# CANINE [CANINE](https://huggingface.co/papers/2103.06874) is a tokenization-free Transformer. It skips the usual step of splitting text into subwords or wordpieces and processes text character by character. That means it works directly with raw Unicode, making it especially useful for languages with complex or inconsistent tokenization rules and even noisy inputs like typos. Since working with characters means handling longer sequences, CANINE uses a smart trick. The model compresses the input early on (called downsampling) so the transformer doesn't have to process every character individually. This keeps things fast and efficient. You can find all the original CANINE checkpoints under the [Google](https://huggingface.co/google?search_models=canine) organization. > [!TIP] > Click on the CANINE models in the right sidebar for more examples of how to apply CANINE to different language tasks. The example below demonstrates how to generate embeddings with [Pipeline](/docs/transformers/v5.8.0/en/main_classes/pipelines#transformers.Pipeline), [AutoModel](/docs/transformers/v5.8.0/en/model_doc/auto#transformers.AutoModel), and from the command line. ```python from transformers import pipeline pipeline = pipeline( task="feature-extraction", model="google/canine-c", device=0, ) pipeline("Plant create energy through a process known as photosynthesis.") ``` ```python import torch from transformers import AutoModel model = AutoModel.from_pretrained("google/canine-c", device_map="auto") text = "Plant create energy through a process known as photosynthesis." input_ids = torch.tensor([[ord(char) for char in text]]) outputs = model(input_ids) pooled_output = outputs.pooler_output sequence_output = outputs.last_hidden_state ``` ## Notes - CANINE skips tokenization entirely — it works directly on raw characters, not subwords. You can use it with or without a tokenizer. For batched inference and training, it is recommended to use the tokenizer to pad and truncate all sequences to the same length. ```py from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer("google/canine-c") inputs = ["Life is like a box of chocolates.", "You never know what you gonna get."] encoding = tokenizer(inputs, padding="longest", truncation=True, return_tensors="pt").to(model.device) ``` - CANINE is primarily designed to be fine-tuned on a downstream task. The pretrained model can be used for either masked language modeling or next sentence prediction. ## CanineConfig[[transformers.CanineConfig]] #### transformers.CanineConfig[[transformers.CanineConfig]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/configuration_canine.py#L24) This is the configuration class to store the configuration of a CanineModel. It is used to instantiate a Canine 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 [google/canine-s](https://huggingface.co/google/canine-s) Configuration objects inherit from [PreTrainedConfig](/docs/transformers/v5.8.0/en/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the documentation from [PreTrainedConfig](/docs/transformers/v5.8.0/en/main_classes/configuration#transformers.PreTrainedConfig) for more information. Example: ```python >>> from transformers import CanineConfig, CanineModel >>> # Initializing a CANINE google/canine-s style configuration >>> configuration = CanineConfig() >>> # Initializing a model (with random weights) from the google/canine-s style configuration >>> model = CanineModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` **Parameters:** hidden_size (`int`, *optional*, defaults to `768`) : Dimension of the hidden representations. num_hidden_layers (`int`, *optional*, defaults to `12`) : Number of hidden layers in the Transformer decoder. num_attention_heads (`int`, *optional*, defaults to `12`) : Number of attention heads for each attention layer in the Transformer decoder. intermediate_size (`int`, *optional*, defaults to `3072`) : Dimension of the MLP representations. hidden_act (`str`, *optional*, defaults to `gelu`) : The non-linear activation function (function or string) in the decoder. For example, `"gelu"`, `"relu"`, `"silu"`, etc. hidden_dropout_prob (`Union[float, int]`, *optional*, defaults to `0.1`) : The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_probs_dropout_prob (`Union[float, int]`, *optional*, defaults to `0.1`) : The dropout ratio for the attention probabilities. max_position_embeddings (`int`, *optional*, defaults to `16384`) : The maximum sequence length that this model might ever be used with. type_vocab_size (`int`, *optional*, defaults to `16`) : The vocabulary size of the `token_type_ids`. initializer_range (`float`, *optional*, defaults to `0.02`) : The standard deviation of the truncated_normal_initializer for initializing all weight matrices. layer_norm_eps (`float`, *optional*, defaults to `1e-12`) : The epsilon used by the layer normalization layers. pad_token_id (`int`, *optional*, defaults to `0`) : Token id used for padding in the vocabulary. bos_token_id (`int`, *optional*, defaults to `57344`) : Token id used for beginning-of-stream in the vocabulary. eos_token_id (`Union[int, list[int]]`, *optional*, defaults to `57345`) : Token id used for end-of-stream in the vocabulary. downsampling_rate (`int`, *optional*, defaults to 4) : The rate at which to downsample the original character sequence length before applying the deep Transformer encoder. upsampling_kernel_size (`int`, *optional*, defaults to 4) : The kernel size (i.e. the number of characters in each window) of the convolutional projection layer when projecting back from `hidden_size`*2 to `hidden_size`. num_hash_functions (`int`, *optional*, defaults to 8) : The number of hash functions to use. Each hash function has its own embedding matrix. num_hash_buckets (`int`, *optional*, defaults to 16384) : The number of hash buckets to use. local_transformer_stride (`int`, *optional*, defaults to 128) : The stride of the local attention of the first shallow Transformer encoder. Defaults to 128 for good TPU/XLA