Skip to content

layoutlm

mindnlp.transformers.models.layoutlm.modeling_layoutlm

MindSpore LayoutLM model.

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMAttention

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertAttention with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
class LayoutLMAttention(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertAttention with Bert->LayoutLM"""
    def __init__(self, config, position_embedding_type=None):
        """
        Initializes an instance of the LayoutLMAttention class.

        Args:
            self: The instance of the class (automatically passed).
            config:
                An object containing the configuration settings.

                - Type: object
                - Purpose: Provides the configuration settings for the LayoutLMAttention class.
                - Restrictions: None

            position_embedding_type:
                The type of position embedding to use.

                - Type: Any
                - Purpose: Specifies the type of position embedding to be used in the LayoutLMAttention class.
                - Restrictions: None

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__()
        self.self = LayoutLMSelfAttention(config, position_embedding_type=position_embedding_type)
        self.output = LayoutLMSelfOutput(config)
        self.pruned_heads = set()

    def prune_heads(self, heads):
        """
        Prunes the attention heads in the LayoutLMAttention class.

        Args:
            self: The LayoutLMAttention instance.
            heads (list): A list of integers representing the attention heads to be pruned.

        Returns:
            None: The method modifies the LayoutLMAttention instance in-place.

        Raises:
            None.

        This method prunes the specified attention heads from the LayoutLMAttention instance.
        First, it checks if the 'heads' list is empty. If so, the method returns without making any changes.
        Otherwise, it calls the 'find_pruneable_heads_and_indices' function to identify the attention heads and
        their corresponding indices that can be pruned based on the given 'heads' list, the number of attention heads,
        attention head size, and already pruned heads stored in the instance.
        Next, it prunes the 'self.query', 'self.key', 'self.value', and 'self.output.dense' linear layers by calling
        the 'prune_linear_layer' function with the identified indices.
        After each linear layer is pruned, the number of attention heads is updated by subtracting the length of the
        'heads' list from the current number of attention heads.
        The total size of all attention heads, 'self.all_head_size', is then recalculated as the product of the
        attention head size and the updated number of attention heads.
        Finally, the 'pruned_heads' set is updated by adding the attention heads specified in the 'heads' list.
        The method does not return any value but modifies the LayoutLMAttention instance by pruning the specified
        attention heads.
        """
        if len(heads) == 0:
            return
        heads, index = find_pruneable_heads_and_indices(
            heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
        )

        # Prune linear layers
        self.self.query = prune_linear_layer(self.self.query, index)
        self.self.key = prune_linear_layer(self.self.key, index)
        self.self.value = prune_linear_layer(self.self.value, index)
        self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)

        # Update hyper params and store pruned heads
        self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
        self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
        self.pruned_heads = self.pruned_heads.union(heads)

    def forward(
        self,
        hidden_states: mindspore.Tensor,
        attention_mask: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        encoder_hidden_states: Optional[mindspore.Tensor] = None,
        encoder_attention_mask: Optional[mindspore.Tensor] = None,
        past_key_value: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
        output_attentions: Optional[bool] = False,
    ) -> Tuple[mindspore.Tensor]:
        '''
        This method forwards the LayoutLMAttention.

        Args:
            self (LayoutLMAttention): The instance of the LayoutLMAttention class.
            hidden_states (mindspore.Tensor): The input hidden states for the attention mechanism.
            attention_mask (Optional[mindspore.Tensor]): An optional mask for the attention mechanism. Default is None.
            head_mask (Optional[mindspore.Tensor]): An optional mask for the attention heads. Default is None.
            encoder_hidden_states (Optional[mindspore.Tensor]): An optional input for encoder hidden states.
                Default is None.
            encoder_attention_mask (Optional[mindspore.Tensor]): An optional mask for encoder attention.
                Default is None.
            past_key_value (Optional[Tuple[Tuple[mindspore.Tensor]]]): An optional input for past key value.
                Default is None.
            output_attentions (Optional[bool]): A flag to indicate whether to output attentions. Default is False.

        Returns:
            Tuple[mindspore.Tensor]: A tuple containing the attention output and other optional outputs.

        Raises:
            None
        '''
        self_outputs = self.self(
            hidden_states,
            attention_mask,
            head_mask,
            encoder_hidden_states,
            encoder_attention_mask,
            past_key_value,
            output_attentions,
        )
        attention_output = self.output(self_outputs[0], hidden_states)
        outputs = (attention_output,) + self_outputs[1:]  # add attentions if we output them
        return outputs

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMAttention.__init__(config, position_embedding_type=None)

Initializes an instance of the LayoutLMAttention class.

PARAMETER DESCRIPTION
self

The instance of the class (automatically passed).
config

An object containing the configuration settings.

  • Type: object
  • Purpose: Provides the configuration settings for the LayoutLMAttention class.
  • Restrictions: None
position_embedding_type

The type of position embedding to use.

  • Type: Any
  • Purpose: Specifies the type of position embedding to be used in the LayoutLMAttention class.
  • Restrictions: None

DEFAULT: None

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
def __init__(self, config, position_embedding_type=None):
    """
    Initializes an instance of the LayoutLMAttention class.

    Args:
        self: The instance of the class (automatically passed).
        config:
            An object containing the configuration settings.

            - Type: object
            - Purpose: Provides the configuration settings for the LayoutLMAttention class.
            - Restrictions: None

        position_embedding_type:
            The type of position embedding to use.

            - Type: Any
            - Purpose: Specifies the type of position embedding to be used in the LayoutLMAttention class.
            - Restrictions: None

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__()
    self.self = LayoutLMSelfAttention(config, position_embedding_type=position_embedding_type)
    self.output = LayoutLMSelfOutput(config)
    self.pruned_heads = set()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMAttention.forward(hidden_states, attention_mask=None, head_mask=None, encoder_hidden_states=None, encoder_attention_mask=None, past_key_value=None, output_attentions=False)

This method forwards the LayoutLMAttention.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMAttention class.

TYPE: LayoutLMAttention

hidden_states

The input hidden states for the attention mechanism.

TYPE: Tensor

attention_mask

An optional mask for the attention mechanism. Default is None.

TYPE: Optional[Tensor] DEFAULT: None

head_mask

An optional mask for the attention heads. Default is None.

TYPE: Optional[Tensor] DEFAULT: None

encoder_hidden_states

An optional input for encoder hidden states. Default is None.

TYPE: Optional[Tensor] DEFAULT: None

encoder_attention_mask

An optional mask for encoder attention. Default is None.

TYPE: Optional[Tensor] DEFAULT: None

past_key_value

An optional input for past key value. Default is None.

TYPE: Optional[Tuple[Tuple[Tensor]]] DEFAULT: None

output_attentions

A flag to indicate whether to output attentions. Default is False.

TYPE: Optional[bool] DEFAULT: False

RETURNS DESCRIPTION
Tuple[Tensor]

Tuple[mindspore.Tensor]: A tuple containing the attention output and other optional outputs.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
def forward(
    self,
    hidden_states: mindspore.Tensor,
    attention_mask: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    encoder_hidden_states: Optional[mindspore.Tensor] = None,
    encoder_attention_mask: Optional[mindspore.Tensor] = None,
    past_key_value: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
    output_attentions: Optional[bool] = False,
) -> Tuple[mindspore.Tensor]:
    '''
    This method forwards the LayoutLMAttention.

    Args:
        self (LayoutLMAttention): The instance of the LayoutLMAttention class.
        hidden_states (mindspore.Tensor): The input hidden states for the attention mechanism.
        attention_mask (Optional[mindspore.Tensor]): An optional mask for the attention mechanism. Default is None.
        head_mask (Optional[mindspore.Tensor]): An optional mask for the attention heads. Default is None.
        encoder_hidden_states (Optional[mindspore.Tensor]): An optional input for encoder hidden states.
            Default is None.
        encoder_attention_mask (Optional[mindspore.Tensor]): An optional mask for encoder attention.
            Default is None.
        past_key_value (Optional[Tuple[Tuple[mindspore.Tensor]]]): An optional input for past key value.
            Default is None.
        output_attentions (Optional[bool]): A flag to indicate whether to output attentions. Default is False.

    Returns:
        Tuple[mindspore.Tensor]: A tuple containing the attention output and other optional outputs.

    Raises:
        None
    '''
    self_outputs = self.self(
        hidden_states,
        attention_mask,
        head_mask,
        encoder_hidden_states,
        encoder_attention_mask,
        past_key_value,
        output_attentions,
    )
    attention_output = self.output(self_outputs[0], hidden_states)
    outputs = (attention_output,) + self_outputs[1:]  # add attentions if we output them
    return outputs

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMAttention.prune_heads(heads)

Prunes the attention heads in the LayoutLMAttention class.

PARAMETER DESCRIPTION
self

The LayoutLMAttention instance.
heads

A list of integers representing the attention heads to be pruned.

TYPE: list

RETURNS DESCRIPTION
None

The method modifies the LayoutLMAttention instance in-place.

This method prunes the specified attention heads from the LayoutLMAttention instance. First, it checks if the 'heads' list is empty. If so, the method returns without making any changes. Otherwise, it calls the 'find_pruneable_heads_and_indices' function to identify the attention heads and their corresponding indices that can be pruned based on the given 'heads' list, the number of attention heads, attention head size, and already pruned heads stored in the instance. Next, it prunes the 'self.query', 'self.key', 'self.value', and 'self.output.dense' linear layers by calling the 'prune_linear_layer' function with the identified indices. After each linear layer is pruned, the number of attention heads is updated by subtracting the length of the 'heads' list from the current number of attention heads. The total size of all attention heads, 'self.all_head_size', is then recalculated as the product of the attention head size and the updated number of attention heads. Finally, the 'pruned_heads' set is updated by adding the attention heads specified in the 'heads' list. The method does not return any value but modifies the LayoutLMAttention instance by pruning the specified attention heads.

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
def prune_heads(self, heads):
    """
    Prunes the attention heads in the LayoutLMAttention class.

    Args:
        self: The LayoutLMAttention instance.
        heads (list): A list of integers representing the attention heads to be pruned.

    Returns:
        None: The method modifies the LayoutLMAttention instance in-place.

    Raises:
        None.

    This method prunes the specified attention heads from the LayoutLMAttention instance.
    First, it checks if the 'heads' list is empty. If so, the method returns without making any changes.
    Otherwise, it calls the 'find_pruneable_heads_and_indices' function to identify the attention heads and
    their corresponding indices that can be pruned based on the given 'heads' list, the number of attention heads,
    attention head size, and already pruned heads stored in the instance.
    Next, it prunes the 'self.query', 'self.key', 'self.value', and 'self.output.dense' linear layers by calling
    the 'prune_linear_layer' function with the identified indices.
    After each linear layer is pruned, the number of attention heads is updated by subtracting the length of the
    'heads' list from the current number of attention heads.
    The total size of all attention heads, 'self.all_head_size', is then recalculated as the product of the
    attention head size and the updated number of attention heads.
    Finally, the 'pruned_heads' set is updated by adding the attention heads specified in the 'heads' list.
    The method does not return any value but modifies the LayoutLMAttention instance by pruning the specified
    attention heads.
    """
    if len(heads) == 0:
        return
    heads, index = find_pruneable_heads_and_indices(
        heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
    )

    # Prune linear layers
    self.self.query = prune_linear_layer(self.self.query, index)
    self.self.key = prune_linear_layer(self.self.key, index)
    self.self.value = prune_linear_layer(self.self.value, index)
    self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)

    # Update hyper params and store pruned heads
    self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
    self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
    self.pruned_heads = self.pruned_heads.union(heads)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMEmbeddings

Bases: Module

forward the embeddings from word, position and token_type embeddings.

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
class LayoutLMEmbeddings(nn.Module):
    """forward the embeddings from word, position and token_type embeddings."""
    def __init__(self, config):
        """
        Args:
            self (LayoutLMEmbeddings): The instance of the LayoutLMEmbeddings class.
            config (object): An object containing configuration parameters,
                including vocab_size, hidden_size, max_position_embeddings,
                max_2d_position_embeddings, type_vocab_size, pad_token_id, layer_norm_eps, and hidden_dropout_prob. 

        Returns:
            None.

        Raises:
            None
        """
        super(LayoutLMEmbeddings, self).__init__()
        self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
        self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.hidden_size)
        self.x_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
        self.y_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
        self.h_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
        self.w_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
        self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)

        self.LayerNorm = LayoutLMLayerNorm([config.hidden_size], eps=config.layer_norm_eps)
        self.dropout = nn.Dropout(p=config.hidden_dropout_prob)

        self.position_ids = ops.arange(config.max_position_embeddings).broadcast_to((1, -1))

    def forward(
        self,
        input_ids=None,
        bbox=None,
        token_type_ids=None,
        position_ids=None,
        inputs_embeds=None,
    ):
        """
        Constructs the LayoutLM embeddings.

        Args:
            self (LayoutLMEmbeddings): An instance of the LayoutLMEmbeddings class.
            input_ids (Tensor, optional): The input tensor of token indices. Defaults to None.
            bbox (Tensor, optional): The bounding box tensor. Defaults to None.
            token_type_ids (Tensor, optional): The input tensor of token type indices. Defaults to None.
            position_ids (Tensor, optional): The input tensor of position indices. Defaults to None.
            inputs_embeds (Tensor, optional): The input tensor of embeddings. Defaults to None.

        Returns:
            Tensor: The computed embeddings.

        Raises:
            IndexError: If the `bbox` coordinate values are not within the 0-1000 range.

        Note:
            The method calculates the embeddings by adding various embeddings such as words, position, 
            bounding box, token type, etc. It also performs layer normalization and dropout on the embeddings.
        """
        if input_ids is not None:
            input_shape = input_ids.shape
        else:
            input_shape = inputs_embeds.shape[:-1]

        seq_length = input_shape[1]

        if position_ids is None:
            position_ids = self.position_ids[:, :seq_length]

        if token_type_ids is None:
            token_type_ids = ops.zeros(input_shape, dtype=mindspore.int64)

        if inputs_embeds is None:
            inputs_embeds = self.word_embeddings(input_ids)

        words_embeddings = inputs_embeds
        position_embeddings = self.position_embeddings(position_ids)

        try:
            left_position_embeddings = self.x_position_embeddings(bbox[:, :, 0])
            upper_position_embeddings = self.y_position_embeddings(bbox[:, :, 1])
            right_position_embeddings = self.x_position_embeddings(bbox[:, :, 2])
            lower_position_embeddings = self.y_position_embeddings(bbox[:, :, 3])
        except IndexError as e:
            raise IndexError("The `bbox`coordinate values should be within 0-1000 range.") from e

        h_position_embeddings = self.h_position_embeddings(bbox[:, :, 3] - bbox[:, :, 1])
        w_position_embeddings = self.w_position_embeddings(bbox[:, :, 2] - bbox[:, :, 0])
        token_type_embeddings = self.token_type_embeddings(token_type_ids)

        embeddings = (
            words_embeddings
            + position_embeddings
            + left_position_embeddings
            + upper_position_embeddings
            + right_position_embeddings
            + lower_position_embeddings
            + h_position_embeddings
            + w_position_embeddings
            + token_type_embeddings
        )
        embeddings = self.LayerNorm(embeddings)
        embeddings = self.dropout(embeddings)
        return embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMEmbeddings.__init__(config)

PARAMETER DESCRIPTION
self

The instance of the LayoutLMEmbeddings class.

TYPE: LayoutLMEmbeddings

config

An object containing configuration parameters, including vocab_size, hidden_size, max_position_embeddings, max_2d_position_embeddings, type_vocab_size, pad_token_id, layer_norm_eps, and hidden_dropout_prob.

TYPE: object

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
def __init__(self, config):
    """
    Args:
        self (LayoutLMEmbeddings): The instance of the LayoutLMEmbeddings class.
        config (object): An object containing configuration parameters,
            including vocab_size, hidden_size, max_position_embeddings,
            max_2d_position_embeddings, type_vocab_size, pad_token_id, layer_norm_eps, and hidden_dropout_prob. 

    Returns:
        None.

    Raises:
        None
    """
    super(LayoutLMEmbeddings, self).__init__()
    self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
    self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.hidden_size)
    self.x_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
    self.y_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
    self.h_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
    self.w_position_embeddings = nn.Embedding(config.max_2d_position_embeddings, config.hidden_size)
    self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)

    self.LayerNorm = LayoutLMLayerNorm([config.hidden_size], eps=config.layer_norm_eps)
    self.dropout = nn.Dropout(p=config.hidden_dropout_prob)

    self.position_ids = ops.arange(config.max_position_embeddings).broadcast_to((1, -1))

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMEmbeddings.forward(input_ids=None, bbox=None, token_type_ids=None, position_ids=None, inputs_embeds=None)

Constructs the LayoutLM embeddings.

PARAMETER DESCRIPTION
self

An instance of the LayoutLMEmbeddings class.

TYPE: LayoutLMEmbeddings

input_ids

The input tensor of token indices. Defaults to None.

TYPE: Tensor DEFAULT: None

bbox

The bounding box tensor. Defaults to None.

TYPE: Tensor DEFAULT: None

token_type_ids

The input tensor of token type indices. Defaults to None.

TYPE: Tensor DEFAULT: None

position_ids

The input tensor of position indices. Defaults to None.

TYPE: Tensor DEFAULT: None

inputs_embeds

The input tensor of embeddings. Defaults to None.

TYPE: Tensor DEFAULT: None

RETURNS DESCRIPTION
Tensor

The computed embeddings.
RAISES DESCRIPTION
IndexError

If the bbox coordinate values are not within the 0-1000 range.
Note

The method calculates the embeddings by adding various embeddings such as words, position, bounding box, token type, etc. It also performs layer normalization and dropout on the embeddings.

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
def forward(
    self,
    input_ids=None,
    bbox=None,
    token_type_ids=None,
    position_ids=None,
    inputs_embeds=None,
):
    """
    Constructs the LayoutLM embeddings.

    Args:
        self (LayoutLMEmbeddings): An instance of the LayoutLMEmbeddings class.
        input_ids (Tensor, optional): The input tensor of token indices. Defaults to None.
        bbox (Tensor, optional): The bounding box tensor. Defaults to None.
        token_type_ids (Tensor, optional): The input tensor of token type indices. Defaults to None.
        position_ids (Tensor, optional): The input tensor of position indices. Defaults to None.
        inputs_embeds (Tensor, optional): The input tensor of embeddings. Defaults to None.

    Returns:
        Tensor: The computed embeddings.

    Raises:
        IndexError: If the `bbox` coordinate values are not within the 0-1000 range.

