralForSequenceClassification.from_pretrained("mistralai/Mistral-7B-v0.1", 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 = MistralForSequenceClassification.from_pretrained(
... "mistralai/Mistral-7B-v0.1", 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
MistralForTokenClassification
class transformers.MistralForTokenClassification
< source >( config )
Parameters
- config (MistralForTokenClassification) — 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 Mistral transformer 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, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[transformers.cache_utils.Cache] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None **kwargs ) → transformers.modeling_outputs.TokenClassifierOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
torch.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- 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]
. - past_key_values (
~cache_utils.Cache
, 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
.Only Cache instance is allowed as input, see our kv cache guide. If no
past_key_values
are passed, DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_values
are used, the user is expected to input only unprocessedinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, unprocessed_length)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - 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). - 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
).
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 (MistralConfig) 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 MistralForTokenClassification 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, MistralForTokenClassification
>>> import torch
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> model = MistralForTokenClassification.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> 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]]
>>> predicted_tokens_classes
...
>>> labels = predicted_token_class_ids
>>> loss = model(**inputs, labels=labels).loss
>>> round(loss.item(), 2)
...
MistralForQuestionAnswering
class transformers.MistralForQuestionAnswering
< source >( config )
Parameters
- config (MistralForQuestionAnswering) — 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 Mistral transformer with a span classification head on top for extractive question-answering tasks like
SQuAD (a linear layer 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, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward
< source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Union[transformers.cache_utils.Cache, list[torch.FloatTensor], NoneType] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None start_positions: typing.Optional[torch.LongTensor] = None end_positions: typing.Optional[torch.LongTensor] = None **kwargs ) → transformers.modeling_outputs.QuestionAnsweringModelOutput or tuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
torch.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
- 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]
. - past_key_values (
Union[~cache_utils.Cache, list[torch.FloatTensor], NoneType]
) — 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
.Only Cache instance is allowed as input, see our kv cache guide. If no
past_key_values
are passed, DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_values
are used, the user is expected to input only unprocessedinput_ids
(those that don’t have their past key value states given to this model) of shape(batch_size, unprocessed_length)
instead of allinput_ids
of shape(batch_size, sequence_length)
. - inputs_embeds (
torch.FloatTensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - 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.
Returns
transformers.modeling_outputs.QuestionAnsweringModelOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.QuestionAnsweringModelOutput 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 (MistralConfig) and inputs.
-
loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenlabels
is provided) — Total span extraction loss is the sum of a Cross-Entropy for the start and end positions. -
start_logits (
torch.FloatTensor
of shape(batch_size, sequence_length)
) — Span-start scores (before SoftMax). -
end_logits (
torch.FloatTensor
of shape(batch_size, sequence_length)
) — Span-end 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 MistralForQuestionAnswering 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, MistralForQuestionAnswering
>>> import torch
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> model = MistralForQuestionAnswering.from_pretrained("mistralai/Mistral-7B-v0.1")
>>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
>>> inputs = tokenizer(question, text, return_tensors="pt")
>>> with torch.no_grad():
... outputs = model(**inputs)
>>> answer_start_index = outputs.start_logits.argmax()
>>> answer_end_index = outputs.end_logits.argmax()
>>> predict_answer_tokens = inputs.input_ids[0, answer_start_index : answer_end_index + 1]
>>> tokenizer.decode(predict_answer_tokens, skip_special_tokens=True)
...
>>> # target is "nice puppet"
>>> target_start_index = torch.tensor([14])
>>> target_end_index = torch.tensor([15])
>>> outputs = model(**inputs, start_positions=target_start_index, end_positions=target_end_index)
>>> loss = outputs.loss
>>> round(loss.item(), 2)
...
- forward
FlaxMistralModel
class transformers.FlaxMistralModel
< source >( config: MistralConfig input_shape: tuple = (1, 1) seed: int = 0 dtype: dtype = <class 'jax.numpy.float32'> _do_init: bool = True **kwargs )
Parameters
- config (MistralConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
- dtype (
jax.numpy.dtype
, optional, defaults tojax.numpy.float32
) — The data type of the computation. Can be one ofjax.numpy.float32
,jax.numpy.float16
, orjax.numpy.bfloat16
.This can be used to enable mixed-precision training or half-precision inference on GPUs or TPUs. If specified all the computation will be performed with the given
dtype
.Note that this only specifies the dtype of the computation and does not influence the dtype of model parameters.