memory alignment. ## CanineTokenizer[[transformers.CanineTokenizer]] #### transformers.CanineTokenizer[[transformers.CanineTokenizer]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/tokenization_canine.py#L55) Construct a CANINE tokenizer (i.e. a character splitter). It turns text into a sequence of characters, and then converts each character into its Unicode code point. [CanineTokenizer](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineTokenizer) inherits from [PreTrainedTokenizer](/docs/transformers/v5.8.0/en/main_classes/tokenizer#transformers.PythonBackend). Refer to superclass [PreTrainedTokenizer](/docs/transformers/v5.8.0/en/main_classes/tokenizer#transformers.PythonBackend) for usage examples and documentation concerning parameters. build_inputs_with_special_tokenstransformers.CanineTokenizer.build_inputs_with_special_tokenshttps://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/tokenization_python.py#L860[{"name": "token_ids_0", "val": ": list"}, {"name": "token_ids_1", "val": ": list[int] | None = None"}]- **token_ids_0** (`list[int]`) -- List of IDs to which the special tokens will be added. - **token_ids_1** (`list[int]`, *optional*) -- Optional second list of IDs for sequence pairs.0`list[int]`List of input IDs with the appropriate special tokens. Build model inputs from a sequence or a pair of sequences by adding special tokens. This method dynamically builds inputs based on the tokenizer's `special_tokens_pattern`: - `"none"`: No special tokens - `"cls_sep"`: [CLS] seq0 [SEP] or [CLS] seq0 [SEP] seq1 [SEP] - `"eos"`: seq0 [EOS] or seq0 [EOS] seq1 [EOS] - `"bos"`: [BOS] seq0 or [BOS] seq0 [BOS] seq1 - `"bos_eos"`: [BOS] seq0 [EOS] or [BOS] seq0 [EOS] seq1 [EOS] - `"cls_double_sep"`: [CLS] seq0 [SEP] or [CLS] seq0 [SEP] [SEP] seq1 [SEP] - `"prefix_suffix"`: ` seq0 [seq1] ` (custom prefix/suffix stored on the tokenizer) **Parameters:** model_max_length (`int`, *optional*, defaults to 2048) : The maximum sentence length the model accepts. **Returns:** ``list[int]`` List of input IDs with the appropriate special tokens. #### get_special_tokens_mask[[transformers.CanineTokenizer.get_special_tokens_mask]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/tokenization_python.py#L965) 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. This method dynamically builds the special tokens mask based on the tokenizer's `special_tokens_pattern`: - `"none"`: No special tokens (default, returns all 0s) - `"cls_sep"`: [CLS] seq0 [SEP] or [CLS] seq0 [SEP] seq1 [SEP] - `"eos"`: seq0 [EOS] or seq0 [EOS] seq1 [EOS] - `"bos"`: [BOS] seq0 or [BOS] seq0 [BOS] seq1 - `"bos_eos"`: [BOS] seq0 [EOS] or [BOS] seq0 [EOS] seq1 [EOS] - `"cls_double_sep"`: [CLS] seq0 [SEP] or [CLS] seq0 [SEP] [SEP] seq1 [SEP] - `"prefix_suffix"`: ` seq0 [seq1] ` **Parameters:** token_ids_0 (`list[int]`) : List of ids of the first sequence. token_ids_1 (`list[int]`, *optional*) : List of ids of the second sequence. already_has_special_tokens (`bool`, *optional*, defaults to `False`) : 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. #### create_token_type_ids_from_sequences[[transformers.CanineTokenizer.create_token_type_ids_from_sequences]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/tokenization_python.py#L1294) Create a mask from the two sequences passed to be used in a sequence-pair classification task. This method dynamically builds the token type IDs based on the tokenizer's configuration attributes: - `token_type_ids_pattern`: Pattern to use ("all_zeros" or "bert_style") - `token_type_ids_include_special_tokens`: Whether to account for special tokens in length calculation Examples: ```python # All zeros pattern (default, used by RoBERTa, BART, etc.) tokenizer.token_type_ids_pattern = "all_zeros" # Returns: [0, 0, 0, ...] for both sequences # BERT-style pattern (first sequence gets 0s, second gets 1s) tokenizer.token_type_ids_pattern = "bert_style" # Returns: [0, 0, 0, ..., 1, 1, 1, ...] for sequence pairs ``` **Parameters:** token_ids_0 (`list[int]`) : List of IDs. token_ids_1 (`list[int]`, *optional*) : Optional second list of IDs for sequence pairs. **Returns:** ``list[int]`` Token type IDs according to the configured pattern. ## CANINE specific outputs[[transformers.models.canine.modeling_canine.CanineModelOutputWithPooling]] #### transformers.models.canine.modeling_canine.CanineModelOutputWithPooling[[transformers.models.canine.modeling_canine.CanineModelOutputWithPooling]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L56) Output type of [CanineModel](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineModel). Based on [BaseModelOutputWithPooling](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling), but with slightly different `hidden_states` and `attentions`, as these also include the hidden states and attentions of the shallow Transformer encoders. **Parameters:** 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 (i.e. the output of the final shallow Transformer encoder). pooler_output (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) : Hidden-state of the first token of the sequence (classification token) at the last layer of the deep Transformer encoder, further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining. hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) : Tuple of `torch.FloatTensor` (one for the input to each encoder + one for the output of each layer of each encoder) of shape `(batch_size, sequence_length, hidden_size)` and `(batch_size, sequence_length // config.downsampling_rate, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial input to each Transformer encoder. The hidden states of the shallow encoders have length `sequence_length`, but the hidden states of the deep encoder have length `sequence_length` // `config.downsampling_rate`. attentions (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) : Tuple of `torch.FloatTensor` (one for each layer) of the 3 Transformer encoders of shape `(batch_size, num_heads, sequence_length, sequence_length)` and `(batch_size, num_heads, sequence_length // config.downsampling_rate, sequence_length // config.downsampling_rate)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. 