    Note:
        The method calculates the embeddings by adding various embeddings such as words, position, 
        bounding box, token type, etc. It also performs layer normalization and dropout on the embeddings.
    """
    if input_ids is not None:
        input_shape = input_ids.shape
    else:
        input_shape = inputs_embeds.shape[:-1]

    seq_length = input_shape[1]

    if position_ids is None:
        position_ids = self.position_ids[:, :seq_length]

    if token_type_ids is None:
        token_type_ids = ops.zeros(input_shape, dtype=mindspore.int64)

    if inputs_embeds is None:
        inputs_embeds = self.word_embeddings(input_ids)

    words_embeddings = inputs_embeds
    position_embeddings = self.position_embeddings(position_ids)

    try:
        left_position_embeddings = self.x_position_embeddings(bbox[:, :, 0])
        upper_position_embeddings = self.y_position_embeddings(bbox[:, :, 1])
        right_position_embeddings = self.x_position_embeddings(bbox[:, :, 2])
        lower_position_embeddings = self.y_position_embeddings(bbox[:, :, 3])
    except IndexError as e:
        raise IndexError("The `bbox`coordinate values should be within 0-1000 range.") from e

    h_position_embeddings = self.h_position_embeddings(bbox[:, :, 3] - bbox[:, :, 1])
    w_position_embeddings = self.w_position_embeddings(bbox[:, :, 2] - bbox[:, :, 0])
    token_type_embeddings = self.token_type_embeddings(token_type_ids)

    embeddings = (
        words_embeddings
        + position_embeddings
        + left_position_embeddings
        + upper_position_embeddings
        + right_position_embeddings
        + lower_position_embeddings
        + h_position_embeddings
        + w_position_embeddings
        + token_type_embeddings
    )
    embeddings = self.LayerNorm(embeddings)
    embeddings = self.dropout(embeddings)
    return embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMEncoder

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertEncoder with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
class LayoutLMEncoder(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertEncoder with Bert->LayoutLM"""
    def __init__(self, config):
        """Initializes an instance of the LayoutLMEncoder class.

        Args:
            self (LayoutLMEncoder): The instance of the LayoutLMEncoder class.
            config (object): The configuration object containing the necessary parameters for the LayoutLMEncoder.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__()
        self.config = config
        self.layer = nn.ModuleList([LayoutLMLayer(config) for _ in range(config.num_hidden_layers)])
        self.gradient_checkpointing = False

    def forward(
        self,
        hidden_states: mindspore.Tensor,
        attention_mask: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        encoder_hidden_states: Optional[mindspore.Tensor] = None,
        encoder_attention_mask: Optional[mindspore.Tensor] = None,
        past_key_values: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = False,
        output_hidden_states: Optional[bool] = False,
        return_dict: Optional[bool] = True,
    ) -> Union[Tuple[mindspore.Tensor], BaseModelOutputWithPastAndCrossAttentions]:
        """
        This method forwards the LayoutLM encoder using the specified parameters.

        Args:
            self: The instance of the LayoutLMEncoder class.
            hidden_states (mindspore.Tensor): The input hidden states for encoding.
            attention_mask (Optional[mindspore.Tensor]): Mask to avoid attending to certain positions.
            head_mask (Optional[mindspore.Tensor]): Mask to specify which heads to disable in the attention computation.
            encoder_hidden_states (Optional[mindspore.Tensor]): Hidden states of the encoder to be used in
                cross-attention layers.
            encoder_attention_mask (Optional[mindspore.Tensor]): Mask for encoder attention mechanism.
            past_key_values (Optional[Tuple[Tuple[mindspore.Tensor]]]): Cached key/values for previous decoding steps.
            use_cache (Optional[bool]): Flag to indicate whether to use caching for decoding.
            output_attentions (Optional[bool]): Flag to output attention scores.
            output_hidden_states (Optional[bool]): Flag to output hidden states for each layer.
            return_dict (Optional[bool]): Flag to indicate returning the output as a dictionary.

        Returns:
            Union[Tuple[mindspore.Tensor], BaseModelOutputWithPastAndCrossAttentions]: The output of the encoder,
                which is either a tuple of hidden states or a complex object containing past key values and attentions.

        Raises:
            Warning: If `use_cache=True` is incompatible with gradient checkpointing,
                it will issue a warning and set `use_cache=False`.
        """
        all_hidden_states = () if output_hidden_states else None
        all_self_attentions = () if output_attentions else None
        all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None

        if self.gradient_checkpointing and self.training:
            if use_cache:
                logger.warning_once(
                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
                )
                use_cache = False

        next_decoder_cache = () if use_cache else None
        for i, layer_module in enumerate(self.layer):
            if output_hidden_states:
                all_hidden_states = all_hidden_states + (hidden_states,)

            layer_head_mask = head_mask[i] if head_mask is not None else None
            past_key_value = past_key_values[i] if past_key_values is not None else None

            layer_outputs = layer_module(
                hidden_states,
                attention_mask,
                layer_head_mask,
                encoder_hidden_states,
                encoder_attention_mask,
                past_key_value,
                output_attentions,
            )

            hidden_states = layer_outputs[0]
            if use_cache:
                next_decoder_cache += (layer_outputs[-1],)
            if output_attentions:
                all_self_attentions = all_self_attentions + (layer_outputs[1],)
                if self.config.add_cross_attention:
                    all_cross_attentions = all_cross_attentions + (layer_outputs[2],)

        if output_hidden_states:
            all_hidden_states = all_hidden_states + (hidden_states,)

        if not return_dict:
            return tuple(
                v
                for v in [
                    hidden_states,
                    next_decoder_cache,
                    all_hidden_states,
                    all_self_attentions,
                    all_cross_attentions,
                ]
                if v is not None
            )
        return BaseModelOutputWithPastAndCrossAttentions(
            last_hidden_state=hidden_states,
            past_key_values=next_decoder_cache,
            hidden_states=all_hidden_states,
            attentions=all_self_attentions,
            cross_attentions=all_cross_attentions,
        )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMEncoder.__init__(config)

Initializes an instance of the LayoutLMEncoder class.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMEncoder class.

TYPE: LayoutLMEncoder

config

The configuration object containing the necessary parameters for the LayoutLMEncoder.

TYPE: object

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
def __init__(self, config):
    """Initializes an instance of the LayoutLMEncoder class.

    Args:
        self (LayoutLMEncoder): The instance of the LayoutLMEncoder class.
        config (object): The configuration object containing the necessary parameters for the LayoutLMEncoder.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__()
    self.config = config
    self.layer = nn.ModuleList([LayoutLMLayer(config) for _ in range(config.num_hidden_layers)])
    self.gradient_checkpointing = False

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMEncoder.forward(hidden_states, attention_mask=None, head_mask=None, encoder_hidden_states=None, encoder_attention_mask=None, past_key_values=None, use_cache=None, output_attentions=False, output_hidden_states=False, return_dict=True)

This method forwards the LayoutLM encoder using the specified parameters.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMEncoder class.
hidden_states

The input hidden states for encoding.

TYPE: Tensor

attention_mask

Mask to avoid attending to certain positions.

TYPE: Optional[Tensor] DEFAULT: None

head_mask

Mask to specify which heads to disable in the attention computation.

TYPE: Optional[Tensor] DEFAULT: None

encoder_hidden_states

Hidden states of the encoder to be used in cross-attention layers.

TYPE: Optional[Tensor] DEFAULT: None

encoder_attention_mask

Mask for encoder attention mechanism.

TYPE: Optional[Tensor] DEFAULT: None

past_key_values

Cached key/values for previous decoding steps.

TYPE: Optional[Tuple[Tuple[Tensor]]] DEFAULT: None

use_cache

Flag to indicate whether to use caching for decoding.

TYPE: Optional[bool] DEFAULT: None

output_attentions

Flag to output attention scores.

TYPE: Optional[bool] DEFAULT: False

output_hidden_states

Flag to output hidden states for each layer.

TYPE: Optional[bool] DEFAULT: False

return_dict

Flag to indicate returning the output as a dictionary.

TYPE: Optional[bool] DEFAULT: True

RETURNS DESCRIPTION
Union[Tuple[Tensor], BaseModelOutputWithPastAndCrossAttentions]

Union[Tuple[mindspore.Tensor], BaseModelOutputWithPastAndCrossAttentions]: The output of the encoder, which is either a tuple of hidden states or a complex object containing past key values and attentions.
RAISES DESCRIPTION
Warning

If use_cache=True is incompatible with gradient checkpointing, it will issue a warning and set use_cache=False.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
def forward(
    self,
    hidden_states: mindspore.Tensor,
    attention_mask: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    encoder_hidden_states: Optional[mindspore.Tensor] = None,
    encoder_attention_mask: Optional[mindspore.Tensor] = None,
    past_key_values: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
    use_cache: Optional[bool] = None,
    output_attentions: Optional[bool] = False,
    output_hidden_states: Optional[bool] = False,
    return_dict: Optional[bool] = True,
) -> Union[Tuple[mindspore.Tensor], BaseModelOutputWithPastAndCrossAttentions]:
    """
    This method forwards the LayoutLM encoder using the specified parameters.

    Args:
        self: The instance of the LayoutLMEncoder class.
        hidden_states (mindspore.Tensor): The input hidden states for encoding.
        attention_mask (Optional[mindspore.Tensor]): Mask to avoid attending to certain positions.
        head_mask (Optional[mindspore.Tensor]): Mask to specify which heads to disable in the attention computation.
        encoder_hidden_states (Optional[mindspore.Tensor]): Hidden states of the encoder to be used in
            cross-attention layers.
        encoder_attention_mask (Optional[mindspore.Tensor]): Mask for encoder attention mechanism.
        past_key_values (Optional[Tuple[Tuple[mindspore.Tensor]]]): Cached key/values for previous decoding steps.
        use_cache (Optional[bool]): Flag to indicate whether to use caching for decoding.
        output_attentions (Optional[bool]): Flag to output attention scores.
        output_hidden_states (Optional[bool]): Flag to output hidden states for each layer.
        return_dict (Optional[bool]): Flag to indicate returning the output as a dictionary.

    Returns:
        Union[Tuple[mindspore.Tensor], BaseModelOutputWithPastAndCrossAttentions]: The output of the encoder,
            which is either a tuple of hidden states or a complex object containing past key values and attentions.

    Raises:
        Warning: If `use_cache=True` is incompatible with gradient checkpointing,
            it will issue a warning and set `use_cache=False`.
    """
    all_hidden_states = () if output_hidden_states else None
    all_self_attentions = () if output_attentions else None
    all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None

    if self.gradient_checkpointing and self.training:
        if use_cache:
            logger.warning_once(
                "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
            )
            use_cache = False

    next_decoder_cache = () if use_cache else None
    for i, layer_module in enumerate(self.layer):
        if output_hidden_states:
            all_hidden_states = all_hidden_states + (hidden_states,)

        layer_head_mask = head_mask[i] if head_mask is not None else None
        past_key_value = past_key_values[i] if past_key_values is not None else None

        layer_outputs = layer_module(
            hidden_states,
            attention_mask,
            layer_head_mask,
            encoder_hidden_states,
            encoder_attention_mask,
            past_key_value,
            output_attentions,
        )

        hidden_states = layer_outputs[0]
        if use_cache:
            next_decoder_cache += (layer_outputs[-1],)
        if output_attentions:
            all_self_attentions = all_self_attentions + (layer_outputs[1],)
            if self.config.add_cross_attention:
                all_cross_attentions = all_cross_attentions + (layer_outputs[2],)

    if output_hidden_states:
        all_hidden_states = all_hidden_states + (hidden_states,)

    if not return_dict:
        return tuple(
            v
            for v in [
                hidden_states,
                next_decoder_cache,
                all_hidden_states,
                all_self_attentions,
                all_cross_attentions,
            ]
            if v is not None
        )
    return BaseModelOutputWithPastAndCrossAttentions(
        last_hidden_state=hidden_states,
        past_key_values=next_decoder_cache,
        hidden_states=all_hidden_states,
        attentions=all_self_attentions,
        cross_attentions=all_cross_attentions,
    )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForMaskedLM

Bases: LayoutLMPreTrainedModel

LayoutLMForMaskedLM Model

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
class LayoutLMForMaskedLM(LayoutLMPreTrainedModel):
    """LayoutLMForMaskedLM Model"""
    _tied_weights_keys = ["cls.predictions.decoder.bias", "cls.predictions.decoder.weight"]

    def __init__(self, config):
        """
        Initializes the LayoutLMForMaskedLM class.

        Args:
            self: The instance of the class.
            config: The configuration object that contains the model configuration settings.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__(config)

        self.layoutlm = LayoutLMModel(config)
        self.cls = LayoutLMOnlyMLMHead(config)

        # Initialize weights and apply final processing
        self.post_init()

    def get_input_embeddings(self):
        """
        Method to retrieve the input embeddings from the LayoutLM model for Masked Language Modeling task.

        Args:
            self (LayoutLMForMaskedLM): The instance of the LayoutLMForMaskedLM class.
                It represents the model for Masked Language Modeling.

        Returns:
            word_embeddings: The word embeddings from the LayoutLM model's embeddings.

        Raises:
            None.
        """
        return self.layoutlm.embeddings.word_embeddings

    def get_output_embeddings(self):
        '''
        Returns the output embeddings for the LayoutLM model.

        Args:
            self (LayoutLMForMaskedLM): The LayoutLMForMaskedLM object.

        Returns:
            None.

        Raises:
            None.
        '''
        return self.cls.predictions.decoder

    def set_output_embeddings(self, new_embeddings):
        """
        Sets the output embeddings for the LayoutLMForMaskedLM model.

        Args:
            self (LayoutLMForMaskedLM): The instance of the LayoutLMForMaskedLM class.
            new_embeddings (Any): The new embeddings to set for the model's output layer.

        Returns:
            None.

        Raises:
            None.
        """
        self.cls.predictions.decoder = new_embeddings

    def forward(
        self,
        input_ids: Optional[mindspore.Tensor] = None,
        bbox: Optional[mindspore.Tensor] = None,
        attention_mask: Optional[mindspore.Tensor] = None,
        token_type_ids: Optional[mindspore.Tensor] = None,
        position_ids: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        inputs_embeds: Optional[mindspore.Tensor] = None,
        labels: Optional[mindspore.Tensor] = None,
        encoder_hidden_states: Optional[mindspore.Tensor] = None,
        encoder_attention_mask: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, MaskedLMOutput]:
        r"""
        Args:
            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
                Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
                config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
                loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`

        Returns:
            Union[Tuple, MaskedLMOutput]

        Example:
            ```python
            >>> from transformers import AutoTokenizer, LayoutLMForMaskedLM
            >>> import torch
            ...
            >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
            >>> model = LayoutLMForMaskedLM.from_pretrained("microsoft/layoutlm-base-uncased")
            ...
            >>> words = ["Hello", "[MASK]"]
            >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
            ...
            >>> token_boxes = []
            >>> for word, box in zip(words, normalized_word_boxes):
            ...     word_tokens = tokenizer.tokenize(word)
            ...     token_boxes.extend([box] * len(word_tokens))
            >>> # add bounding boxes of cls + sep tokens
            >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
            ...
            >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
            >>> input_ids = encoding["input_ids"]
            >>> attention_mask = encoding["attention_mask"]
            >>> token_type_ids = encoding["token_type_ids"]
            >>> bbox = torch.tensor([token_boxes])
            ...
            >>> labels = tokenizer("Hello world", return_tensors="pt")["input_ids"]
            ...
            >>> outputs = model(
            ...     input_ids=input_ids,
            ...     bbox=bbox,
            ...     attention_mask=attention_mask,
            ...     token_type_ids=token_type_ids,
            ...     labels=labels,
            ... )
            ...
            >>> loss = outputs.loss
            ```
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.layoutlm(
            input_ids,
            bbox,
            attention_mask=attention_mask,
            token_type_ids=token_type_ids,
            position_ids=position_ids,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            encoder_hidden_states=encoder_hidden_states,
            encoder_attention_mask=encoder_attention_mask,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        sequence_output = outputs[0]
        prediction_scores = self.cls(sequence_output)

        masked_lm_loss = None
        if labels is not None:
            masked_lm_loss = ops.cross_entropy(
                prediction_scores.view(-1, self.config.vocab_size),
                labels.view(-1),
            )

        if not return_dict:
            output = (prediction_scores,) + outputs[2:]
            return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output

        return MaskedLMOutput(
            loss=masked_lm_loss,
            logits=prediction_scores,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForMaskedLM.__init__(config)

Initializes the LayoutLMForMaskedLM class.

PARAMETER DESCRIPTION
self

The instance of the class.
config

The configuration object that contains the model configuration settings.
RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
def __init__(self, config):
    """
    Initializes the LayoutLMForMaskedLM class.

    Args:
        self: The instance of the class.
        config: The configuration object that contains the model configuration settings.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__(config)

    self.layoutlm = LayoutLMModel(config)
    self.cls = LayoutLMOnlyMLMHead(config)

    # Initialize weights and apply final processing
    self.post_init()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForMaskedLM.forward(input_ids=None, bbox=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, encoder_hidden_states=None, encoder_attention_mask=None, output_attentions=None, output_hidden_states=None, return_dict=None)

PARAMETER DESCRIPTION
labels

Labels for computing the masked language modeling loss. Indices should be in [-100, 0, ..., config.vocab_size] (see input_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]

TYPE: `torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional* DEFAULT: None

RETURNS DESCRIPTION
Union[Tuple, MaskedLMOutput]

Union[Tuple, MaskedLMOutput]
Example 
>>> from transformers import AutoTokenizer, LayoutLMForMaskedLM
>>> import torch
...
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
>>> model = LayoutLMForMaskedLM.from_pretrained("microsoft/layoutlm-base-uncased")
...
>>> words = ["Hello", "[MASK]"]
>>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
...
>>> token_boxes = []
>>> for word, box in zip(words, normalized_word_boxes):
...     word_tokens = tokenizer.tokenize(word)
...     token_boxes.extend([box] * len(word_tokens))
>>> # add bounding boxes of cls + sep tokens
>>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
...
>>> encoding = tokenizer(" ".join(words), return_tensors="pt")
>>> input_ids = encoding["input_ids"]
>>> attention_mask = encoding["attention_mask"]
>>> token_type_ids = encoding["token_type_ids"]
>>> bbox = torch.tensor([token_boxes])
...
>>> labels = tokenizer("Hello world", return_tensors="pt")["input_ids"]
...
>>> outputs = model(
...     input_ids=input_ids,
...     bbox=bbox,
...     attention_mask=attention_mask,
...     token_type_ids=token_type_ids,
...     labels=labels,
... )
...
>>> loss = outputs.loss
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
def forward(
    self,
    input_ids: Optional[mindspore.Tensor] = None,
    bbox: Optional[mindspore.Tensor] = None,
    attention_mask: Optional[mindspore.Tensor] = None,
    token_type_ids: Optional[mindspore.Tensor] = None,
    position_ids: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    inputs_embeds: Optional[mindspore.Tensor] = None,
    labels: Optional[mindspore.Tensor] = None,
    encoder_hidden_states: Optional[mindspore.Tensor] = None,
    encoder_attention_mask: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, MaskedLMOutput]:
    r"""
    Args:
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
            config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
            loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`

    Returns:
        Union[Tuple, MaskedLMOutput]

    Example:
        ```python
        >>> from transformers import AutoTokenizer, LayoutLMForMaskedLM
        >>> import torch
        ...
        >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
        >>> model = LayoutLMForMaskedLM.from_pretrained("microsoft/layoutlm-base-uncased")
        ...
        >>> words = ["Hello", "[MASK]"]
        >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
        ...
        >>> token_boxes = []
        >>> for word, box in zip(words, normalized_word_boxes):
        ...     word_tokens = tokenizer.tokenize(word)
        ...     token_boxes.extend([box] * len(word_tokens))
        >>> # add bounding boxes of cls + sep tokens
        >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
        ...
        >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
        >>> input_ids = encoding["input_ids"]
        >>> attention_mask = encoding["attention_mask"]
        >>> token_type_ids = encoding["token_type_ids"]
        >>> bbox = torch.tensor([token_boxes])
        ...
        >>> labels = tokenizer("Hello world", return_tensors="pt")["input_ids"]
        ...
        >>> outputs = model(
        ...     input_ids=input_ids,
        ...     bbox=bbox,
        ...     attention_mask=attention_mask,
        ...     token_type_ids=token_type_ids,
        ...     labels=labels,
        ... )
        ...
        >>> loss = outputs.loss
        ```
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.layoutlm(
        input_ids,
        bbox,
        attention_mask=attention_mask,
        token_type_ids=token_type_ids,
        position_ids=position_ids,
        head_mask=head_mask,
        inputs_embeds=inputs_embeds,
        encoder_hidden_states=encoder_hidden_states,
        encoder_attention_mask=encoder_attention_mask,
        output_attentions=output_attentions,
        output_hidden_states=output_hidden_states,
        return_dict=return_dict,
    )

    sequence_output = outputs[0]
    prediction_scores = self.cls(sequence_output)

    masked_lm_loss = None
    if labels is not None:
        masked_lm_loss = ops.cross_entropy(
            prediction_scores.view(-1, self.config.vocab_size),
            labels.view(-1),
        )

    if not return_dict:
        output = (prediction_scores,) + outputs[2:]
        return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output

    return MaskedLMOutput(
        loss=masked_lm_loss,
        logits=prediction_scores,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForMaskedLM.get_input_embeddings()

Method to retrieve the input embeddings from the LayoutLM model for Masked Language Modeling task.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMForMaskedLM class. It represents the model for Masked Language Modeling.