If you wish to change the dtype of the model parameters, see to_fp16() and to_bf16().
The bare Mistral Model transformer outputting raw hidden-states without any specific head on top.
This model inherits from FlaxPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a Flax Linen flax.nn.Module subclass. Use it as a regular Flax Module and refer to the Flax documentation for all matter related to general usage and behavior.
Finally, this model supports inherent JAX features such as:
__call__
< source >( input_ids attention_mask = None position_ids = None params: typing.Optional[dict] = None past_key_values: typing.Optional[dict] = None dropout_rng: <function PRNGKey at 0x7f34b2313b50> = None train: bool = False output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) → transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPast or tuple(torch.FloatTensor)
Parameters
- input_ids (
numpy.ndarray
of shape(batch_size, input_ids_length)
) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
numpy.ndarray
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If
past_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
).If you want to change padding behavior, you should read
modeling_opt._prepare_decoder_attention_mask
and modify to your needs. See diagram 1 in the paper for more information on the default strategy.- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- position_ids (
numpy.ndarray
of shape(batch_size, input_ids_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - past_key_values (
dict[str, np.ndarray]
, optional, returned byinit_cache
or when passing previouspast_key_values
) — Dictionary of pre-computed hidden-states (key and values in the attention blocks) that can be used for fast auto-regressive decoding. Pre-computed key and value hidden-states are of shape [batch_size, max_length]. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPast or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPast 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 (MistralConfig) and inputs.
-
last_hidden_state (
jnp.ndarray
of shape(batch_size, sequence_length, hidden_size)
) — Sequence of hidden-states at the output of the last layer of the model. -
past_key_values (
dict[str, jnp.ndarray]
) — Dictionary of pre-computed hidden-states (key and values in the attention blocks) that can be used for fast auto-regressive decoding. Pre-computed key and value hidden-states are of shape [batch_size, max_length]. -
hidden_states (
tuple(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
The FlaxMistralPreTrainedModel
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.
This example uses a random model as the real ones are all very big. To get proper results, you should use
mistralai/Mistral-7B-v0.1 instead of ksmcg/Mistral-tiny. If you get out-of-memory when loading that checkpoint, you can try
adding device_map="auto"
in the from_pretrained
call.
Example:
>>> from transformers import AutoTokenizer, FlaxMistralModel
>>> tokenizer = AutoTokenizer.from_pretrained("ksmcg/Mistral-tiny")
>>> model = FlaxMistralModel.from_pretrained("ksmcg/Mistral-tiny")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="jax")
>>> outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_state
FlaxMistralForCausalLM
class transformers.FlaxMistralForCausalLM
< source >( config: MistralConfig input_shape: tuple = (1, 1) seed: int = 0 dtype: dtype = <class 'jax.numpy.float32'> _do_init: bool = True **kwargs )
Parameters
- config (MistralConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
- dtype (
jax.numpy.dtype
, optional, defaults tojax.numpy.float32
) — The data type of the computation. Can be one ofjax.numpy.float32
,jax.numpy.float16
, orjax.numpy.bfloat16
.This can be used to enable mixed-precision training or half-precision inference on GPUs or TPUs. If specified all the computation will be performed with the given
dtype
.Note that this only specifies the dtype of the computation and does not influence the dtype of model parameters.
If you wish to change the dtype of the model parameters, see to_fp16() and to_bf16().
The Mistral Model transformer with a language modeling head (linear layer) on top.
This model inherits from FlaxPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a Flax Linen flax.nn.Module subclass. Use it as a regular Flax Module and refer to the Flax documentation for all matter related to general usage and behavior.