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 (i.e. the output of the final shallow Transformer encoder). pooler_output (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) : Hidden-state of the first token of the sequence (classification token) at the last layer of the deep Transformer encoder, further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining. hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) : Tuple of `torch.FloatTensor` (one for the input to each encoder + one for the output of each layer of each encoder) of shape `(batch_size, sequence_length, hidden_size)` and `(batch_size, sequence_length // config.downsampling_rate, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial input to each Transformer encoder. The hidden states of the shallow encoders have length `sequence_length`, but the hidden states of the deep encoder have length `sequence_length` // `config.downsampling_rate`. attentions (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) : Tuple of `torch.FloatTensor` (one for each layer) of the 3 Transformer encoders of shape `(batch_size, num_heads, sequence_length, sequence_length)` and `(batch_size, num_heads, sequence_length // config.downsampling_rate, sequence_length // config.downsampling_rate)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## CanineModel[[transformers.CanineModel]] #### transformers.CanineModel[[transformers.CanineModel]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L728) The bare Canine Model outputting raw hidden-states without any specific head on top. This model inherits from [PreTrainedModel](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.CanineModel.forwardhttps://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L832[{"name": "input_ids", "val": ": torch.LongTensor | None = None"}, {"name": "attention_mask", "val": ": torch.FloatTensor | None = None"}, {"name": "token_type_ids", "val": ": torch.LongTensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "inputs_embeds", "val": ": torch.FloatTensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "**kwargs", "val": ""}]- **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](/docs/transformers/v5.8.0/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[CanineModelOutputWithPooling](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.models.canine.modeling_canine.CanineModelOutputWithPooling) or `tuple(torch.FloatTensor)`A [CanineModelOutputWithPooling](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.models.canine.modeling_canine.CanineModelOutputWithPooling) 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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. The [CanineModel](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineModel) 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. - **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 (i.e. the output of the final shallow Transformer encoder). - **pooler_output** (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) -- Hidden-state of the first token of the sequence (classification token) at the last layer of the deep Transformer encoder, further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining. - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the input to each encoder + one for the output of each layer of each encoder) of shape `(batch_size, sequence_length, hidden_size)` and `(batch_size, sequence_length // config.downsampling_rate, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial input to each Transformer encoder. The hidden states of the shallow encoders have length `sequence_length`, but the hidden states of the deep encoder have length `sequence_length` // `config.downsampling_rate`. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of the 3 Transformer encoders of shape `(batch_size, num_heads, sequence_length, sequence_length)` and `(batch_size, num_heads, sequence_length // config.downsampling_rate, sequence_length // config.downsampling_rate)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. **Parameters:** config ([CanineModel](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineModel)) : 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()](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. add_pooling_layer (`bool`, *optional*, defaults to `True`) : Whether to add a pooling layer **Returns:** `[CanineModelOutputWithPooling](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.models.canine.modeling_canine.CanineModelOutputWithPooling) or `tuple(torch.FloatTensor)`` A [CanineModelOutputWithPooling](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.models.canine.modeling_canine.CanineModelOutputWithPooling) 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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. ## CanineForSequenceClassification[[transformers.CanineForSequenceClassification]] #### transformers.CanineForSequenceClassification[[transformers.CanineForSequenceClassification]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L990) CANINE Model transformer with a sequence classification/regression head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks. This model inherits from [PreTrainedModel](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.CanineForSequenceClassification.forwardhttps://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1002[{"name": "input_ids", "val": ": torch.LongTensor | None = None"}, {"name": "attention_mask", "val": ": torch.FloatTensor | None = None"}, {"name": "token_type_ids", "val": ": torch.LongTensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "inputs_embeds", "val": ": torch.FloatTensor | None = None"}, {"name": "labels", "val": ": torch.LongTensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "**kwargs", "val": ""}]- **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](/docs/transformers/v5.8.0/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **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]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If `config.num_labels > 1` a classification loss is computed (Cross-Entropy). - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[SequenceClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or `tuple(torch.FloatTensor)`A [SequenceClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) 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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. The [CanineForSequenceClassification](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForSequenceClassification) 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. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` 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). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, 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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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. Example of single-label classification: ```python >>> import torch >>> from transformers import AutoTokenizer, CanineForSequenceClassification >>> tokenizer = AutoTokenizer.from_pretrained("google/canine-s") >>> model = CanineForSequenceClassification.from_pretrained("google/canine-s") >>> 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 = CanineForSequenceClassification.from_pretrained("google/canine-s", num_labels=num_labels) >>> labels = torch.tensor([1]) >>> loss = model(**inputs, labels=labels).loss >>> round(loss.item(), 2) ... ``` Example of multi-label classification: ```python >>> import torch >>> from transformers import AutoTokenizer, CanineForSequenceClassification >>> tokenizer = AutoTokenizer.from_pretrained("google/canine-s") >>> model = CanineForSequenceClassification.from_pretrained("google/canine-s", 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 = CanineForSequenceClassification.from_pretrained( ... "google/canine-s", 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 ``` **Parameters:** config ([CanineForSequenceClassification](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForSequenceClassification)) : 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()](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[SequenceClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or `tuple(torch.FloatTensor)`` A [SequenceClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) 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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. ## CanineForMultipleChoice[[transformers.CanineForMultipleChoice]] #### transformers.CanineForMultipleChoice[[transformers.CanineForMultipleChoice]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1075) The Canine Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a softmax) e.g. for RocStories/SWAG tasks. This model inherits from [PreTrainedModel](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.CanineForMultipleChoice.forwardhttps://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1086[{"name": "input_ids", "val": ": torch.LongTensor | None = None"}, {"name": "attention_mask", "val": ": torch.FloatTensor | None = None"}, {"name": "token_type_ids", "val": ": torch.LongTensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "inputs_embeds", "val": ": torch.FloatTensor | None = None"}, {"name": "labels", "val": ": torch.LongTensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "**kwargs", "val": ""}]- **input_ids** (`torch.LongTensor` of shape `(batch_size, num_choices, sequence_length)`) -- Indices of input sequence tokens in the vocabulary. Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.8.0/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.LongTensor` of shape `(batch_size, num_choices, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.LongTensor` of shape `(batch_size, num_choices, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.max_position_embeddings - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.FloatTensor` of shape `(batch_size, num_choices, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert *input_ids* indices into associated vectors than the model's internal embedding lookup matrix. - **labels** (`torch.LongTensor` of shape `(batch_size,)`, *optional*) -- Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See `input_ids` above) - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[MultipleChoiceModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) or `tuple(torch.FloatTensor)`A [MultipleChoiceModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) 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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. The [CanineForMultipleChoice](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForMultipleChoice) 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. - **loss** (`torch.FloatTensor` of shape *(1,)*, *optional*, returned when `labels` is provided) -- Classification loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, num_choices)`) -- *num_choices* is the second dimension of the input tensors. (see *input_ids* above). Classification scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, 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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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. Example: ```python >>> from transformers import AutoTokenizer, CanineForMultipleChoice >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("google/canine-s") >>> model = CanineForMultipleChoice.from_pretrained("google/canine-s") >>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced." >>> choice0 = "It is eaten with a fork and a knife." >>> choice1 = "It is eaten while held in the hand." >>> labels = torch.tensor(0).unsqueeze(0) # choice0 is correct (according to Wikipedia ;)), batch size 1 >>> encoding = tokenizer([prompt, prompt], [choice0, choice1], return_tensors="pt", padding=True) >>> outputs = model(**{k: v.unsqueeze(0) for k, v in encoding.items()}, labels=labels) # batch size is 1 >>> # the linear classifier still needs to be trained >>> loss = outputs.loss >>> logits = outputs.logits ``` **Parameters:** config ([CanineForMultipleChoice](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForMultipleChoice)) : 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()](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[MultipleChoiceModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) or `tuple(torch.FloatTensor)`` A [MultipleChoiceModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) 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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. ## CanineForTokenClassification[[transformers.CanineForTokenClassification]] #### transformers.CanineForTokenClassification[[transformers.CanineForTokenClassification]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1178) The Canine 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](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.CanineForTokenClassification.forwardhttps://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1190[{"name": "input_ids", "val": ": torch.LongTensor | None = None"}, {"name": "attention_mask", "val": ": torch.FloatTensor | None = None"}, {"name": "token_type_ids", "val": ": torch.LongTensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "inputs_embeds", "val": ": torch.FloatTensor | None = None"}, {"name": "labels", "val": ": torch.LongTensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "**kwargs", "val": ""}]- **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](/docs/transformers/v5.8.0/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **labels** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[TokenClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or `tuple(torch.FloatTensor)`A [TokenClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. The [CanineForTokenClassification](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForTokenClassification) 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. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, 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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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. Example: ```python >>> from transformers import AutoTokenizer, CanineForTokenClassification >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("google/canine-s") >>> model = CanineForTokenClassification.from_pretrained("google/canine-s") >>> 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 ``` ```python >>> labels = predicted_token_class_ids >>> loss = model(**inputs, labels=labels).loss >>> round(loss.item(), 2) ``` **Parameters:** config ([CanineForTokenClassification](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForTokenClassification)) : 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()](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[TokenClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or `tuple(torch.FloatTensor)`` A [TokenClassifierOutput](/docs/transformers/v5.8.0/en/main_classes/output#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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. ## CanineForQuestionAnswering[[transformers.CanineForQuestionAnswering]] #### transformers.CanineForQuestionAnswering[[transformers.CanineForQuestionAnswering]] [Source](https://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1274) The Canine 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](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.CanineForQuestionAnswering.forwardhttps://github.com/huggingface/transformers/blob/v5.8.0/src/transformers/models/canine/modeling_canine.py#L1285[{"name": "input_ids", "val": ": torch.LongTensor | None = None"}, {"name": "attention_mask", "val": ": torch.FloatTensor | None = None"}, {"name": "token_type_ids", "val": ": torch.LongTensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "inputs_embeds", "val": ": torch.FloatTensor | None = None"}, {"name": "start_positions", "val": ": torch.LongTensor | None = None"}, {"name": "end_positions", "val": ": torch.LongTensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "**kwargs", "val": ""}]- **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](/docs/transformers/v5.8.0/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.8.0/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **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. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[QuestionAnsweringModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`A [QuestionAnsweringModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs. The [CanineForQuestionAnswering](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForQuestionAnswering) 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. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, 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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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. Example: ```python >>> from transformers import AutoTokenizer, CanineForQuestionAnswering >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("google/canine-s") >>> model = CanineForQuestionAnswering.from_pretrained("google/canine-s") >>> 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) ... ``` **Parameters:** config ([CanineForQuestionAnswering](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineForQuestionAnswering)) : 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()](/docs/transformers/v5.8.0/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[QuestionAnsweringModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`` A [QuestionAnsweringModelOutput](/docs/transformers/v5.8.0/en/main_classes/output#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 ([CanineConfig](/docs/transformers/v5.8.0/en/model_doc/canine#transformers.CanineConfig)) and inputs.