TYPE: LayoutLMForMaskedLM

RETURNS DESCRIPTION
word_embeddings

The word embeddings from the LayoutLM model's embeddings.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
def get_input_embeddings(self):
    """
    Method to retrieve the input embeddings from the LayoutLM model for Masked Language Modeling task.

    Args:
        self (LayoutLMForMaskedLM): The instance of the LayoutLMForMaskedLM class.
            It represents the model for Masked Language Modeling.

    Returns:
        word_embeddings: The word embeddings from the LayoutLM model's embeddings.

    Raises:
        None.
    """
    return self.layoutlm.embeddings.word_embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForMaskedLM.get_output_embeddings()

Returns the output embeddings for the LayoutLM model.

PARAMETER DESCRIPTION
self

The LayoutLMForMaskedLM object.

TYPE: LayoutLMForMaskedLM

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
def get_output_embeddings(self):
    '''
    Returns the output embeddings for the LayoutLM model.

    Args:
        self (LayoutLMForMaskedLM): The LayoutLMForMaskedLM object.

    Returns:
        None.

    Raises:
        None.
    '''
    return self.cls.predictions.decoder

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForMaskedLM.set_output_embeddings(new_embeddings)

Sets the output embeddings for the LayoutLMForMaskedLM model.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMForMaskedLM class.

TYPE: LayoutLMForMaskedLM

new_embeddings

The new embeddings to set for the model's output layer.

TYPE: Any

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
def set_output_embeddings(self, new_embeddings):
    """
    Sets the output embeddings for the LayoutLMForMaskedLM model.

    Args:
        self (LayoutLMForMaskedLM): The instance of the LayoutLMForMaskedLM class.
        new_embeddings (Any): The new embeddings to set for the model's output layer.

    Returns:
        None.

    Raises:
        None.
    """
    self.cls.predictions.decoder = new_embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForQuestionAnswering

Bases: LayoutLMPreTrainedModel

LayoutLMForQuestionAnswering Model

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
class LayoutLMForQuestionAnswering(LayoutLMPreTrainedModel):
    """LayoutLMForQuestionAnswering Model"""
    def __init__(self, config, has_visual_segment_embedding=True):
        """
        Initializes an instance of the LayoutLMForQuestionAnswering class.

        Args:
            self: The instance of the class.
            config (object): An object containing configuration settings.
            has_visual_segment_embedding (bool, optional): Flag indicating whether visual segment embedding is present.
                Defaults to True.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__(config)
        self.num_labels = config.num_labels

        self.layoutlm = LayoutLMModel(config)
        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)

        # Initialize weights and apply final processing
        self.post_init()

    def get_input_embeddings(self):
        """
        Method to retrieve the input embeddings from the LayoutLM model for question answering.

        Args:
            self: An instance of the LayoutLMForQuestionAnswering class.
                It represents the current instance of the model and is used to access the embeddings.

        Returns:
            word_embeddings:
                The word embeddings:

                from the LayoutLM model for input sequences.
                The embeddings are used for processing the input data during question answering tasks.

        Raises:
            None.
        """
        return self.layoutlm.embeddings.word_embeddings

    def forward(
        self,
        input_ids: Optional[mindspore.Tensor] = None,
        bbox: Optional[mindspore.Tensor] = None,
        attention_mask: Optional[mindspore.Tensor] = None,
        token_type_ids: Optional[mindspore.Tensor] = None,
        position_ids: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        inputs_embeds: Optional[mindspore.Tensor] = None,
        start_positions: Optional[mindspore.Tensor] = None,
        end_positions: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, QuestionAnsweringModelOutput]:
        r"""
        Args:
            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:
            Union[Tuple, QuestionAnsweringModelOutput]

        In the example below, we prepare a question + context pair for the LayoutLM model. It will give us a prediction
        of what it thinks the answer is (the span of the answer within the texts parsed from the image).

        Example:
            ```python
            >>> from transformers import AutoTokenizer, LayoutLMForQuestionAnswering
            >>> from datasets import load_dataset
            >>> import torch
            ...
            >>> tokenizer = AutoTokenizer.from_pretrained("impira/layoutlm-document-qa", add_prefix_space=True)
            >>> model = LayoutLMForQuestionAnswering.from_pretrained("impira/layoutlm-document-qa", revision="1e3ebac")
            ...
            >>> dataset = load_dataset("nielsr/funsd", split="train")
            >>> example = dataset[0]
            >>> question = "what's his name?"
            >>> words = example["words"]
            >>> boxes = example["bboxes"]
            ...
            >>> encoding = tokenizer(
            ...     question.split(), words, is_split_into_words=True, return_token_type_ids=True, return_tensors="pt"
            ... )
            >>> bbox = []
            >>> for i, s, w in zip(encoding.input_ids[0], encoding.sequence_ids(0), encoding.word_ids(0)):
            ...     if s == 1:
            ...         bbox.append(boxes[w])
            ...     elif i == tokenizer.sep_token_id:
            ...         bbox.append([1000] * 4)
            ...     else:
            ...         bbox.append([0] * 4)
            >>> encoding["bbox"] = torch.tensor([bbox])
            ...
            >>> word_ids = encoding.word_ids(0)
            >>> outputs = model(**encoding)
            >>> loss = outputs.loss
            >>> start_scores = outputs.start_logits
            >>> end_scores = outputs.end_logits
            >>> start, end = word_ids[start_scores.argmax(-1)], word_ids[end_scores.argmax(-1)]
            >>> print(" ".join(words[start : end + 1]))
            M. Hamann P. Harper, P. Martinez
            ```
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.layoutlm(
            input_ids=input_ids,
            bbox=bbox,
            attention_mask=attention_mask,
            token_type_ids=token_type_ids,
            position_ids=position_ids,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        sequence_output = outputs[0]

        logits = self.qa_outputs(sequence_output)
        start_logits, end_logits = logits.split(1, axis=-1)
        start_logits = start_logits.squeeze(-1)
        end_logits = end_logits.squeeze(-1)

        total_loss = None
        if start_positions is not None and end_positions is not None:
            # If we are on multi-GPU, split add a dimension
            if len(start_positions.shape) > 1:
                start_positions = start_positions.squeeze(-1)
            if len(end_positions.shape) > 1:
                end_positions = end_positions.squeeze(-1)
            # sometimes the start/end positions are outside our model inputs, we ignore these terms
            ignored_index = start_logits.shape[1]
            start_positions = start_positions.clamp(0, ignored_index)
            end_positions = end_positions.clamp(0, ignored_index)

            start_loss = ops.cross_entropy(start_logits, start_positions, ignore_index=ignored_index)
            end_loss = ops.cross_entropy(end_logits, end_positions, ignore_index=ignored_index)
            total_loss = (start_loss + end_loss) / 2

        if not return_dict:
            output = (start_logits, end_logits) + outputs[2:]
            return ((total_loss,) + output) if total_loss is not None else output

        return QuestionAnsweringModelOutput(
            loss=total_loss,
            start_logits=start_logits,
            end_logits=end_logits,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForQuestionAnswering.__init__(config, has_visual_segment_embedding=True)

Initializes an instance of the LayoutLMForQuestionAnswering class.

PARAMETER DESCRIPTION
self

The instance of the class.
config

An object containing configuration settings.

TYPE: object

has_visual_segment_embedding

Flag indicating whether visual segment embedding is present. Defaults to True.

TYPE: bool DEFAULT: True

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
def __init__(self, config, has_visual_segment_embedding=True):
    """
    Initializes an instance of the LayoutLMForQuestionAnswering class.

    Args:
        self: The instance of the class.
        config (object): An object containing configuration settings.
        has_visual_segment_embedding (bool, optional): Flag indicating whether visual segment embedding is present.
            Defaults to True.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__(config)
    self.num_labels = config.num_labels

    self.layoutlm = LayoutLMModel(config)
    self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)

    # Initialize weights and apply final processing
    self.post_init()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForQuestionAnswering.forward(input_ids=None, bbox=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, start_positions=None, end_positions=None, output_attentions=None, output_hidden_states=None, return_dict=None)

PARAMETER DESCRIPTION
start_positions

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.

TYPE: `torch.LongTensor` of shape `(batch_size,)`, *optional* DEFAULT: None

end_positions

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.

TYPE: `torch.LongTensor` of shape `(batch_size,)`, *optional* DEFAULT: None

RETURNS DESCRIPTION
Union[Tuple, QuestionAnsweringModelOutput]

Union[Tuple, QuestionAnsweringModelOutput]

In the example below, we prepare a question + context pair for the LayoutLM model. It will give us a prediction of what it thinks the answer is (the span of the answer within the texts parsed from the image).

Example 
>>> from transformers import AutoTokenizer, LayoutLMForQuestionAnswering
>>> from datasets import load_dataset
>>> import torch
...
>>> tokenizer = AutoTokenizer.from_pretrained("impira/layoutlm-document-qa", add_prefix_space=True)
>>> model = LayoutLMForQuestionAnswering.from_pretrained("impira/layoutlm-document-qa", revision="1e3ebac")
...
>>> dataset = load_dataset("nielsr/funsd", split="train")
>>> example = dataset[0]
>>> question = "what's his name?"
>>> words = example["words"]
>>> boxes = example["bboxes"]
...
>>> encoding = tokenizer(
...     question.split(), words, is_split_into_words=True, return_token_type_ids=True, return_tensors="pt"
... )
>>> bbox = []
>>> for i, s, w in zip(encoding.input_ids[0], encoding.sequence_ids(0), encoding.word_ids(0)):
...     if s == 1:
...         bbox.append(boxes[w])
...     elif i == tokenizer.sep_token_id:
...         bbox.append([1000] * 4)
...     else:
...         bbox.append([0] * 4)
>>> encoding["bbox"] = torch.tensor([bbox])
...
>>> word_ids = encoding.word_ids(0)
>>> outputs = model(**encoding)
>>> loss = outputs.loss
>>> start_scores = outputs.start_logits
>>> end_scores = outputs.end_logits
>>> start, end = word_ids[start_scores.argmax(-1)], word_ids[end_scores.argmax(-1)]
>>> print(" ".join(words[start : end + 1]))
M. Hamann P. Harper, P. Martinez
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
def forward(
    self,
    input_ids: Optional[mindspore.Tensor] = None,
    bbox: Optional[mindspore.Tensor] = None,
    attention_mask: Optional[mindspore.Tensor] = None,
    token_type_ids: Optional[mindspore.Tensor] = None,
    position_ids: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    inputs_embeds: Optional[mindspore.Tensor] = None,
    start_positions: Optional[mindspore.Tensor] = None,
    end_positions: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, QuestionAnsweringModelOutput]:
    r"""
    Args:
        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:
        Union[Tuple, QuestionAnsweringModelOutput]

    In the example below, we prepare a question + context pair for the LayoutLM model. It will give us a prediction
    of what it thinks the answer is (the span of the answer within the texts parsed from the image).

    Example:
        ```python
        >>> from transformers import AutoTokenizer, LayoutLMForQuestionAnswering
        >>> from datasets import load_dataset
        >>> import torch
        ...
        >>> tokenizer = AutoTokenizer.from_pretrained("impira/layoutlm-document-qa", add_prefix_space=True)
        >>> model = LayoutLMForQuestionAnswering.from_pretrained("impira/layoutlm-document-qa", revision="1e3ebac")
        ...
        >>> dataset = load_dataset("nielsr/funsd", split="train")
        >>> example = dataset[0]
        >>> question = "what's his name?"
        >>> words = example["words"]
        >>> boxes = example["bboxes"]
        ...
        >>> encoding = tokenizer(
        ...     question.split(), words, is_split_into_words=True, return_token_type_ids=True, return_tensors="pt"
        ... )
        >>> bbox = []
        >>> for i, s, w in zip(encoding.input_ids[0], encoding.sequence_ids(0), encoding.word_ids(0)):
        ...     if s == 1:
        ...         bbox.append(boxes[w])
        ...     elif i == tokenizer.sep_token_id:
        ...         bbox.append([1000] * 4)
        ...     else:
        ...         bbox.append([0] * 4)
        >>> encoding["bbox"] = torch.tensor([bbox])
        ...
        >>> word_ids = encoding.word_ids(0)
        >>> outputs = model(**encoding)
        >>> loss = outputs.loss
        >>> start_scores = outputs.start_logits
        >>> end_scores = outputs.end_logits
        >>> start, end = word_ids[start_scores.argmax(-1)], word_ids[end_scores.argmax(-1)]
        >>> print(" ".join(words[start : end + 1]))
        M. Hamann P. Harper, P. Martinez
        ```
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.layoutlm(
        input_ids=input_ids,
        bbox=bbox,
        attention_mask=attention_mask,
        token_type_ids=token_type_ids,
        position_ids=position_ids,
        head_mask=head_mask,
        inputs_embeds=inputs_embeds,
        output_attentions=output_attentions,
        output_hidden_states=output_hidden_states,
        return_dict=return_dict,
    )

    sequence_output = outputs[0]

    logits = self.qa_outputs(sequence_output)
    start_logits, end_logits = logits.split(1, axis=-1)
    start_logits = start_logits.squeeze(-1)
    end_logits = end_logits.squeeze(-1)

    total_loss = None
    if start_positions is not None and end_positions is not None:
        # If we are on multi-GPU, split add a dimension
        if len(start_positions.shape) > 1:
            start_positions = start_positions.squeeze(-1)
        if len(end_positions.shape) > 1:
            end_positions = end_positions.squeeze(-1)
        # sometimes the start/end positions are outside our model inputs, we ignore these terms
        ignored_index = start_logits.shape[1]
        start_positions = start_positions.clamp(0, ignored_index)
        end_positions = end_positions.clamp(0, ignored_index)

        start_loss = ops.cross_entropy(start_logits, start_positions, ignore_index=ignored_index)
        end_loss = ops.cross_entropy(end_logits, end_positions, ignore_index=ignored_index)
        total_loss = (start_loss + end_loss) / 2

    if not return_dict:
        output = (start_logits, end_logits) + outputs[2:]
        return ((total_loss,) + output) if total_loss is not None else output

    return QuestionAnsweringModelOutput(
        loss=total_loss,
        start_logits=start_logits,
        end_logits=end_logits,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForQuestionAnswering.get_input_embeddings()

Method to retrieve the input embeddings from the LayoutLM model for question answering.

PARAMETER DESCRIPTION
self

An instance of the LayoutLMForQuestionAnswering class. It represents the current instance of the model and is used to access the embeddings.
RETURNS DESCRIPTION
word_embeddings

The word embeddings:

from the LayoutLM model for input sequences. The embeddings are used for processing the input data during question answering tasks.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
def get_input_embeddings(self):
    """
    Method to retrieve the input embeddings from the LayoutLM model for question answering.

    Args:
        self: An instance of the LayoutLMForQuestionAnswering class.
            It represents the current instance of the model and is used to access the embeddings.

    Returns:
        word_embeddings:
            The word embeddings:

            from the LayoutLM model for input sequences.
            The embeddings are used for processing the input data during question answering tasks.

    Raises:
        None.
    """
    return self.layoutlm.embeddings.word_embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForSequenceClassification

Bases: LayoutLMPreTrainedModel

LayoutLMForSequenceClassification Model

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
class LayoutLMForSequenceClassification(LayoutLMPreTrainedModel):
    """LayoutLMForSequenceClassification Model"""
    def __init__(self, config):
        """
        __init__

        Initializes the LayoutLMForSequenceClassification class.

        Args:
            self: The instance of the class.
            config: An instance of the configuration class containing the model configuration parameters.

        Returns:
            None.

        Raises:
            TypeError: If the config parameter is not of the expected type.
            ValueError: If the num_labels attribute is not present in the config parameter.
            RuntimeError: If an error occurs during the initialization process.
        """
        super().__init__(config)
        self.num_labels = config.num_labels
        self.layoutlm = LayoutLMModel(config)
        self.dropout = nn.Dropout(p=config.hidden_dropout_prob)
        self.classifier = nn.Linear(config.hidden_size, config.num_labels)

        # Initialize weights and apply final processing
        self.post_init()

    def get_input_embeddings(self):
        """
        This method, get_input_embeddings, retrieves the input embeddings from the LayoutLM model for sequence classification.

        Args:
            self (LayoutLMForSequenceClassification): The instance of the LayoutLMForSequenceClassification class.
                This parameter refers to the current instance of the LayoutLMForSequenceClassification class.

        Returns:
            None:
                The input embeddings from the LayoutLM model for sequence classification.

        Raises:
            None.
        """
        return self.layoutlm.embeddings.word_embeddings

    def forward(
        self,
        input_ids: Optional[mindspore.Tensor] = None,
        bbox: Optional[mindspore.Tensor] = None,
        attention_mask: Optional[mindspore.Tensor] = None,
        token_type_ids: Optional[mindspore.Tensor] = None,
        position_ids: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        inputs_embeds: Optional[mindspore.Tensor] = None,
        labels: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, SequenceClassifierOutput]:
        r"""
        Args:
            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).