Finally, this model supports inherent JAX features such as:
__call__
< source >( input_ids attention_mask = None position_ids = None params: typing.Optional[dict] = None past_key_values: typing.Optional[dict] = None dropout_rng: <function PRNGKey at 0x7f34b2313b50> = None train: bool = False output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) → transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
Parameters
- input_ids (
numpy.ndarray
of shape(batch_size, input_ids_length)
) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
numpy.ndarray
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If
past_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
).If you want to change padding behavior, you should read
modeling_opt._prepare_decoder_attention_mask
and modify to your needs. See diagram 1 in the paper for more information on the default strategy.- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- position_ids (
numpy.ndarray
of shape(batch_size, input_ids_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - past_key_values (
dict[str, np.ndarray]
, optional, returned byinit_cache
or when passing previouspast_key_values
) — Dictionary of pre-computed hidden-states (key and values in the attention blocks) that can be used for fast auto-regressive decoding. Pre-computed key and value hidden-states are of shape [batch_size, max_length]. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions or a tuple of
torch.FloatTensor
(if return_dict=False
is passed or when config.return_dict=False
) comprising various
elements depending on the configuration (MistralConfig) and inputs.
-
logits (
jnp.ndarray
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(jnp.ndarray)
, optional, returned whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple ofjnp.ndarray
(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size)
.Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
-
cross_attentions (
tuple(jnp.ndarray)
, optional, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple ofjnp.ndarray
(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length)
.Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
-
past_key_values (
tuple(tuple(jnp.ndarray))
, optional, returned whenuse_cache=True
is passed or whenconfig.use_cache=True
) — Tuple ofjnp.ndarray
tuples of lengthconfig.n_layers
, with each tuple containing the cached key, value states of the self-attention and the cross-attention layers if model is used in encoder-decoder setting. Only relevant ifconfig.is_decoder = True
.Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see
past_key_values
input) to speed up sequential decoding.
The FlaxMistralPreTrainedModel
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.
This example uses a random model as the real ones are all very big. To get proper results, you should use
mistralai/Mistral-7B-v0.1 instead of ksmcg/Mistral-tiny. If you get out-of-memory when loading that checkpoint, you can try
adding device_map="auto"
in the from_pretrained
call.
Example:
>>> from transformers import AutoTokenizer, FlaxMistralForCausalLM
>>> tokenizer = AutoTokenizer.from_pretrained("ksmcg/Mistral-tiny")
>>> model = FlaxMistralForCausalLM.from_pretrained("ksmcg/Mistral-tiny")
>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="np")
>>> outputs = model(**inputs)
>>> # retrieve logts for next token
>>> next_token_logits = outputs.logits[:, -1]
TFMistralModel
class transformers.TFMistralModel
< source >( config: MistralConfig *inputs **kwargs )
Parameters
- config (MistralConfig) — 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 Mistral Model outputting raw hidden-states without any specific head on top.
This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.
TensorFlow models and layers in model
accept two formats as input:
- having all inputs as keyword arguments (like PyTorch models), or
- having all inputs as a list, tuple or dict in the first positional argument.
The reason the second format is supported is that Keras methods prefer this format when passing inputs to models
and layers. Because of this support, when using methods like model.fit()
things should “just work” for you - just
pass your inputs and labels in any format that model.fit()
supports! If, however, you want to use the second
format outside of Keras methods like fit()
and predict()
, such as when creating your own layers or models with
the Keras Functional
API, there are three possibilities you can use to gather all the input Tensors in the first
positional argument:
- a single Tensor with
input_ids
only and nothing else:model(input_ids)
- a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
model([input_ids, attention_mask])
ormodel([input_ids, attention_mask, token_type_ids])
- a dictionary with one or several input Tensors associated to the input names given in the docstring:
model({"input_ids": input_ids, "token_type_ids": token_type_ids})
Note that when creating models and layers with subclassing then you don’t need to worry about any of this, as you can just pass inputs like you would to any other Python function!
call
< source >( input_ids: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None attention_mask: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None position_ids: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None past_key_values: typing.Optional[list[tensorflow.python.framework.tensor.Tensor]] = None inputs_embeds: typing.Optional[tensorflow.python.framework.tensor.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 )
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If
past_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
).If you want to change padding behavior, you should read
modeling_opt._prepare_decoder_attention_mask
and modify to your needs. See diagram 1 in the paper for more information on the default strategy.- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - past_key_values (
Cache
ortuple(tuple(tf.Tensor))
, 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
.One formats is allowed:
- Tuple of
tuple(tf.Tensor)
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)
. - Tuple of
- inputs_embeds (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - use_cache (
bool
, optional) — 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.
The TFMistralModel 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.