        Returns:
            Union[Tuple, SequenceClassifierOutput]

        Example:
            ```python
            >>> from transformers import AutoTokenizer, LayoutLMForSequenceClassification
            >>> import torch
            ...
            >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
            >>> model = LayoutLMForSequenceClassification.from_pretrained("microsoft/layoutlm-base-uncased")
            ...
            >>> words = ["Hello", "world"]
            >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
            ...
            >>> token_boxes = []
            >>> for word, box in zip(words, normalized_word_boxes):
            ...     word_tokens = tokenizer.tokenize(word)
            ...     token_boxes.extend([box] * len(word_tokens))
            >>> # add bounding boxes of cls + sep tokens
            >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
            ...
            >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
            >>> input_ids = encoding["input_ids"]
            >>> attention_mask = encoding["attention_mask"]
            >>> token_type_ids = encoding["token_type_ids"]
            >>> bbox = torch.tensor([token_boxes])
            >>> sequence_label = torch.tensor([1])
            ...
            >>> outputs = model(
            ...     input_ids=input_ids,
            ...     bbox=bbox,
            ...     attention_mask=attention_mask,
            ...     token_type_ids=token_type_ids,
            ...     labels=sequence_label,
            ... )
            ...
            >>> loss = outputs.loss
            >>> logits = outputs.logits
            ```
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.layoutlm(
            input_ids=input_ids,
            bbox=bbox,
            attention_mask=attention_mask,
            token_type_ids=token_type_ids,
            position_ids=position_ids,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        pooled_output = outputs[1]

        pooled_output = self.dropout(pooled_output)
        logits = self.classifier(pooled_output)

        loss = None
        if labels is not None:
            if self.config.problem_type is None:
                if self.num_labels == 1:
                    self.config.problem_type = "regression"
                elif self.num_labels > 1 and labels.dtype in (mindspore.int32, mindspore.int64):
                    self.config.problem_type = "single_label_classification"
                else:
                    self.config.problem_type = "multi_label_classification"

            if self.config.problem_type == "regression":
                if self.num_labels == 1:
                    loss = ops.mse_loss(logits.squeeze(), labels.squeeze())
                else:
                    loss = ops.mse_loss(logits, labels)
            elif self.config.problem_type == "single_label_classification":
                loss = ops.cross_entropy(logits.view(-1, self.num_labels), labels.view(-1))
            elif self.config.problem_type == "multi_label_classification":
                loss = ops.binary_cross_entropy_with_logits(logits, labels)
        if not return_dict:
            output = (logits,) + outputs[2:]
            return ((loss,) + output) if loss is not None else output

        return SequenceClassifierOutput(
            loss=loss,
            logits=logits,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForSequenceClassification.__init__(config)

init

Initializes the LayoutLMForSequenceClassification class.

PARAMETER DESCRIPTION
self

The instance of the class.
config

An instance of the configuration class containing the model configuration parameters.
RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
TypeError

If the config parameter is not of the expected type.
ValueError

If the num_labels attribute is not present in the config parameter.
RuntimeError

If an error occurs during the initialization process.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
def __init__(self, config):
    """
    __init__

    Initializes the LayoutLMForSequenceClassification class.

    Args:
        self: The instance of the class.
        config: An instance of the configuration class containing the model configuration parameters.

    Returns:
        None.

    Raises:
        TypeError: If the config parameter is not of the expected type.
        ValueError: If the num_labels attribute is not present in the config parameter.
        RuntimeError: If an error occurs during the initialization process.
    """
    super().__init__(config)
    self.num_labels = config.num_labels
    self.layoutlm = LayoutLMModel(config)
    self.dropout = nn.Dropout(p=config.hidden_dropout_prob)
    self.classifier = nn.Linear(config.hidden_size, config.num_labels)

    # Initialize weights and apply final processing
    self.post_init()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForSequenceClassification.forward(input_ids=None, bbox=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, output_attentions=None, output_hidden_states=None, return_dict=None)

PARAMETER DESCRIPTION
labels

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).

TYPE: `torch.LongTensor` of shape `(batch_size,)`, *optional* DEFAULT: None

RETURNS DESCRIPTION
Union[Tuple, SequenceClassifierOutput]

Union[Tuple, SequenceClassifierOutput]
Example 
>>> from transformers import AutoTokenizer, LayoutLMForSequenceClassification
>>> import torch
...
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
>>> model = LayoutLMForSequenceClassification.from_pretrained("microsoft/layoutlm-base-uncased")
...
>>> words = ["Hello", "world"]
>>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
...
>>> token_boxes = []
>>> for word, box in zip(words, normalized_word_boxes):
...     word_tokens = tokenizer.tokenize(word)
...     token_boxes.extend([box] * len(word_tokens))
>>> # add bounding boxes of cls + sep tokens
>>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
...
>>> encoding = tokenizer(" ".join(words), return_tensors="pt")
>>> input_ids = encoding["input_ids"]
>>> attention_mask = encoding["attention_mask"]
>>> token_type_ids = encoding["token_type_ids"]
>>> bbox = torch.tensor([token_boxes])
>>> sequence_label = torch.tensor([1])
...
>>> outputs = model(
...     input_ids=input_ids,
...     bbox=bbox,
...     attention_mask=attention_mask,
...     token_type_ids=token_type_ids,
...     labels=sequence_label,
... )
...
>>> loss = outputs.loss
>>> logits = outputs.logits
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
def forward(
    self,
    input_ids: Optional[mindspore.Tensor] = None,
    bbox: Optional[mindspore.Tensor] = None,
    attention_mask: Optional[mindspore.Tensor] = None,
    token_type_ids: Optional[mindspore.Tensor] = None,
    position_ids: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    inputs_embeds: Optional[mindspore.Tensor] = None,
    labels: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, SequenceClassifierOutput]:
    r"""
    Args:
        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).

    Returns:
        Union[Tuple, SequenceClassifierOutput]

    Example:
        ```python
        >>> from transformers import AutoTokenizer, LayoutLMForSequenceClassification
        >>> import torch
        ...
        >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
        >>> model = LayoutLMForSequenceClassification.from_pretrained("microsoft/layoutlm-base-uncased")
        ...
        >>> words = ["Hello", "world"]
        >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
        ...
        >>> token_boxes = []
        >>> for word, box in zip(words, normalized_word_boxes):
        ...     word_tokens = tokenizer.tokenize(word)
        ...     token_boxes.extend([box] * len(word_tokens))
        >>> # add bounding boxes of cls + sep tokens
        >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
        ...
        >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
        >>> input_ids = encoding["input_ids"]
        >>> attention_mask = encoding["attention_mask"]
        >>> token_type_ids = encoding["token_type_ids"]
        >>> bbox = torch.tensor([token_boxes])
        >>> sequence_label = torch.tensor([1])
        ...
        >>> outputs = model(
        ...     input_ids=input_ids,
        ...     bbox=bbox,
        ...     attention_mask=attention_mask,
        ...     token_type_ids=token_type_ids,
        ...     labels=sequence_label,
        ... )
        ...
        >>> loss = outputs.loss
        >>> logits = outputs.logits
        ```
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.layoutlm(
        input_ids=input_ids,
        bbox=bbox,
        attention_mask=attention_mask,
        token_type_ids=token_type_ids,
        position_ids=position_ids,
        head_mask=head_mask,
        inputs_embeds=inputs_embeds,
        output_attentions=output_attentions,
        output_hidden_states=output_hidden_states,
        return_dict=return_dict,
    )

    pooled_output = outputs[1]

    pooled_output = self.dropout(pooled_output)
    logits = self.classifier(pooled_output)

    loss = None
    if labels is not None:
        if self.config.problem_type is None:
            if self.num_labels == 1:
                self.config.problem_type = "regression"
            elif self.num_labels > 1 and labels.dtype in (mindspore.int32, mindspore.int64):
                self.config.problem_type = "single_label_classification"
            else:
                self.config.problem_type = "multi_label_classification"

        if self.config.problem_type == "regression":
            if self.num_labels == 1:
                loss = ops.mse_loss(logits.squeeze(), labels.squeeze())
            else:
                loss = ops.mse_loss(logits, labels)
        elif self.config.problem_type == "single_label_classification":
            loss = ops.cross_entropy(logits.view(-1, self.num_labels), labels.view(-1))
        elif self.config.problem_type == "multi_label_classification":
            loss = ops.binary_cross_entropy_with_logits(logits, labels)
    if not return_dict:
        output = (logits,) + outputs[2:]
        return ((loss,) + output) if loss is not None else output

    return SequenceClassifierOutput(
        loss=loss,
        logits=logits,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForSequenceClassification.get_input_embeddings()

This method, get_input_embeddings, retrieves the input embeddings from the LayoutLM model for sequence classification.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMForSequenceClassification class. This parameter refers to the current instance of the LayoutLMForSequenceClassification class.

TYPE: LayoutLMForSequenceClassification

RETURNS DESCRIPTION
None

The input embeddings from the LayoutLM model for sequence classification.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
def get_input_embeddings(self):
    """
    This method, get_input_embeddings, retrieves the input embeddings from the LayoutLM model for sequence classification.

    Args:
        self (LayoutLMForSequenceClassification): The instance of the LayoutLMForSequenceClassification class.
            This parameter refers to the current instance of the LayoutLMForSequenceClassification class.

    Returns:
        None:
            The input embeddings from the LayoutLM model for sequence classification.

    Raises:
        None.
    """
    return self.layoutlm.embeddings.word_embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForTokenClassification

Bases: LayoutLMPreTrainedModel

LayoutLMForTokenClassification Model

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
class LayoutLMForTokenClassification(LayoutLMPreTrainedModel):
    """LayoutLMForTokenClassification Model"""
    def __init__(self, config):
        """
        Initializes an instance of the LayoutLMForTokenClassification class.

        Args:
            self: The instance of the LayoutLMForTokenClassification class.
            config:
                An object of the LayoutLMConfig class containing the configuration parameters for the LayoutLM model.

                - Type: LayoutLMConfig
                - Purpose: Specifies the configuration parameters for the LayoutLM model.
                - Restrictions: None

        Returns:
            None

        Raises:
            None
        """
        super().__init__(config)
        self.num_labels = config.num_labels
        self.layoutlm = LayoutLMModel(config)
        self.dropout = nn.Dropout(p=config.hidden_dropout_prob)
        self.classifier = nn.Linear(config.hidden_size, config.num_labels)

        # Initialize weights and apply final processing
        self.post_init()

    def get_input_embeddings(self):
        """
        This method returns the word embeddings from the LayoutLM model for token classification.

        Args:
            self: The instance of the LayoutLMForTokenClassification class.

        Returns:
            word_embeddings: The word embeddings from the LayoutLM model for token classification.

        Raises:
            None.
        """
        return self.layoutlm.embeddings.word_embeddings

    def forward(
        self,
        input_ids: Optional[mindspore.Tensor] = None,
        bbox: Optional[mindspore.Tensor] = None,
        attention_mask: Optional[mindspore.Tensor] = None,
        token_type_ids: Optional[mindspore.Tensor] = None,
        position_ids: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        inputs_embeds: Optional[mindspore.Tensor] = None,
        labels: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, TokenClassifierOutput]:
        r"""
        Args:
            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]`.

        Returns:
            Union[Tuple, TokenClassifierOutput]

        Example:
            ```python
            >>> from transformers import AutoTokenizer, LayoutLMForTokenClassification
            >>> import torch
            ...
            >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
            >>> model = LayoutLMForTokenClassification.from_pretrained("microsoft/layoutlm-base-uncased")
            ...
            >>> words = ["Hello", "world"]
            >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
            ...
            >>> token_boxes = []
            >>> for word, box in zip(words, normalized_word_boxes):
            ...     word_tokens = tokenizer.tokenize(word)
            ...     token_boxes.extend([box] * len(word_tokens))
            >>> # add bounding boxes of cls + sep tokens
            >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
            ...
            >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
            >>> input_ids = encoding["input_ids"]
            >>> attention_mask = encoding["attention_mask"]
            >>> token_type_ids = encoding["token_type_ids"]
            >>> bbox = torch.tensor([token_boxes])
            >>> token_labels = torch.tensor([1, 1, 0, 0]).unsqueeze(0)  # batch size of 1
            ...
            >>> outputs = model(
            ...     input_ids=input_ids,
            ...     bbox=bbox,
            ...     attention_mask=attention_mask,
            ...     token_type_ids=token_type_ids,
            ...     labels=token_labels,
            ... )
            ...
            >>> loss = outputs.loss
            >>> logits = outputs.logits
            ```
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.layoutlm(
            input_ids=input_ids,
            bbox=bbox,
            attention_mask=attention_mask,
            token_type_ids=token_type_ids,
            position_ids=position_ids,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        sequence_output = outputs[0]

        sequence_output = self.dropout(sequence_output)
        logits = self.classifier(sequence_output)

        loss = None
        if labels is not None:
            loss = ops.cross_entropy(logits.view(-1, self.num_labels), labels.view(-1))

        if not return_dict:
            output = (logits,) + outputs[2:]
            return ((loss,) + output) if loss is not None else output

        return TokenClassifierOutput(
            loss=loss,
            logits=logits,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForTokenClassification.__init__(config)

Initializes an instance of the LayoutLMForTokenClassification class.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMForTokenClassification class.
config

An object of the LayoutLMConfig class containing the configuration parameters for the LayoutLM model.

  • Type: LayoutLMConfig
  • Purpose: Specifies the configuration parameters for the LayoutLM model.
  • Restrictions: None
RETURNS DESCRIPTION

None
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
def __init__(self, config):
    """
    Initializes an instance of the LayoutLMForTokenClassification class.

    Args:
        self: The instance of the LayoutLMForTokenClassification class.
        config:
            An object of the LayoutLMConfig class containing the configuration parameters for the LayoutLM model.

            - Type: LayoutLMConfig
            - Purpose: Specifies the configuration parameters for the LayoutLM model.
            - Restrictions: None

    Returns:
        None

    Raises:
        None
    """
    super().__init__(config)
    self.num_labels = config.num_labels
    self.layoutlm = LayoutLMModel(config)
    self.dropout = nn.Dropout(p=config.hidden_dropout_prob)
    self.classifier = nn.Linear(config.hidden_size, config.num_labels)

    # Initialize weights and apply final processing
    self.post_init()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForTokenClassification.forward(input_ids=None, bbox=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, output_attentions=None, output_hidden_states=None, return_dict=None)

PARAMETER DESCRIPTION
labels

Labels for computing the token classification loss. Indices should be in [0, ..., config.num_labels - 1].

TYPE: `torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional* DEFAULT: None

RETURNS DESCRIPTION
Union[Tuple, TokenClassifierOutput]

Union[Tuple, TokenClassifierOutput]
Example 
>>> from transformers import AutoTokenizer, LayoutLMForTokenClassification
>>> import torch
...
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
>>> model = LayoutLMForTokenClassification.from_pretrained("microsoft/layoutlm-base-uncased")
...
>>> words = ["Hello", "world"]
>>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
...
>>> token_boxes = []
>>> for word, box in zip(words, normalized_word_boxes):
...     word_tokens = tokenizer.tokenize(word)
...     token_boxes.extend([box] * len(word_tokens))
>>> # add bounding boxes of cls + sep tokens
>>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
...
>>> encoding = tokenizer(" ".join(words), return_tensors="pt")
>>> input_ids = encoding["input_ids"]
>>> attention_mask = encoding["attention_mask"]
>>> token_type_ids = encoding["token_type_ids"]
>>> bbox = torch.tensor([token_boxes])
>>> token_labels = torch.tensor([1, 1, 0, 0]).unsqueeze(0)  # batch size of 1
...
>>> outputs = model(
...     input_ids=input_ids,
...     bbox=bbox,
...     attention_mask=attention_mask,
...     token_type_ids=token_type_ids,
...     labels=token_labels,
... )
...
>>> loss = outputs.loss
>>> logits = outputs.logits
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
def forward(
    self,
    input_ids: Optional[mindspore.Tensor] = None,
    bbox: Optional[mindspore.Tensor] = None,
    attention_mask: Optional[mindspore.Tensor] = None,
    token_type_ids: Optional[mindspore.Tensor] = None,
    position_ids: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    inputs_embeds: Optional[mindspore.Tensor] = None,
    labels: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, TokenClassifierOutput]:
    r"""
    Args:
        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]`.

    Returns:
        Union[Tuple, TokenClassifierOutput]

    Example:
        ```python
        >>> from transformers import AutoTokenizer, LayoutLMForTokenClassification
        >>> import torch
        ...
        >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
        >>> model = LayoutLMForTokenClassification.from_pretrained("microsoft/layoutlm-base-uncased")
        ...
        >>> words = ["Hello", "world"]
        >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
        ...
        >>> token_boxes = []
        >>> for word, box in zip(words, normalized_word_boxes):
        ...     word_tokens = tokenizer.tokenize(word)
        ...     token_boxes.extend([box] * len(word_tokens))
        >>> # add bounding boxes of cls + sep tokens
        >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
        ...
        >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
        >>> input_ids = encoding["input_ids"]
        >>> attention_mask = encoding["attention_mask"]
        >>> token_type_ids = encoding["token_type_ids"]
        >>> bbox = torch.tensor([token_boxes])
        >>> token_labels = torch.tensor([1, 1, 0, 0]).unsqueeze(0)  # batch size of 1
        ...
        >>> outputs = model(
        ...     input_ids=input_ids,
        ...     bbox=bbox,
        ...     attention_mask=attention_mask,
        ...     token_type_ids=token_type_ids,
        ...     labels=token_labels,
        ... )
        ...
        >>> loss = outputs.loss
        >>> logits = outputs.logits
        ```
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.layoutlm(
        input_ids=input_ids,
        bbox=bbox,
        attention_mask=attention_mask,
        token_type_ids=token_type_ids,
        position_ids=position_ids,
        head_mask=head_mask,
        inputs_embeds=inputs_embeds,
        output_attentions=output_attentions,
        output_hidden_states=output_hidden_states,
        return_dict=return_dict,
    )

    sequence_output = outputs[0]

    sequence_output = self.dropout(sequence_output)
    logits = self.classifier(sequence_output)

    loss = None
    if labels is not None:
        loss = ops.cross_entropy(logits.view(-1, self.num_labels), labels.view(-1))

    if not return_dict:
        output = (logits,) + outputs[2:]
        return ((loss,) + output) if loss is not None else output

    return TokenClassifierOutput(
        loss=loss,
        logits=logits,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMForTokenClassification.get_input_embeddings()

This method returns the word embeddings from the LayoutLM model for token classification.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMForTokenClassification class.
RETURNS DESCRIPTION
word_embeddings

The word embeddings from the LayoutLM model for token classification.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
def get_input_embeddings(self):
    """
    This method returns the word embeddings from the LayoutLM model for token classification.

    Args:
        self: The instance of the LayoutLMForTokenClassification class.

    Returns:
        word_embeddings: The word embeddings from the LayoutLM model for token classification.

    Raises:
        None.
    """
    return self.layoutlm.embeddings.word_embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMIntermediate

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertIntermediate

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
class LayoutLMIntermediate(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertIntermediate"""
    def __init__(self, config):
        """
        Initializes an instance of the LayoutLMIntermediate class.

        Args:
            self: The current instance of the class.
            config: An object of type 'config' containing the configuration settings for the intermediate layer.
                This parameter is required and has no default value.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__()
        self.dense = nn.Linear(config.hidden_size, config.intermediate_size)
        if isinstance(config.hidden_act, str):
            self.intermediate_act_fn = ACT2FN[config.hidden_act]
        else:
            self.intermediate_act_fn = config.hidden_act

    def forward(self, hidden_states: mindspore.Tensor) -> mindspore.Tensor:
        """
        Constructs the intermediate layer in the LayoutLM model.

        Args:
            self (LayoutLMIntermediate): An instance of the LayoutLMIntermediate class.
            hidden_states (mindspore.Tensor): The input tensor representing hidden states.

        Returns:
            mindspore.Tensor: The output tensor after passing through the intermediate layer.

        Raises:
            None.
        """
        hidden_states = self.dense(hidden_states)
        hidden_states = self.intermediate_act_fn(hidden_states)
        return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMIntermediate.__init__(config)

Initializes an instance of the LayoutLMIntermediate class.

PARAMETER DESCRIPTION
self

The current instance of the class.
config

An object of type 'config' containing the configuration settings for the intermediate layer. This parameter is required and has no default value.
RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
def __init__(self, config):
    """
    Initializes an instance of the LayoutLMIntermediate class.

    Args:
        self: The current instance of the class.
        config: An object of type 'config' containing the configuration settings for the intermediate layer.
            This parameter is required and has no default value.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__()
    self.dense = nn.Linear(config.hidden_size, config.intermediate_size)
    if isinstance(config.hidden_act, str):
        self.intermediate_act_fn = ACT2FN[config.hidden_act]
    else:
        self.intermediate_act_fn = config.hidden_act

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMIntermediate.forward(hidden_states)

Constructs the intermediate layer in the LayoutLM model.

PARAMETER DESCRIPTION
self

An instance of the LayoutLMIntermediate class.

TYPE: LayoutLMIntermediate

hidden_states

The input tensor representing hidden states.

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: The output tensor after passing through the intermediate layer.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
def forward(self, hidden_states: mindspore.Tensor) -> mindspore.Tensor:
    """
    Constructs the intermediate layer in the LayoutLM model.

    Args:
        self (LayoutLMIntermediate): An instance of the LayoutLMIntermediate class.
        hidden_states (mindspore.Tensor): The input tensor representing hidden states.

    Returns:
        mindspore.Tensor: The output tensor after passing through the intermediate layer.

    Raises:
        None.
    """
    hidden_states = self.dense(hidden_states)
    hidden_states = self.intermediate_act_fn(hidden_states)
    return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLMPredictionHead

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
class LayoutLMLMPredictionHead(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead with Bert->LayoutLM"""
    def __init__(self, config):
        """
        Initializes an instance of the LayoutLMLMPredictionHead class.

        Args:
            self: The instance of the LayoutLMLMPredictionHead class.
            config: An object containing configuration parameters for the LayoutLMLMPredictionHead.
                It is expected to be an instance of a class that holds information such as hidden size and vocabulary size.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__()
        self.transform = LayoutLMPredictionHeadTransform(config)

        # The output weights are the same as the input embeddings, but there is
        # an output-only bias for each token.
        self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)

        self.bias = mindspore.Parameter(ops.zeros(config.vocab_size))

        # Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
        self.decoder.bias = self.bias

    def forward(self, hidden_states):
        """
        Constructs the LayoutLMLMPredictionHead by transforming and decoding hidden states.

        Args:
            self (LayoutLMLMPredictionHead): An instance of the LayoutLMLMPredictionHead class.
            hidden_states (torch.Tensor): The input hidden states to be processed by the prediction head.

        Returns:
            None.

        Raises:
            None
        """
        hidden_states = self.transform(hidden_states)
        hidden_states = self.decoder(hidden_states)
        return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLMPredictionHead.__init__(config)

Initializes an instance of the LayoutLMLMPredictionHead class.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMLMPredictionHead class.
config

An object containing configuration parameters for the LayoutLMLMPredictionHead. It is expected to be an instance of a class that holds information such as hidden size and vocabulary size.
RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
def __init__(self, config):
    """
    Initializes an instance of the LayoutLMLMPredictionHead class.

    Args:
        self: The instance of the LayoutLMLMPredictionHead class.
        config: An object containing configuration parameters for the LayoutLMLMPredictionHead.
            It is expected to be an instance of a class that holds information such as hidden size and vocabulary size.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__()
    self.transform = LayoutLMPredictionHeadTransform(config)

    # The output weights are the same as the input embeddings, but there is
    # an output-only bias for each token.
    self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)

    self.bias = mindspore.Parameter(ops.zeros(config.vocab_size))

    # Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
    self.decoder.bias = self.bias

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLMPredictionHead.forward(hidden_states)

Constructs the LayoutLMLMPredictionHead by transforming and decoding hidden states.

PARAMETER DESCRIPTION
self

An instance of the LayoutLMLMPredictionHead class.

TYPE: LayoutLMLMPredictionHead

hidden_states

The input hidden states to be processed by the prediction head.

TYPE: Tensor

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
def forward(self, hidden_states):
    """
    Constructs the LayoutLMLMPredictionHead by transforming and decoding hidden states.

    Args:
        self (LayoutLMLMPredictionHead): An instance of the LayoutLMLMPredictionHead class.
        hidden_states (torch.Tensor): The input hidden states to be processed by the prediction head.

    Returns:
        None.

    Raises:
        None
    """
    hidden_states = self.transform(hidden_states)
    hidden_states = self.decoder(hidden_states)
    return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLayer

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertLayer with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
class LayoutLMLayer(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertLayer with Bert->LayoutLM"""
    def __init__(self, config):
        """
        This method initializes an instance of the LayoutLMLayer class.

        Args:
            self (LayoutLMLayer): The instance of the LayoutLMLayer class.
            config: A configuration object containing parameters for the LayoutLMLayer.

        Returns:
            None.

        Raises:
            ValueError: If the cross attention is added and the model is not used as a decoder, a ValueError is raised.
        """
        super().__init__()
        self.chunk_size_feed_forward = config.chunk_size_feed_forward
        self.seq_len_dim = 1
        self.attention = LayoutLMAttention(config)
        self.is_decoder = config.is_decoder
        self.add_cross_attention = config.add_cross_attention
        if self.add_cross_attention:
            if not self.is_decoder:
                raise ValueError(f"{self} should be used as a decoder model if cross attention is added")
            self.crossattention = LayoutLMAttention(config, position_embedding_type="absolute")
        self.intermediate = LayoutLMIntermediate(config)
        self.output = LayoutLMOutput(config)

    def forward(
        self,
        hidden_states: mindspore.Tensor,
        attention_mask: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        encoder_hidden_states: Optional[mindspore.Tensor] = None,
        encoder_attention_mask: Optional[mindspore.Tensor] = None,
        past_key_value: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
        output_attentions: Optional[bool] = False,
    ) -> Tuple[mindspore.Tensor]:
        """
        Constructs the LayoutLMLayer.

        Args:
            self: The object instance.
            hidden_states (mindspore.Tensor): The input hidden states tensor of shape
                (batch_size, seq_length, hidden_size).
            attention_mask (Optional[mindspore.Tensor]): The attention mask tensor of shape
                (batch_size, seq_length) or (batch_size, seq_length, seq_length). Defaults to None.
            head_mask (Optional[mindspore.Tensor]): The head mask tensor of shape (num_heads,) or
                (num_layers, num_heads), where num_heads and num_layers are derived from the configuration.
                Defaults to None.
            encoder_hidden_states (Optional[mindspore.Tensor]): The encoder hidden states tensor of shape
                (batch_size, seq_length, hidden_size). Defaults to None.
            encoder_attention_mask (Optional[mindspore.Tensor]): The encoder attention mask tensor of shape
                (batch_size, seq_length) or (batch_size, seq_length, seq_length). Defaults to None.
            past_key_value (Optional[Tuple[Tuple[mindspore.Tensor]]]): The past key-value tensor of shape
                (2, batch_size, num_heads, past_seq_length, head_dim), where past_seq_length is the length of past
                sequence. Defaults to None.
            output_attentions (Optional[bool]): Whether to output attentions. Defaults to False.

        Returns:
            Tuple[mindspore.Tensor]: A tuple containing the output tensor(s) of the layer.
                The first element is the layer output tensor of shape (batch_size, seq_length, hidden_size).
                If the layer is a decoder, the tuple also includes the present key-value tensor(s) of shape
                (2, batch_size, num_heads, seq_length, head_dim).

        Raises:
            ValueError: If `encoder_hidden_states` are passed and the cross-attention layers are not instantiated
                by setting `config.add_cross_attention=True`.

        """
        # decoder uni-directional self-attention cached key/values tuple is at positions 1,2
        self_attn_past_key_value = past_key_value[:2] if past_key_value is not None else None
        self_attention_outputs = self.attention(
            hidden_states,
            attention_mask,
            head_mask,
            output_attentions=output_attentions,
            past_key_value=self_attn_past_key_value,
        )
        attention_output = self_attention_outputs[0]

        # if decoder, the last output is tuple of self-attn cache
        if self.is_decoder:
            outputs = self_attention_outputs[1:-1]
            present_key_value = self_attention_outputs[-1]
        else:
            outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights

        cross_attn_present_key_value = None
        if self.is_decoder and encoder_hidden_states is not None:
            if not hasattr(self, "crossattention"):
                raise ValueError(
                    f"If `encoder_hidden_states` are passed, {self} has to be instantiated with cross-attention layers"
                    " by setting `config.add_cross_attention=True`"
                )

            # cross_attn cached key/values tuple is at positions 3,4 of past_key_value tuple
            cross_attn_past_key_value = past_key_value[-2:] if past_key_value is not None else None
            cross_attention_outputs = self.crossattention(
                attention_output,
                attention_mask,
                head_mask,
                encoder_hidden_states,
                encoder_attention_mask,
                cross_attn_past_key_value,
                output_attentions,
            )
            attention_output = cross_attention_outputs[0]
            outputs = outputs + cross_attention_outputs[1:-1]  # add cross attentions if we output attention weights

            # add cross-attn cache to positions 3,4 of present_key_value tuple
            cross_attn_present_key_value = cross_attention_outputs[-1]
            present_key_value = present_key_value + cross_attn_present_key_value

        layer_output = apply_chunking_to_forward(
            self.feed_forward_chunk, self.chunk_size_feed_forward, self.seq_len_dim, attention_output
        )
        outputs = (layer_output,) + outputs

        # if decoder, return the attn key/values as the last output
        if self.is_decoder:
            outputs = outputs + (present_key_value,)

        return outputs

    def feed_forward_chunk(self, attention_output):
        """
        Class: LayoutLMLayer

        Description:
            This class represents a layer in a layout LM model.

        Method:
            feed_forward_chunk

        Description:
            This method applies a feed-forward operation to the given attention output.

        Args:
            self (LayoutLMLayer): The instance of the LayoutLMLayer class.
            attention_output (Tensor): The attention output tensor to be processed by the feed-forward operation.

        Returns:
            layer_output (Tensor): The output tensor after applying the feed-forward operation.

        Raises:
            None.
        """
        intermediate_output = self.intermediate(attention_output)
        layer_output = self.output(intermediate_output, attention_output)
        return layer_output

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLayer.__init__(config)

This method initializes an instance of the LayoutLMLayer class.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMLayer class.

TYPE: LayoutLMLayer

config

A configuration object containing parameters for the LayoutLMLayer.
RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
ValueError

If the cross attention is added and the model is not used as a decoder, a ValueError is raised.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
def __init__(self, config):
    """
    This method initializes an instance of the LayoutLMLayer class.

    Args:
        self (LayoutLMLayer): The instance of the LayoutLMLayer class.
        config: A configuration object containing parameters for the LayoutLMLayer.

    Returns:
        None.

    Raises:
        ValueError: If the cross attention is added and the model is not used as a decoder, a ValueError is raised.
    """
    super().__init__()
    self.chunk_size_feed_forward = config.chunk_size_feed_forward
    self.seq_len_dim = 1
    self.attention = LayoutLMAttention(config)
    self.is_decoder = config.is_decoder
    self.add_cross_attention = config.add_cross_attention
    if self.add_cross_attention:
        if not self.is_decoder:
            raise ValueError(f"{self} should be used as a decoder model if cross attention is added")
        self.crossattention = LayoutLMAttention(config, position_embedding_type="absolute")
    self.intermediate = LayoutLMIntermediate(config)
    self.output = LayoutLMOutput(config)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLayer.feed_forward_chunk(attention_output)

Description

This class represents a layer in a layout LM model.

Method

feed_forward_chunk

Description

This method applies a feed-forward operation to the given attention output.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMLayer class.

TYPE: LayoutLMLayer

attention_output

The attention output tensor to be processed by the feed-forward operation.

TYPE: Tensor

RETURNS DESCRIPTION
layer_output

The output tensor after applying the feed-forward operation.

TYPE: Tensor

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
def feed_forward_chunk(self, attention_output):
    """
    Class: LayoutLMLayer

    Description:
        This class represents a layer in a layout LM model.

    Method:
        feed_forward_chunk

    Description:
        This method applies a feed-forward operation to the given attention output.

    Args:
        self (LayoutLMLayer): The instance of the LayoutLMLayer class.
        attention_output (Tensor): The attention output tensor to be processed by the feed-forward operation.

    Returns:
        layer_output (Tensor): The output tensor after applying the feed-forward operation.

    Raises:
        None.
    """
    intermediate_output = self.intermediate(attention_output)
    layer_output = self.output(intermediate_output, attention_output)
    return layer_output

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMLayer.forward(hidden_states, attention_mask=None, head_mask=None, encoder_hidden_states=None, encoder_attention_mask=None, past_key_value=None, output_attentions=False)

Constructs the LayoutLMLayer.

PARAMETER DESCRIPTION
self

The object instance.
hidden_states

The input hidden states tensor of shape (batch_size, seq_length, hidden_size).

TYPE: Tensor

attention_mask

The attention mask tensor of shape (batch_size, seq_length) or (batch_size, seq_length, seq_length). Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

head_mask

The head mask tensor of shape (num_heads,) or (num_layers, num_heads), where num_heads and num_layers are derived from the configuration. Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

encoder_hidden_states

The encoder hidden states tensor of shape (batch_size, seq_length, hidden_size). Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

encoder_attention_mask

The encoder attention mask tensor of shape (batch_size, seq_length) or (batch_size, seq_length, seq_length). Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

past_key_value

The past key-value tensor of shape (2, batch_size, num_heads, past_seq_length, head_dim), where past_seq_length is the length of past sequence. Defaults to None.

TYPE: Optional[Tuple[Tuple[Tensor]]] DEFAULT: None

output_attentions

Whether to output attentions. Defaults to False.

TYPE: Optional[bool] DEFAULT: False

RETURNS DESCRIPTION
Tuple[Tensor]

Tuple[mindspore.Tensor]: A tuple containing the output tensor(s) of the layer. The first element is the layer output tensor of shape (batch_size, seq_length, hidden_size). If the layer is a decoder, the tuple also includes the present key-value tensor(s) of shape (2, batch_size, num_heads, seq_length, head_dim).
RAISES DESCRIPTION
ValueError

If encoder_hidden_states are passed and the cross-attention layers are not instantiated by setting config.add_cross_attention=True.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
def forward(
    self,
    hidden_states: mindspore.Tensor,
    attention_mask: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    encoder_hidden_states: Optional[mindspore.Tensor] = None,
    encoder_attention_mask: Optional[mindspore.Tensor] = None,
    past_key_value: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
    output_attentions: Optional[bool] = False,
) -> Tuple[mindspore.Tensor]:
    """
    Constructs the LayoutLMLayer.

    Args:
        self: The object instance.
        hidden_states (mindspore.Tensor): The input hidden states tensor of shape
            (batch_size, seq_length, hidden_size).
        attention_mask (Optional[mindspore.Tensor]): The attention mask tensor of shape
            (batch_size, seq_length) or (batch_size, seq_length, seq_length). Defaults to None.
        head_mask (Optional[mindspore.Tensor]): The head mask tensor of shape (num_heads,) or
            (num_layers, num_heads), where num_heads and num_layers are derived from the configuration.
            Defaults to None.
        encoder_hidden_states (Optional[mindspore.Tensor]): The encoder hidden states tensor of shape
            (batch_size, seq_length, hidden_size). Defaults to None.
        encoder_attention_mask (Optional[mindspore.Tensor]): The encoder attention mask tensor of shape
            (batch_size, seq_length) or (batch_size, seq_length, seq_length). Defaults to None.
        past_key_value (Optional[Tuple[Tuple[mindspore.Tensor]]]): The past key-value tensor of shape
            (2, batch_size, num_heads, past_seq_length, head_dim), where past_seq_length is the length of past
            sequence. Defaults to None.
        output_attentions (Optional[bool]): Whether to output attentions. Defaults to False.

    Returns:
        Tuple[mindspore.Tensor]: A tuple containing the output tensor(s) of the layer.
            The first element is the layer output tensor of shape (batch_size, seq_length, hidden_size).
            If the layer is a decoder, the tuple also includes the present key-value tensor(s) of shape
            (2, batch_size, num_heads, seq_length, head_dim).

    Raises:
        ValueError: If `encoder_hidden_states` are passed and the cross-attention layers are not instantiated
            by setting `config.add_cross_attention=True`.

    """
    # decoder uni-directional self-attention cached key/values tuple is at positions 1,2
    self_attn_past_key_value = past_key_value[:2] if past_key_value is not None else None
    self_attention_outputs = self.attention(
        hidden_states,
        attention_mask,
        head_mask,
        output_attentions=output_attentions,
        past_key_value=self_attn_past_key_value,
    )
    attention_output = self_attention_outputs[0]

    # if decoder, the last output is tuple of self-attn cache
    if self.is_decoder:
        outputs = self_attention_outputs[1:-1]
        present_key_value = self_attention_outputs[-1]
    else:
        outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights

    cross_attn_present_key_value = None
    if self.is_decoder and encoder_hidden_states is not None:
        if not hasattr(self, "crossattention"):
            raise ValueError(
                f"If `encoder_hidden_states` are passed, {self} has to be instantiated with cross-attention layers"
                " by setting `config.add_cross_attention=True`"
            )

        # cross_attn cached key/values tuple is at positions 3,4 of past_key_value tuple
        cross_attn_past_key_value = past_key_value[-2:] if past_key_value is not None else None
        cross_attention_outputs = self.crossattention(
            attention_output,
            attention_mask,
            head_mask,
            encoder_hidden_states,
            encoder_attention_mask,
            cross_attn_past_key_value,
            output_attentions,
        )
        attention_output = cross_attention_outputs[0]
        outputs = outputs + cross_attention_outputs[1:-1]  # add cross attentions if we output attention weights

        # add cross-attn cache to positions 3,4 of present_key_value tuple
        cross_attn_present_key_value = cross_attention_outputs[-1]
        present_key_value = present_key_value + cross_attn_present_key_value

    layer_output = apply_chunking_to_forward(
        self.feed_forward_chunk, self.chunk_size_feed_forward, self.seq_len_dim, attention_output
    )
    outputs = (layer_output,) + outputs

    # if decoder, return the attn key/values as the last output
    if self.is_decoder:
        outputs = outputs + (present_key_value,)

    return outputs

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMModel

Bases: LayoutLMPreTrainedModel

LayoutLM Model

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
class LayoutLMModel(LayoutLMPreTrainedModel):
    """LayoutLM Model"""
    def __init__(self, config):
        """
        Initializes a LayoutLMModel instance.

        Args:
            self: The instance of the LayoutLMModel class.
            config: A dictionary containing the configuration settings for the LayoutLMModel.
                The config should include parameters for initializing the LayoutLMModel,
                such as hidden size, number of layers, etc.

        Returns:
            None.

        Raises:
            None.
        """
        super(LayoutLMModel, self).__init__(config)
        self.config = config

        self.embeddings = LayoutLMEmbeddings(config)
        self.encoder = LayoutLMEncoder(config)
        self.pooler = LayoutLMPooler(config)

        # Initialize weights and apply final processing
        self.post_init()

    def get_input_embeddings(self):
        """
        Retrieves the input embeddings from the LayoutLMModel.