TFMistralForCausalLM
call
< source >( input_ids: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None attention_mask: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None position_ids: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None past_key_values: typing.Optional[list[tensorflow.python.framework.tensor.Tensor]] = None inputs_embeds: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None labels: typing.Optional[tensorflow.python.framework.tensor.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 )
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If
past_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
).If you want to change padding behavior, you should read
modeling_opt._prepare_decoder_attention_mask
and modify to your needs. See diagram 1 in the paper for more information on the default strategy.- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - past_key_values (
Cache
ortuple(tuple(tf.Tensor))
, 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
.One formats is allowed:
- Tuple of
tuple(tf.Tensor)
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)
. - Tuple of
- inputs_embeds (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - labels (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Labels for computing the masked language modeling loss. Indices should either be in[0, ..., config.vocab_size]
or -100 (seeinput_ids
docstring). Tokens with indices set to-100
are ignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size]
.
The TFMistralForCausalLM 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.
TFMistralForSequenceClassification
class transformers.TFMistralForSequenceClassification
< source >( config *inputs **kwargs )
Parameters
- config (MistralConfig) — 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 Mistral Model transformer with a sequence classification head on top (linear layer).
MistralForSequenceClassification uses the last token in order to do the classification, as other causal models (e.g. GPT-2) 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 TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.
TensorFlow models and layers in model
accept two formats as input:
- having all inputs as keyword arguments (like PyTorch models), or
- having all inputs as a list, tuple or dict in the first positional argument.
The reason the second format is supported is that Keras methods prefer this format when passing inputs to models
and layers. Because of this support, when using methods like model.fit()
things should “just work” for you - just
pass your inputs and labels in any format that model.fit()
supports! If, however, you want to use the second
format outside of Keras methods like fit()
and predict()
, such as when creating your own layers or models with
the Keras Functional
API, there are three possibilities you can use to gather all the input Tensors in the first
positional argument:
- a single Tensor with
input_ids
only and nothing else:model(input_ids)
- a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
model([input_ids, attention_mask])
ormodel([input_ids, attention_mask, token_type_ids])
- a dictionary with one or several input Tensors associated to the input names given in the docstring:
model({"input_ids": input_ids, "token_type_ids": token_type_ids})
Note that when creating models and layers with subclassing then you don’t need to worry about any of this, as you can just pass inputs like you would to any other Python function!
call
< source >( input_ids: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None attention_mask: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None position_ids: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None past_key_values: typing.Optional[list[tensorflow.python.framework.tensor.Tensor]] = None inputs_embeds: typing.Optional[tensorflow.python.framework.tensor.Tensor] = None labels: typing.Optional[tensorflow.python.framework.tensor.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 )
Parameters
- input_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- attention_mask (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Mask to avoid performing attention on padding token indices. Mask values selected in[0, 1]
:- 1 for tokens that are not masked,
- 0 for tokens that are masked.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If
past_key_values
is used, optionally only the lastdecoder_input_ids
have to be input (seepast_key_values
).If you want to change padding behavior, you should read
modeling_opt._prepare_decoder_attention_mask
and modify to your needs. See diagram 1 in the paper for more information on the default strategy.- 1 indicates the head is not masked,
- 0 indicates the head is masked.
- position_ids (
torch.LongTensor
of shape(batch_size, sequence_length)
, optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]
. - past_key_values (
Cache
ortuple(tuple(tf.Tensor))
, 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
.One formats is allowed:
- Tuple of
tuple(tf.Tensor)
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)
. - Tuple of
- inputs_embeds (
tf.Tensor
of shape(batch_size, sequence_length, hidden_size)
, optional) — Optionally, instead of passinginput_ids
you can choose to directly pass an embedded representation. This is useful if you want more control over how to convertinput_ids
indices into associated vectors than the model’s internal embedding lookup matrix. - use_cache (
bool
, optional) — If set toTrue
,past_key_values
key value states are returned and can be used to speed up decoding (seepast_key_values
). - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_states
under returned tensors for more detail. - return_dict (
bool
, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - labels (
tf.Tensor
of shape(batch_size, sequence_length)
, optional) — Labels for computing the masked language modeling loss. Indices should either be in[0, ..., config.vocab_size]
or -100 (seeinput_ids
docstring). Tokens with indices set to-100
are ignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size]
.
The TFMistralForSequenceClassification 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.