        Args:
            self (LayoutLMModel): The LayoutLMModel instance.

        Returns:
            None.

        Raises:
            None.
        """
        return self.embeddings.word_embeddings

    def set_input_embeddings(self, value):
        """
        Sets the input embeddings for the LayoutLMModel.

        Args:
            self (LayoutLMModel): The LayoutLMModel instance.
            value: The input embeddings to be set.
                It should be of type torch.Tensor and have the same shape as the word_embeddings.

        Returns:
            None

        Raises:
            None
        """
        self.embeddings.word_embeddings = value

    def _prune_heads(self, heads_to_prune):
        """
        Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
        class PreTrainedModel
        """
        for layer, heads in heads_to_prune.items():
            self.encoder.layer[layer].attention.prune_heads(heads)

    def forward(
        self,
        input_ids: Optional[mindspore.Tensor] = None,
        bbox: Optional[mindspore.Tensor] = None,
        attention_mask: Optional[mindspore.Tensor] = None,
        token_type_ids: Optional[mindspore.Tensor] = None,
        position_ids: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        inputs_embeds: Optional[mindspore.Tensor] = None,
        encoder_hidden_states: Optional[mindspore.Tensor] = None,
        encoder_attention_mask: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]:
        r"""

        Returns:
            Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]

        Example:
            ```python
            >>> from transformers import AutoTokenizer, LayoutLMModel
            >>> import torch
            ...
            >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
            >>> model = LayoutLMModel.from_pretrained("microsoft/layoutlm-base-uncased")
            ...
            >>> words = ["Hello", "world"]
            >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
            ...
            >>> token_boxes = []
            >>> for word, box in zip(words, normalized_word_boxes):
            ...     word_tokens = tokenizer.tokenize(word)
            ...     token_boxes.extend([box] * len(word_tokens))
            >>> # add bounding boxes of cls + sep tokens
            >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
            ...
            >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
            >>> input_ids = encoding["input_ids"]
            >>> attention_mask = encoding["attention_mask"]
            >>> token_type_ids = encoding["token_type_ids"]
            >>> bbox = torch.tensor([token_boxes])
            ...
            >>> outputs = model(
            ...     input_ids=input_ids, bbox=bbox, attention_mask=attention_mask, token_type_ids=token_type_ids
            ... )
            ...
            >>> last_hidden_states = outputs.last_hidden_state
            ```
        """
        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
        output_hidden_states = (
            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
        )
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        if input_ids is not None and inputs_embeds is not None:
            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
        elif input_ids is not None:
            self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
            input_shape = input_ids.shape
        elif inputs_embeds is not None:
            input_shape = inputs_embeds.shape[:-1]
        else:
            raise ValueError("You have to specify either input_ids or inputs_embeds")

        if attention_mask is None:
            attention_mask = ops.ones(input_shape)
        if token_type_ids is None:
            token_type_ids = ops.zeros(input_shape, dtype=mindspore.int64)

        if bbox is None:
            bbox = ops.zeros(input_shape + (4,), dtype=mindspore.int64)

        extended_attention_mask = attention_mask.unsqueeze(1).unsqueeze(2)

        extended_attention_mask = extended_attention_mask.to(dtype=self.dtype)
        extended_attention_mask = (1.0 - extended_attention_mask) * finfo(self.dtype, 'min')

        if head_mask is not None:
            if head_mask.ndimension() == 1:
                head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
                head_mask = head_mask.broadcast_to((self.config.num_hidden_layers, -1, -1, -1, -1))
            elif head_mask.ndimension() == 2:
                head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)
            head_mask = head_mask.to(dtype=next(self.get_parameters()).dtype)
        else:
            head_mask = [None] * self.config.num_hidden_layers

        embedding_output = self.embeddings(
            input_ids=input_ids,
            bbox=bbox,
            position_ids=position_ids,
            token_type_ids=token_type_ids,
            inputs_embeds=inputs_embeds,
        )
        encoder_outputs = self.encoder(
            embedding_output,
            extended_attention_mask,
            head_mask=head_mask,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        sequence_output = encoder_outputs[0]
        pooled_output = self.pooler(sequence_output)

        if not return_dict:
            return (sequence_output, pooled_output) + encoder_outputs[1:]

        return BaseModelOutputWithPoolingAndCrossAttentions(
            last_hidden_state=sequence_output,
            pooler_output=pooled_output,
            hidden_states=encoder_outputs.hidden_states,
            attentions=encoder_outputs.attentions,
            cross_attentions=encoder_outputs.cross_attentions,
        )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMModel.__init__(config)

Initializes a LayoutLMModel instance.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMModel class.
config

A dictionary containing the configuration settings for the LayoutLMModel. The config should include parameters for initializing the LayoutLMModel, such as hidden size, number of layers, etc.
RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
def __init__(self, config):
    """
    Initializes a LayoutLMModel instance.

    Args:
        self: The instance of the LayoutLMModel class.
        config: A dictionary containing the configuration settings for the LayoutLMModel.
            The config should include parameters for initializing the LayoutLMModel,
            such as hidden size, number of layers, etc.

    Returns:
        None.

    Raises:
        None.
    """
    super(LayoutLMModel, self).__init__(config)
    self.config = config

    self.embeddings = LayoutLMEmbeddings(config)
    self.encoder = LayoutLMEncoder(config)
    self.pooler = LayoutLMPooler(config)

    # Initialize weights and apply final processing
    self.post_init()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMModel.forward(input_ids=None, bbox=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, encoder_hidden_states=None, encoder_attention_mask=None, output_attentions=None, output_hidden_states=None, return_dict=None)

RETURNS DESCRIPTION
Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]

Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]
Example 
>>> from transformers import AutoTokenizer, LayoutLMModel
>>> import torch
...
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
>>> model = LayoutLMModel.from_pretrained("microsoft/layoutlm-base-uncased")
...
>>> words = ["Hello", "world"]
>>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
...
>>> token_boxes = []
>>> for word, box in zip(words, normalized_word_boxes):
...     word_tokens = tokenizer.tokenize(word)
...     token_boxes.extend([box] * len(word_tokens))
>>> # add bounding boxes of cls + sep tokens
>>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
...
>>> encoding = tokenizer(" ".join(words), return_tensors="pt")
>>> input_ids = encoding["input_ids"]
>>> attention_mask = encoding["attention_mask"]
>>> token_type_ids = encoding["token_type_ids"]
>>> bbox = torch.tensor([token_boxes])
...
>>> outputs = model(
...     input_ids=input_ids, bbox=bbox, attention_mask=attention_mask, token_type_ids=token_type_ids
... )
...
>>> last_hidden_states = outputs.last_hidden_state
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
def forward(
    self,
    input_ids: Optional[mindspore.Tensor] = None,
    bbox: Optional[mindspore.Tensor] = None,
    attention_mask: Optional[mindspore.Tensor] = None,
    token_type_ids: Optional[mindspore.Tensor] = None,
    position_ids: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    inputs_embeds: Optional[mindspore.Tensor] = None,
    encoder_hidden_states: Optional[mindspore.Tensor] = None,
    encoder_attention_mask: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]:
    r"""

    Returns:
        Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]

    Example:
        ```python
        >>> from transformers import AutoTokenizer, LayoutLMModel
        >>> import torch
        ...
        >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/layoutlm-base-uncased")
        >>> model = LayoutLMModel.from_pretrained("microsoft/layoutlm-base-uncased")
        ...
        >>> words = ["Hello", "world"]
        >>> normalized_word_boxes = [637, 773, 693, 782], [698, 773, 733, 782]
        ...
        >>> token_boxes = []
        >>> for word, box in zip(words, normalized_word_boxes):
        ...     word_tokens = tokenizer.tokenize(word)
        ...     token_boxes.extend([box] * len(word_tokens))
        >>> # add bounding boxes of cls + sep tokens
        >>> token_boxes = [[0, 0, 0, 0]] + token_boxes + [[1000, 1000, 1000, 1000]]
        ...
        >>> encoding = tokenizer(" ".join(words), return_tensors="pt")
        >>> input_ids = encoding["input_ids"]
        >>> attention_mask = encoding["attention_mask"]
        >>> token_type_ids = encoding["token_type_ids"]
        >>> bbox = torch.tensor([token_boxes])
        ...
        >>> outputs = model(
        ...     input_ids=input_ids, bbox=bbox, attention_mask=attention_mask, token_type_ids=token_type_ids
        ... )
        ...
        >>> last_hidden_states = outputs.last_hidden_state
        ```
    """
    output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
    output_hidden_states = (
        output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
    )
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    if input_ids is not None and inputs_embeds is not None:
        raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
    elif input_ids is not None:
        self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
        input_shape = input_ids.shape
    elif inputs_embeds is not None:
        input_shape = inputs_embeds.shape[:-1]
    else:
        raise ValueError("You have to specify either input_ids or inputs_embeds")

    if attention_mask is None:
        attention_mask = ops.ones(input_shape)
    if token_type_ids is None:
        token_type_ids = ops.zeros(input_shape, dtype=mindspore.int64)

    if bbox is None:
        bbox = ops.zeros(input_shape + (4,), dtype=mindspore.int64)

    extended_attention_mask = attention_mask.unsqueeze(1).unsqueeze(2)

    extended_attention_mask = extended_attention_mask.to(dtype=self.dtype)
    extended_attention_mask = (1.0 - extended_attention_mask) * finfo(self.dtype, 'min')

    if head_mask is not None:
        if head_mask.ndimension() == 1:
            head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
            head_mask = head_mask.broadcast_to((self.config.num_hidden_layers, -1, -1, -1, -1))
        elif head_mask.ndimension() == 2:
            head_mask = head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1)
        head_mask = head_mask.to(dtype=next(self.get_parameters()).dtype)
    else:
        head_mask = [None] * self.config.num_hidden_layers

    embedding_output = self.embeddings(
        input_ids=input_ids,
        bbox=bbox,
        position_ids=position_ids,
        token_type_ids=token_type_ids,
        inputs_embeds=inputs_embeds,
    )
    encoder_outputs = self.encoder(
        embedding_output,
        extended_attention_mask,
        head_mask=head_mask,
        output_attentions=output_attentions,
        output_hidden_states=output_hidden_states,
        return_dict=return_dict,
    )

    sequence_output = encoder_outputs[0]
    pooled_output = self.pooler(sequence_output)

    if not return_dict:
        return (sequence_output, pooled_output) + encoder_outputs[1:]

    return BaseModelOutputWithPoolingAndCrossAttentions(
        last_hidden_state=sequence_output,
        pooler_output=pooled_output,
        hidden_states=encoder_outputs.hidden_states,
        attentions=encoder_outputs.attentions,
        cross_attentions=encoder_outputs.cross_attentions,
    )

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMModel.get_input_embeddings()

Retrieves the input embeddings from the LayoutLMModel.

PARAMETER DESCRIPTION
self

The LayoutLMModel instance.

TYPE: LayoutLMModel

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
def get_input_embeddings(self):
    """
    Retrieves the input embeddings from the LayoutLMModel.

    Args:
        self (LayoutLMModel): The LayoutLMModel instance.

    Returns:
        None.

    Raises:
        None.
    """
    return self.embeddings.word_embeddings

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMModel.set_input_embeddings(value)

Sets the input embeddings for the LayoutLMModel.

PARAMETER DESCRIPTION
self

The LayoutLMModel instance.

TYPE: LayoutLMModel

value

The input embeddings to be set. It should be of type torch.Tensor and have the same shape as the word_embeddings.
RETURNS DESCRIPTION

None
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
def set_input_embeddings(self, value):
    """
    Sets the input embeddings for the LayoutLMModel.

    Args:
        self (LayoutLMModel): The LayoutLMModel instance.
        value: The input embeddings to be set.
            It should be of type torch.Tensor and have the same shape as the word_embeddings.

    Returns:
        None

    Raises:
        None
    """
    self.embeddings.word_embeddings = value

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMOnlyMLMHead

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertOnlyMLMHead with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
class LayoutLMOnlyMLMHead(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertOnlyMLMHead with Bert->LayoutLM"""
    def __init__(self, config):
        """
        Initializes a LayoutLMOnlyMLMHead object.

        Args:
            self (LayoutLMOnlyMLMHead): The current instance of the LayoutLMOnlyMLMHead class.
            config: The configuration parameters for the LayoutLMOnlyMLMHead.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__()
        self.predictions = LayoutLMLMPredictionHead(config)

    def forward(self, sequence_output: mindspore.Tensor) -> mindspore.Tensor:
        """
        Constructs the LayoutLMOnlyMLMHead.

        This method takes two parameters: self and sequence_output.

        Args:
            self: An instance of the LayoutLMOnlyMLMHead class.
            sequence_output (mindspore.Tensor): The output tensor from the sequence modeling layer.
                It is the input to the prediction layer.

        Returns:
            mindspore.Tensor: The prediction scores tensor generated by the prediction layer.
                It represents the predicted scores for each token in the input sequence.

        Raises:
            None.
        """
        prediction_scores = self.predictions(sequence_output)
        return prediction_scores

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMOnlyMLMHead.__init__(config)

Initializes a LayoutLMOnlyMLMHead object.

PARAMETER DESCRIPTION
self

The current instance of the LayoutLMOnlyMLMHead class.

TYPE: LayoutLMOnlyMLMHead

config

The configuration parameters for the LayoutLMOnlyMLMHead.
RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
def __init__(self, config):
    """
    Initializes a LayoutLMOnlyMLMHead object.

    Args:
        self (LayoutLMOnlyMLMHead): The current instance of the LayoutLMOnlyMLMHead class.
        config: The configuration parameters for the LayoutLMOnlyMLMHead.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__()
    self.predictions = LayoutLMLMPredictionHead(config)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMOnlyMLMHead.forward(sequence_output)

Constructs the LayoutLMOnlyMLMHead.

This method takes two parameters: self and sequence_output.

PARAMETER DESCRIPTION
self

An instance of the LayoutLMOnlyMLMHead class.
sequence_output

The output tensor from the sequence modeling layer. It is the input to the prediction layer.

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: The prediction scores tensor generated by the prediction layer. It represents the predicted scores for each token in the input sequence.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
def forward(self, sequence_output: mindspore.Tensor) -> mindspore.Tensor:
    """
    Constructs the LayoutLMOnlyMLMHead.

    This method takes two parameters: self and sequence_output.

    Args:
        self: An instance of the LayoutLMOnlyMLMHead class.
        sequence_output (mindspore.Tensor): The output tensor from the sequence modeling layer.
            It is the input to the prediction layer.

    Returns:
        mindspore.Tensor: The prediction scores tensor generated by the prediction layer.
            It represents the predicted scores for each token in the input sequence.

    Raises:
        None.
    """
    prediction_scores = self.predictions(sequence_output)
    return prediction_scores

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMOutput

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertOutput with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
class LayoutLMOutput(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertOutput with Bert->LayoutLM"""
    def __init__(self, config):
        """
        Initializes a new instance of the LayoutLMOutput class.

        Args:
            self (object): The instance of the class.
            config (object): An object containing configuration parameters for the LayoutLMOutput.
                This parameter is required to configure the dense layer, layer normalization, and dropout.
                It should be an instance of a class that contains the following attributes:

                - intermediate_size (int): The size of the intermediate layer.
                - hidden_size (int): The size of the hidden layer.
                - layer_norm_eps (float): The epsilon value for layer normalization.
                - hidden_dropout_prob (float): The dropout probability for the hidden layer.

        Returns:
            None.

        Raises:
            TypeError: If the config parameter is not provided or is not an instance of the expected class.
            ValueError: If the attributes intermediate_size, hidden_size, layer_norm_eps, or hidden_dropout_prob
                are missing from the config object.
        """
        super().__init__()
        self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
        self.dropout = nn.Dropout(p=config.hidden_dropout_prob)

    def forward(self, hidden_states: mindspore.Tensor, input_tensor: mindspore.Tensor) -> mindspore.Tensor:
        """
        Construct method in the LayoutLMOutput class.

        Args:
            self: LayoutLMOutput instance.
            hidden_states (mindspore.Tensor): The hidden states tensor to be processed.
            input_tensor (mindspore.Tensor): The input tensor to be added to the processed hidden states.

        Returns:
            mindspore.Tensor: A tensor representing the processed hidden states with the input tensor added.

        Raises:
            None
        """
        hidden_states = self.dense(hidden_states)
        hidden_states = self.dropout(hidden_states)
        hidden_states = self.LayerNorm(hidden_states + input_tensor)
        return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMOutput.__init__(config)

Initializes a new instance of the LayoutLMOutput class.

PARAMETER DESCRIPTION
self

The instance of the class.

TYPE: object

config

An object containing configuration parameters for the LayoutLMOutput. This parameter is required to configure the dense layer, layer normalization, and dropout. It should be an instance of a class that contains the following attributes:

  • intermediate_size (int): The size of the intermediate layer.
  • hidden_size (int): The size of the hidden layer.
  • layer_norm_eps (float): The epsilon value for layer normalization.
  • hidden_dropout_prob (float): The dropout probability for the hidden layer.

TYPE: object

RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
TypeError

If the config parameter is not provided or is not an instance of the expected class.
ValueError

If the attributes intermediate_size, hidden_size, layer_norm_eps, or hidden_dropout_prob are missing from the config object.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
def __init__(self, config):
    """
    Initializes a new instance of the LayoutLMOutput class.

    Args:
        self (object): The instance of the class.
        config (object): An object containing configuration parameters for the LayoutLMOutput.
            This parameter is required to configure the dense layer, layer normalization, and dropout.
            It should be an instance of a class that contains the following attributes:

            - intermediate_size (int): The size of the intermediate layer.
            - hidden_size (int): The size of the hidden layer.
            - layer_norm_eps (float): The epsilon value for layer normalization.
            - hidden_dropout_prob (float): The dropout probability for the hidden layer.

    Returns:
        None.

    Raises:
        TypeError: If the config parameter is not provided or is not an instance of the expected class.
        ValueError: If the attributes intermediate_size, hidden_size, layer_norm_eps, or hidden_dropout_prob
            are missing from the config object.
    """
    super().__init__()
    self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
    self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
    self.dropout = nn.Dropout(p=config.hidden_dropout_prob)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMOutput.forward(hidden_states, input_tensor)

Construct method in the LayoutLMOutput class.

PARAMETER DESCRIPTION
self

LayoutLMOutput instance.
hidden_states

The hidden states tensor to be processed.

TYPE: Tensor

input_tensor

The input tensor to be added to the processed hidden states.

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: A tensor representing the processed hidden states with the input tensor added.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
def forward(self, hidden_states: mindspore.Tensor, input_tensor: mindspore.Tensor) -> mindspore.Tensor:
    """
    Construct method in the LayoutLMOutput class.

    Args:
        self: LayoutLMOutput instance.
        hidden_states (mindspore.Tensor): The hidden states tensor to be processed.
        input_tensor (mindspore.Tensor): The input tensor to be added to the processed hidden states.

    Returns:
        mindspore.Tensor: A tensor representing the processed hidden states with the input tensor added.

    Raises:
        None
    """
    hidden_states = self.dense(hidden_states)
    hidden_states = self.dropout(hidden_states)
    hidden_states = self.LayerNorm(hidden_states + input_tensor)
    return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPooler

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertPooler

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
class LayoutLMPooler(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertPooler"""
    def __init__(self, config):
        """
        Initializes a LayoutLMPooler instance.

        Args:
            self: The instance of LayoutLMPooler.
            config: The configuration object containing parameters for the LayoutLMPooler initialization.
                It should be an instance of the Config class.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__()
        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
        self.activation = nn.Tanh()

    def forward(self, hidden_states: mindspore.Tensor) -> mindspore.Tensor:
        """
        This method 'forward' in the class 'LayoutLMPooler' forwards a pooled output tensor based on
        the hidden states provided.

        Args:
            self (LayoutLMPooler): The instance of the LayoutLMPooler class.
            hidden_states (mindspore.Tensor): The input tensor containing hidden states.
                It should have the shape (batch_size, sequence_length, hidden_size).
                This tensor holds the hidden states generated by the model for each token in the input sequence.

        Returns:
            mindspore.Tensor: A tensor representing the pooled output. It is the result of applying dense and
                activation layers on the first token's hidden state.
                The shape of the returned tensor is (batch_size, hidden_size).

        Raises:
            None.
        """
        # We "pool" the model by simply taking the hidden state corresponding
        # to the first token.
        first_token_tensor = hidden_states[:, 0]
        pooled_output = self.dense(first_token_tensor)
        pooled_output = self.activation(pooled_output)
        return pooled_output

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPooler.__init__(config)

Initializes a LayoutLMPooler instance.

PARAMETER DESCRIPTION
self

The instance of LayoutLMPooler.
config

The configuration object containing parameters for the LayoutLMPooler initialization. It should be an instance of the Config class.
RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
def __init__(self, config):
    """
    Initializes a LayoutLMPooler instance.

    Args:
        self: The instance of LayoutLMPooler.
        config: The configuration object containing parameters for the LayoutLMPooler initialization.
            It should be an instance of the Config class.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__()
    self.dense = nn.Linear(config.hidden_size, config.hidden_size)
    self.activation = nn.Tanh()

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPooler.forward(hidden_states)

This method 'forward' in the class 'LayoutLMPooler' forwards a pooled output tensor based on the hidden states provided.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMPooler class.

TYPE: LayoutLMPooler

hidden_states

The input tensor containing hidden states. It should have the shape (batch_size, sequence_length, hidden_size). This tensor holds the hidden states generated by the model for each token in the input sequence.

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: A tensor representing the pooled output. It is the result of applying dense and activation layers on the first token's hidden state. The shape of the returned tensor is (batch_size, hidden_size).
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
def forward(self, hidden_states: mindspore.Tensor) -> mindspore.Tensor:
    """
    This method 'forward' in the class 'LayoutLMPooler' forwards a pooled output tensor based on
    the hidden states provided.

    Args:
        self (LayoutLMPooler): The instance of the LayoutLMPooler class.
        hidden_states (mindspore.Tensor): The input tensor containing hidden states.
            It should have the shape (batch_size, sequence_length, hidden_size).
            This tensor holds the hidden states generated by the model for each token in the input sequence.

    Returns:
        mindspore.Tensor: A tensor representing the pooled output. It is the result of applying dense and
            activation layers on the first token's hidden state.
            The shape of the returned tensor is (batch_size, hidden_size).

    Raises:
        None.
    """
    # We "pool" the model by simply taking the hidden state corresponding
    # to the first token.
    first_token_tensor = hidden_states[:, 0]
    pooled_output = self.dense(first_token_tensor)
    pooled_output = self.activation(pooled_output)
    return pooled_output

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPreTrainedModel

Bases: PreTrainedModel

An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained models.

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
class LayoutLMPreTrainedModel(PreTrainedModel):
    """
    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
    models.
    """
    config_class = LayoutLMConfig
    pretrained_model_archive_map = LAYOUTLM_PRETRAINED_MODEL_ARCHIVE_LIST
    base_model_prefix = "layoutlm"
    supports_gradient_checkpointing = True

    def _init_weights(self, cell):
        """Initialize the weights"""
        if isinstance(cell, nn.Linear):
            # Slightly different from the TF version which uses truncated_normal for initialization
            # cf https://github.com/pytorch/pytorch/pull/5617
            cell.weight.set_data(initializer(Normal(self.config.initializer_range),
                                                    cell.weight.shape, cell.weight.dtype))
            if cell.bias is not None:
                cell.bias.set_data(initializer('zeros', cell.bias.shape, cell.bias.dtype))
        elif isinstance(cell, nn.Embedding):
            weight = np.random.normal(0.0, self.config.initializer_range, cell.weight.shape)
            if cell.padding_idx:
                weight[cell.padding_idx] = 0

            cell.weight.set_data(mindspore.Tensor(weight, cell.weight.dtype))
        elif isinstance(cell, LayoutLMLayerNorm):
            cell.weight.set_data(initializer('ones', cell.weight.shape, cell.weight.dtype))
            cell.bias.set_data(initializer('zeros', cell.bias.shape, cell.bias.dtype))

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPredictionHeadTransform

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertPredictionHeadTransform with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
class LayoutLMPredictionHeadTransform(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertPredictionHeadTransform with Bert->LayoutLM"""
    def __init__(self, config):
        """
        Initialize the LayoutLMPredictionHeadTransform class.

        Args:
            self: The instance of the class.
            config:
                An object containing configuration parameters for the head transformation.

                - Type: Custom configuration class
                - Purpose: Specifies the configuration settings for the head transformation.
                - Restrictions: Must be a valid configuration object.

        Returns:
            None.

        Raises:
            KeyError: If the specified 'hidden_act' in the configuration is not found in the ACT2FN dictionary.
            AttributeError: If the configuration object does not contain the required attributes.
            ValueError: If there are issues with the provided configuration parameters.
        """
        super().__init__()
        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
        if isinstance(config.hidden_act, str):
            self.transform_act_fn = ACT2FN[config.hidden_act]
        else:
            self.transform_act_fn = config.hidden_act
        self.LayerNorm = nn.LayerNorm([config.hidden_size], eps=config.layer_norm_eps)

    def forward(self, hidden_states: mindspore.Tensor) -> mindspore.Tensor:
        """
        This method 'forward' in the class 'LayoutLMPredictionHeadTransform'
        performs transformations on the input hidden states tensor.

        Args:
            self: An instance of the class 'LayoutLMPredictionHeadTransform'.
            hidden_states (mindspore.Tensor): The input tensor representing the hidden states.
                It is expected to be a tensor of shape (batch_size, sequence_length, hidden_size).

        Returns:
            mindspore.Tensor: A tensor containing the transformed hidden states after passing through dense layers,
                activation function, and layer normalization. The shape of the output tensor is the same as
                the input hidden_states.

        Raises:
            None.
        """
        hidden_states = self.dense(hidden_states)
        hidden_states = self.transform_act_fn(hidden_states)
        hidden_states = self.LayerNorm(hidden_states)
        return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPredictionHeadTransform.__init__(config)

Initialize the LayoutLMPredictionHeadTransform class.

PARAMETER DESCRIPTION
self

The instance of the class.
config

An object containing configuration parameters for the head transformation.

  • Type: Custom configuration class
  • Purpose: Specifies the configuration settings for the head transformation.
  • Restrictions: Must be a valid configuration object.
RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
KeyError

If the specified 'hidden_act' in the configuration is not found in the ACT2FN dictionary.
AttributeError

If the configuration object does not contain the required attributes.
ValueError

If there are issues with the provided configuration parameters.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
def __init__(self, config):
    """
    Initialize the LayoutLMPredictionHeadTransform class.

    Args:
        self: The instance of the class.
        config:
            An object containing configuration parameters for the head transformation.

            - Type: Custom configuration class
            - Purpose: Specifies the configuration settings for the head transformation.
            - Restrictions: Must be a valid configuration object.

    Returns:
        None.

    Raises:
        KeyError: If the specified 'hidden_act' in the configuration is not found in the ACT2FN dictionary.
        AttributeError: If the configuration object does not contain the required attributes.
        ValueError: If there are issues with the provided configuration parameters.
    """
    super().__init__()
    self.dense = nn.Linear(config.hidden_size, config.hidden_size)
    if isinstance(config.hidden_act, str):
        self.transform_act_fn = ACT2FN[config.hidden_act]
    else:
        self.transform_act_fn = config.hidden_act
    self.LayerNorm = nn.LayerNorm([config.hidden_size], eps=config.layer_norm_eps)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMPredictionHeadTransform.forward(hidden_states)

This method 'forward' in the class 'LayoutLMPredictionHeadTransform' performs transformations on the input hidden states tensor.

PARAMETER DESCRIPTION
self

An instance of the class 'LayoutLMPredictionHeadTransform'.
hidden_states

The input tensor representing the hidden states. It is expected to be a tensor of shape (batch_size, sequence_length, hidden_size).

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: A tensor containing the transformed hidden states after passing through dense layers, activation function, and layer normalization. The shape of the output tensor is the same as the input hidden_states.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
def forward(self, hidden_states: mindspore.Tensor) -> mindspore.Tensor:
    """
    This method 'forward' in the class 'LayoutLMPredictionHeadTransform'
    performs transformations on the input hidden states tensor.

    Args:
        self: An instance of the class 'LayoutLMPredictionHeadTransform'.
        hidden_states (mindspore.Tensor): The input tensor representing the hidden states.
            It is expected to be a tensor of shape (batch_size, sequence_length, hidden_size).

    Returns:
        mindspore.Tensor: A tensor containing the transformed hidden states after passing through dense layers,
            activation function, and layer normalization. The shape of the output tensor is the same as
            the input hidden_states.

    Raises:
        None.
    """
    hidden_states = self.dense(hidden_states)
    hidden_states = self.transform_act_fn(hidden_states)
    hidden_states = self.LayerNorm(hidden_states)
    return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfAttention

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertSelfAttention with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
class LayoutLMSelfAttention(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertSelfAttention with Bert->LayoutLM"""
    def __init__(self, config, position_embedding_type=None):
        """
        Initializes a LayoutLMSelfAttention instance with the provided configuration and optional position embedding type.

        Args:
            self: The instance of LayoutLMSelfAttention.
            config: An object containing the configuration parameters for the self-attention layer.
                Expected attributes:

                - hidden_size (int): The size of the hidden layer.
                - num_attention_heads (int): The number of attention heads.
                - embedding_size (int, optional): The size of the embedding layer.
                - attention_probs_dropout_prob (float): The dropout probability for attention probabilities.
                - position_embedding_type (str, optional): The type of position embedding to use ('absolute' by default).
                - max_position_embeddings (int): The maximum number of position embeddings.
                - is_decoder (bool): Indicates if the self-attention layer is part of a decoder.
            position_embedding_type (str, optional): The type of position embedding to use (default is None).
                Accepted values: 'absolute', 'relative_key', 'relative_key_query'.

        Returns:
            None.

        Raises:
            ValueError:
                If the hidden size is not a multiple of the number of attention heads and no embedding size is provided.
        """
        super().__init__()
        if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
            raise ValueError(
                f"The hidden size ({config.hidden_size}) is not a multiple of the number of attention "
                f"heads ({config.num_attention_heads})"
            )

        self.num_attention_heads = config.num_attention_heads
        self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
        self.all_head_size = self.num_attention_heads * self.attention_head_size

        self.query = nn.Linear(config.hidden_size, self.all_head_size)
        self.key = nn.Linear(config.hidden_size, self.all_head_size)
        self.value = nn.Linear(config.hidden_size, self.all_head_size)

        self.dropout = nn.Dropout(p=config.attention_probs_dropout_prob)
        self.position_embedding_type = position_embedding_type or getattr(
            config, "position_embedding_type", "absolute"
        )
        if self.position_embedding_type in ('relative_key', 'relative_key_query'):
            self.max_position_embeddings = config.max_position_embeddings
            self.distance_embedding = nn.Embedding(2 * config.max_position_embeddings - 1, self.attention_head_size)

        self.is_decoder = config.is_decoder

    def transpose_for_scores(self, x: mindspore.Tensor) -> mindspore.Tensor:
        """
        Transposes the input tensor `x` for calculating self-attention scores.

        Args:
            self (LayoutLMSelfAttention): The current instance of the LayoutLMSelfAttention class.
            x (mindspore.Tensor): The input tensor of shape `(batch_size, sequence_length, hidden_size)`.

        Returns:
            mindspore.Tensor:
                The transposed tensor of shape `(batch_size, num_attention_heads, sequence_length, attention_head_size)`.

        Raises:
            None

        This method transposes the input tensor `x` to prepare it for calculating self-attention scores in the
        LayoutLMSelfAttention model. The transposition is performed by reshaping the tensor to include the number of
        attention heads and the size of each attention head. The resulting tensor is then permuted to match the desired
        shape `(batch_size, num_attention_heads, sequence_length, attention_head_size)`.

        Note that this method assumes that the input tensor `x` has a rank of at least 3, where the last dimension
        represents the hidden size. The number of attention heads and the size of each attention head are obtained
        from the attributes `num_attention_heads` and `attention_head_size` of the LayoutLMSelfAttention instance,
        respectively.
        """
        new_x_shape = x.shape[:-1] + (self.num_attention_heads, self.attention_head_size)
        x = x.view(new_x_shape)
        return x.permute(0, 2, 1, 3)

    def forward(
        self,
        hidden_states: mindspore.Tensor,
        attention_mask: Optional[mindspore.Tensor] = None,
        head_mask: Optional[mindspore.Tensor] = None,
        encoder_hidden_states: Optional[mindspore.Tensor] = None,
        encoder_attention_mask: Optional[mindspore.Tensor] = None,
        past_key_value: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
        output_attentions: Optional[bool] = False,
    ) -> Tuple[mindspore.Tensor]:
        """
        This method forwards the self-attention mechanism for the LayoutLMSelfAttention class.

        Args:
            self: The instance of the LayoutLMSelfAttention class.
            hidden_states (mindspore.Tensor): The input hidden states tensor.
            attention_mask (Optional[mindspore.Tensor], optional): Mask tensor to prevent attention to certain
                positions. Defaults to None.
            head_mask (Optional[mindspore.Tensor], optional): Mask tensor to control the heads involved in the
                attention computation. Defaults to None.
            encoder_hidden_states (Optional[mindspore.Tensor], optional): Hidden states of the encoder in case of
                cross-attention. Defaults to None.
            encoder_attention_mask (Optional[mindspore.Tensor], optional): Mask tensor for encoder attention.
                Defaults to None.
            past_key_value (Optional[Tuple[Tuple[mindspore.Tensor]]], optional): Cached key and value tensors from
                previous attention calculations. Defaults to None.
            output_attentions (Optional[bool], optional): Flag to indicate whether to output attentions.
                Defaults to False.

        Returns:
            Tuple[mindspore.Tensor]: A tuple containing the context layer tensor and optionally attention
                probabilities tensor.

        Raises:
            ValueError: If the input tensor shapes are incompatible for matrix multiplication.
            RuntimeError: If there are runtime issues during tensor operations.
            TypeError: If the input types are not as expected.
        """
        mixed_query_layer = self.query(hidden_states)

        # If this is instantiated as a cross-attention module, the keys
        # and values come from an encoder; the attention mask needs to be
        # such that the encoder's padding tokens are not attended to.
        is_cross_attention = encoder_hidden_states is not None

        if is_cross_attention and past_key_value is not None:
            # reuse k,v, cross_attentions
            key_layer = past_key_value[0]
            value_layer = past_key_value[1]
            attention_mask = encoder_attention_mask
        elif is_cross_attention:
            key_layer = self.transpose_for_scores(self.key(encoder_hidden_states))
            value_layer = self.transpose_for_scores(self.value(encoder_hidden_states))
            attention_mask = encoder_attention_mask
        elif past_key_value is not None:
            key_layer = self.transpose_for_scores(self.key(hidden_states))
            value_layer = self.transpose_for_scores(self.value(hidden_states))
            key_layer = ops.cat([past_key_value[0], key_layer], axis=2)
            value_layer = ops.cat([past_key_value[1], value_layer], axis=2)
        else:
            key_layer = self.transpose_for_scores(self.key(hidden_states))
            value_layer = self.transpose_for_scores(self.value(hidden_states))

        query_layer = self.transpose_for_scores(mixed_query_layer)

        use_cache = past_key_value is not None
        if self.is_decoder:
            # if cross_attention save Tuple(torch.Tensor, torch.Tensor) of all cross attention key/value_states.
            # Further calls to cross_attention layer can then reuse all cross-attention
            # key/value_states (first "if" case)
            # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of
            # all previous decoder key/value_states. Further calls to uni-directional self-attention
            # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
            # if encoder bi-directional self-attention `past_key_value` is always `None`
            past_key_value = (key_layer, value_layer)

        # Take the dot product between "query" and "key" to get the raw attention scores.
        attention_scores = ops.matmul(query_layer, key_layer.swapaxes(-1, -2))

        if self.position_embedding_type in ('relative_key', 'relative_key_query'):
            query_length, key_length = query_layer.shape[2], key_layer.shape[2]
            if use_cache:
                position_ids_l = mindspore.tensor(key_length - 1, dtype=mindspore.int64).view(-1, 1)
            else:
                position_ids_l = ops.arange(query_length, dtype=mindspore.int64).view(-1, 1)
            position_ids_r = ops.arange(key_length, dtype=mindspore.int64).view(1, -1)
            distance = position_ids_l - position_ids_r

            positional_embedding = self.distance_embedding(distance + self.max_position_embeddings - 1)
            positional_embedding = positional_embedding.to(dtype=query_layer.dtype)  # fp16 compatibility

            if self.position_embedding_type == "relative_key":
                relative_position_scores = ops.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
                attention_scores = attention_scores + relative_position_scores
            elif self.position_embedding_type == "relative_key_query":
                relative_position_scores_query = ops.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
                relative_position_scores_key = ops.einsum("bhrd,lrd->bhlr", key_layer, positional_embedding)
                attention_scores = attention_scores + relative_position_scores_query + relative_position_scores_key

        attention_scores = attention_scores / math.sqrt(self.attention_head_size)
        if attention_mask is not None:
            # Apply the attention mask is (precomputed for all layers in LayoutLMModel forward() function)
            attention_scores = attention_scores + attention_mask

        # Normalize the attention scores to probabilities.
        attention_probs = ops.softmax(attention_scores, axis=-1)

        # This is actually dropping out entire tokens to attend to, which might
        # seem a bit unusual, but is taken from the original Transformer paper.
        attention_probs = self.dropout(attention_probs)

        # Mask heads if we want to
        if head_mask is not None:
            attention_probs = attention_probs * head_mask

        context_layer = ops.matmul(attention_probs, value_layer)

        context_layer = context_layer.permute(0, 2, 1, 3)
        new_context_layer_shape = context_layer.shape[:-2] + (self.all_head_size,)
        context_layer = context_layer.view(new_context_layer_shape)

        outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)

        if self.is_decoder:
            outputs = outputs + (past_key_value,)
        return outputs

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfAttention.__init__(config, position_embedding_type=None)

Initializes a LayoutLMSelfAttention instance with the provided configuration and optional position embedding type.

PARAMETER DESCRIPTION
self

The instance of LayoutLMSelfAttention.
config

An object containing the configuration parameters for the self-attention layer. Expected attributes:

  • hidden_size (int): The size of the hidden layer.
  • num_attention_heads (int): The number of attention heads.
  • embedding_size (int, optional): The size of the embedding layer.
  • attention_probs_dropout_prob (float): The dropout probability for attention probabilities.
  • position_embedding_type (str, optional): The type of position embedding to use ('absolute' by default).
  • max_position_embeddings (int): The maximum number of position embeddings.
  • is_decoder (bool): Indicates if the self-attention layer is part of a decoder.
position_embedding_type

The type of position embedding to use (default is None). Accepted values: 'absolute', 'relative_key', 'relative_key_query'.

TYPE: str DEFAULT: None

RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
ValueError

If the hidden size is not a multiple of the number of attention heads and no embedding size is provided.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
def __init__(self, config, position_embedding_type=None):
    """
    Initializes a LayoutLMSelfAttention instance with the provided configuration and optional position embedding type.

    Args:
        self: The instance of LayoutLMSelfAttention.
        config: An object containing the configuration parameters for the self-attention layer.
            Expected attributes:

            - hidden_size (int): The size of the hidden layer.
            - num_attention_heads (int): The number of attention heads.
            - embedding_size (int, optional): The size of the embedding layer.
            - attention_probs_dropout_prob (float): The dropout probability for attention probabilities.
            - position_embedding_type (str, optional): The type of position embedding to use ('absolute' by default).
            - max_position_embeddings (int): The maximum number of position embeddings.
            - is_decoder (bool): Indicates if the self-attention layer is part of a decoder.
        position_embedding_type (str, optional): The type of position embedding to use (default is None).
            Accepted values: 'absolute', 'relative_key', 'relative_key_query'.

    Returns:
        None.

    Raises:
        ValueError:
            If the hidden size is not a multiple of the number of attention heads and no embedding size is provided.
    """
    super().__init__()
    if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
        raise ValueError(
            f"The hidden size ({config.hidden_size}) is not a multiple of the number of attention "
            f"heads ({config.num_attention_heads})"
        )

    self.num_attention_heads = config.num_attention_heads
    self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
    self.all_head_size = self.num_attention_heads * self.attention_head_size

    self.query = nn.Linear(config.hidden_size, self.all_head_size)
    self.key = nn.Linear(config.hidden_size, self.all_head_size)
    self.value = nn.Linear(config.hidden_size, self.all_head_size)

    self.dropout = nn.Dropout(p=config.attention_probs_dropout_prob)
    self.position_embedding_type = position_embedding_type or getattr(
        config, "position_embedding_type", "absolute"
    )
    if self.position_embedding_type in ('relative_key', 'relative_key_query'):
        self.max_position_embeddings = config.max_position_embeddings
        self.distance_embedding = nn.Embedding(2 * config.max_position_embeddings - 1, self.attention_head_size)

    self.is_decoder = config.is_decoder

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfAttention.forward(hidden_states, attention_mask=None, head_mask=None, encoder_hidden_states=None, encoder_attention_mask=None, past_key_value=None, output_attentions=False)

This method forwards the self-attention mechanism for the LayoutLMSelfAttention class.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMSelfAttention class.
hidden_states

The input hidden states tensor.

TYPE: Tensor

attention_mask

Mask tensor to prevent attention to certain positions. Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

head_mask

Mask tensor to control the heads involved in the attention computation. Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

encoder_hidden_states

Hidden states of the encoder in case of cross-attention. Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

encoder_attention_mask

Mask tensor for encoder attention. Defaults to None.

TYPE: Optional[Tensor] DEFAULT: None

past_key_value

Cached key and value tensors from previous attention calculations. Defaults to None.

TYPE: Optional[Tuple[Tuple[Tensor]]] DEFAULT: None

output_attentions

Flag to indicate whether to output attentions. Defaults to False.

TYPE: Optional[bool] DEFAULT: False

RETURNS DESCRIPTION
Tuple[Tensor]

Tuple[mindspore.Tensor]: A tuple containing the context layer tensor and optionally attention probabilities tensor.
RAISES DESCRIPTION
ValueError

If the input tensor shapes are incompatible for matrix multiplication.
RuntimeError

If there are runtime issues during tensor operations.
TypeError

If the input types are not as expected.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
def forward(
    self,
    hidden_states: mindspore.Tensor,
    attention_mask: Optional[mindspore.Tensor] = None,
    head_mask: Optional[mindspore.Tensor] = None,
    encoder_hidden_states: Optional[mindspore.Tensor] = None,
    encoder_attention_mask: Optional[mindspore.Tensor] = None,
    past_key_value: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
    output_attentions: Optional[bool] = False,
) -> Tuple[mindspore.Tensor]:
    """
    This method forwards the self-attention mechanism for the LayoutLMSelfAttention class.

    Args:
        self: The instance of the LayoutLMSelfAttention class.
        hidden_states (mindspore.Tensor): The input hidden states tensor.
        attention_mask (Optional[mindspore.Tensor], optional): Mask tensor to prevent attention to certain
            positions. Defaults to None.
        head_mask (Optional[mindspore.Tensor], optional): Mask tensor to control the heads involved in the
            attention computation. Defaults to None.
        encoder_hidden_states (Optional[mindspore.Tensor], optional): Hidden states of the encoder in case of
            cross-attention. Defaults to None.
        encoder_attention_mask (Optional[mindspore.Tensor], optional): Mask tensor for encoder attention.
            Defaults to None.
        past_key_value (Optional[Tuple[Tuple[mindspore.Tensor]]], optional): Cached key and value tensors from
            previous attention calculations. Defaults to None.
        output_attentions (Optional[bool], optional): Flag to indicate whether to output attentions.
            Defaults to False.

    Returns:
        Tuple[mindspore.Tensor]: A tuple containing the context layer tensor and optionally attention
            probabilities tensor.

    Raises:
        ValueError: If the input tensor shapes are incompatible for matrix multiplication.
        RuntimeError: If there are runtime issues during tensor operations.
        TypeError: If the input types are not as expected.
    """
    mixed_query_layer = self.query(hidden_states)

    # If this is instantiated as a cross-attention module, the keys
    # and values come from an encoder; the attention mask needs to be
    # such that the encoder's padding tokens are not attended to.
    is_cross_attention = encoder_hidden_states is not None

    if is_cross_attention and past_key_value is not None:
        # reuse k,v, cross_attentions
        key_layer = past_key_value[0]
        value_layer = past_key_value[1]
        attention_mask = encoder_attention_mask
    elif is_cross_attention:
        key_layer = self.transpose_for_scores(self.key(encoder_hidden_states))
        value_layer = self.transpose_for_scores(self.value(encoder_hidden_states))
        attention_mask = encoder_attention_mask
    elif past_key_value is not None:
        key_layer = self.transpose_for_scores(self.key(hidden_states))
        value_layer = self.transpose_for_scores(self.value(hidden_states))
        key_layer = ops.cat([past_key_value[0], key_layer], axis=2)
        value_layer = ops.cat([past_key_value[1], value_layer], axis=2)
    else:
        key_layer = self.transpose_for_scores(self.key(hidden_states))
        value_layer = self.transpose_for_scores(self.value(hidden_states))

    query_layer = self.transpose_for_scores(mixed_query_layer)

    use_cache = past_key_value is not None
    if self.is_decoder:
        # if cross_attention save Tuple(torch.Tensor, torch.Tensor) of all cross attention key/value_states.
        # Further calls to cross_attention layer can then reuse all cross-attention
        # key/value_states (first "if" case)
        # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of
        # all previous decoder key/value_states. Further calls to uni-directional self-attention
        # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
        # if encoder bi-directional self-attention `past_key_value` is always `None`
        past_key_value = (key_layer, value_layer)

    # Take the dot product between "query" and "key" to get the raw attention scores.
    attention_scores = ops.matmul(query_layer, key_layer.swapaxes(-1, -2))

    if self.position_embedding_type in ('relative_key', 'relative_key_query'):
        query_length, key_length = query_layer.shape[2], key_layer.shape[2]
        if use_cache:
            position_ids_l = mindspore.tensor(key_length - 1, dtype=mindspore.int64).view(-1, 1)
        else:
            position_ids_l = ops.arange(query_length, dtype=mindspore.int64).view(-1, 1)
        position_ids_r = ops.arange(key_length, dtype=mindspore.int64).view(1, -1)
        distance = position_ids_l - position_ids_r

        positional_embedding = self.distance_embedding(distance + self.max_position_embeddings - 1)
        positional_embedding = positional_embedding.to(dtype=query_layer.dtype)  # fp16 compatibility

        if self.position_embedding_type == "relative_key":
            relative_position_scores = ops.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
            attention_scores = attention_scores + relative_position_scores
        elif self.position_embedding_type == "relative_key_query":
            relative_position_scores_query = ops.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
            relative_position_scores_key = ops.einsum("bhrd,lrd->bhlr", key_layer, positional_embedding)
            attention_scores = attention_scores + relative_position_scores_query + relative_position_scores_key

    attention_scores = attention_scores / math.sqrt(self.attention_head_size)
    if attention_mask is not None:
        # Apply the attention mask is (precomputed for all layers in LayoutLMModel forward() function)
        attention_scores = attention_scores + attention_mask

    # Normalize the attention scores to probabilities.
    attention_probs = ops.softmax(attention_scores, axis=-1)

    # This is actually dropping out entire tokens to attend to, which might
    # seem a bit unusual, but is taken from the original Transformer paper.
    attention_probs = self.dropout(attention_probs)

    # Mask heads if we want to
    if head_mask is not None:
        attention_probs = attention_probs * head_mask

    context_layer = ops.matmul(attention_probs, value_layer)

    context_layer = context_layer.permute(0, 2, 1, 3)
    new_context_layer_shape = context_layer.shape[:-2] + (self.all_head_size,)
    context_layer = context_layer.view(new_context_layer_shape)

    outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)

    if self.is_decoder:
        outputs = outputs + (past_key_value,)
    return outputs

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfAttention.transpose_for_scores(x)

Transposes the input tensor x for calculating self-attention scores.

PARAMETER DESCRIPTION
self

The current instance of the LayoutLMSelfAttention class.

TYPE: LayoutLMSelfAttention

x

The input tensor of shape (batch_size, sequence_length, hidden_size).

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: The transposed tensor of shape (batch_size, num_attention_heads, sequence_length, attention_head_size).

This method transposes the input tensor x to prepare it for calculating self-attention scores in the LayoutLMSelfAttention model. The transposition is performed by reshaping the tensor to include the number of attention heads and the size of each attention head. The resulting tensor is then permuted to match the desired shape (batch_size, num_attention_heads, sequence_length, attention_head_size).

Note that this method assumes that the input tensor x has a rank of at least 3, where the last dimension represents the hidden size. The number of attention heads and the size of each attention head are obtained from the attributes num_attention_heads and attention_head_size of the LayoutLMSelfAttention instance, respectively.

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
def transpose_for_scores(self, x: mindspore.Tensor) -> mindspore.Tensor:
    """
    Transposes the input tensor `x` for calculating self-attention scores.

    Args:
        self (LayoutLMSelfAttention): The current instance of the LayoutLMSelfAttention class.
        x (mindspore.Tensor): The input tensor of shape `(batch_size, sequence_length, hidden_size)`.

    Returns:
        mindspore.Tensor:
            The transposed tensor of shape `(batch_size, num_attention_heads, sequence_length, attention_head_size)`.

    Raises:
        None

    This method transposes the input tensor `x` to prepare it for calculating self-attention scores in the
    LayoutLMSelfAttention model. The transposition is performed by reshaping the tensor to include the number of
    attention heads and the size of each attention head. The resulting tensor is then permuted to match the desired
    shape `(batch_size, num_attention_heads, sequence_length, attention_head_size)`.

    Note that this method assumes that the input tensor `x` has a rank of at least 3, where the last dimension
    represents the hidden size. The number of attention heads and the size of each attention head are obtained
    from the attributes `num_attention_heads` and `attention_head_size` of the LayoutLMSelfAttention instance,
    respectively.
    """
    new_x_shape = x.shape[:-1] + (self.num_attention_heads, self.attention_head_size)
    x = x.view(new_x_shape)
    return x.permute(0, 2, 1, 3)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfOutput

Bases: Module

Copied from transformers.models.bert.modeling_bert.BertSelfOutput with Bert->LayoutLM

Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
class LayoutLMSelfOutput(nn.Module):
    """Copied from transformers.models.bert.modeling_bert.BertSelfOutput with Bert->LayoutLM"""
    def __init__(self, config):
        """
        Initializes the LayoutLMSelfOutput class.

        Args:
            self (object): The instance of the class itself.
            config (object):
                An object containing configuration parameters for the layout model.

                - Type: Custom class
                - Purpose: To provide configuration settings for the layout model.
                - Restrictions: Must be compatible with the defined configuration structure.

        Returns:
            None.

        Raises:
            TypeError: If the provided 'config' parameter is not of the expected type.
            ValueError: If the configuration provided is missing essential parameters.
        """
        super().__init__()
        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
        self.dropout = nn.Dropout(p=config.hidden_dropout_prob)

    def forward(self, hidden_states: mindspore.Tensor, input_tensor: mindspore.Tensor) -> mindspore.Tensor:
        """
        Constructs the output of the LayoutLMSelfOutput layer.

        Args:
            self (LayoutLMSelfOutput): The instance of the LayoutLMSelfOutput class.
            hidden_states (mindspore.Tensor): The hidden states tensor generated by the layer.
                This tensor contains the output of the dense and dropout layers.
            input_tensor (mindspore.Tensor): The input tensor to the layer.
                This tensor represents the input to the layer that needs to be added to the hidden states.

        Returns:
            mindspore.Tensor: The tensor representing the output of the LayerNorm operation.
                This tensor is the result of adding the input tensor to the normalized hidden states.

        Raises:
            None
        """
        hidden_states = self.dense(hidden_states)
        hidden_states = self.dropout(hidden_states)
        hidden_states = self.LayerNorm(hidden_states + input_tensor)
        return hidden_states

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfOutput.__init__(config)

Initializes the LayoutLMSelfOutput class.

PARAMETER DESCRIPTION
self

The instance of the class itself.

TYPE: object

config

An object containing configuration parameters for the layout model.

  • Type: Custom class
  • Purpose: To provide configuration settings for the layout model.
  • Restrictions: Must be compatible with the defined configuration structure.

TYPE: object

RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
TypeError

If the provided 'config' parameter is not of the expected type.
ValueError

If the configuration provided is missing essential parameters.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
def __init__(self, config):
    """
    Initializes the LayoutLMSelfOutput class.

    Args:
        self (object): The instance of the class itself.
        config (object):
            An object containing configuration parameters for the layout model.

            - Type: Custom class
            - Purpose: To provide configuration settings for the layout model.
            - Restrictions: Must be compatible with the defined configuration structure.

    Returns:
        None.

    Raises:
        TypeError: If the provided 'config' parameter is not of the expected type.
        ValueError: If the configuration provided is missing essential parameters.
    """
    super().__init__()
    self.dense = nn.Linear(config.hidden_size, config.hidden_size)
    self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
    self.dropout = nn.Dropout(p=config.hidden_dropout_prob)

mindnlp.transformers.models.layoutlm.modeling_layoutlm.LayoutLMSelfOutput.forward(hidden_states, input_tensor)

Constructs the output of the LayoutLMSelfOutput layer.

PARAMETER DESCRIPTION
self

The instance of the LayoutLMSelfOutput class.

TYPE: LayoutLMSelfOutput

hidden_states

The hidden states tensor generated by the layer. This tensor contains the output of the dense and dropout layers.

TYPE: Tensor

input_tensor

The input tensor to the layer. This tensor represents the input to the layer that needs to be added to the hidden states.

TYPE: Tensor

RETURNS DESCRIPTION
Tensor

mindspore.Tensor: The tensor representing the output of the LayerNorm operation. This tensor is the result of adding the input tensor to the normalized hidden states.
Source code in mindnlp/transformers/models/layoutlm/modeling_layoutlm.py 
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
def forward(self, hidden_states: mindspore.Tensor, input_tensor: mindspore.Tensor) -> mindspore.Tensor:
    """
    Constructs the output of the LayoutLMSelfOutput layer.

    Args:
        self (LayoutLMSelfOutput): The instance of the LayoutLMSelfOutput class.
        hidden_states (mindspore.Tensor): The hidden states tensor generated by the layer.
            This tensor contains the output of the dense and dropout layers.
        input_tensor (mindspore.Tensor): The input tensor to the layer.
            This tensor represents the input to the layer that needs to be added to the hidden states.

    Returns:
        mindspore.Tensor: The tensor representing the output of the LayerNorm operation.
            This tensor is the result of adding the input tensor to the normalized hidden states.

    Raises:
        None
    """
    hidden_states = self.dense(hidden_states)
    hidden_states = self.dropout(hidden_states)
    hidden_states = self.LayerNorm(hidden_states + input_tensor)
    return hidden_states

mindnlp.transformers.models.layoutlm.configuration_layoutlm

LayoutLM Models config

mindnlp.transformers.models.layoutlm.configuration_layoutlm.LayoutLMConfig

Bases: PretrainedConfig

LayoutLMConfig

Source code in mindnlp/transformers/models/layoutlm/configuration_layoutlm.py 
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
class LayoutLMConfig(PretrainedConfig):
    """LayoutLMConfig"""
    model_type = "layoutlm"
    keys_to_ignore_at_inference = ["past_key_values"]

    def __init__(
        self,
        vocab_size=30522,
        hidden_size=768,
        num_hidden_layers=12,
        num_attention_heads=12,
        intermediate_size=3072,
        hidden_act="gelu",
        hidden_dropout_prob=0.1,
        attention_probs_dropout_prob=0.1,
        max_position_embeddings=512,
        type_vocab_size=2,
        initializer_range=0.02,
        layer_norm_eps=1e-12,
        pad_token_id=0,
        position_embedding_type="absolute",
        use_cache=True,
        max_2d_position_embeddings=1024,
        **kwargs,
    ):
        """
        Initializes a LayoutLMConfig object.

        Args:
            self: The instance of the class.
            vocab_size (int, optional): The size of the vocabulary. Defaults to 30522.
            hidden_size (int, optional): The size of the hidden layers. Defaults to 768.
            num_hidden_layers (int, optional): The number of hidden layers. Defaults to 12.
            num_attention_heads (int, optional): The number of attention heads. Defaults to 12.
            intermediate_size (int, optional): The size of the intermediate layer in the transformer encoder. 
                Defaults to 3072.
            hidden_act (str, optional): The activation function for the hidden layers. Defaults to 'gelu'.
            hidden_dropout_prob (float, optional): The dropout probability for the hidden layers. Defaults to 0.1.
            attention_probs_dropout_prob (float, optional): The dropout probability for the attention probabilities. 
                Defaults to 0.1.
            max_position_embeddings (int, optional): The maximum sequence length that this model might ever be used with. 
                Defaults to 512.
            type_vocab_size (int, optional): The size of the token type vocabulary. Defaults to 2.
            initializer_range (float, optional): The standard deviation of the truncated_normal_initializer for 
                initializing all weight matrices. Defaults to 0.02.
            layer_norm_eps (float, optional): The epsilon value to use in LayerNorm layers. Defaults to 1e-12.
            pad_token_id (int, optional): The id of the padding token. Defaults to 0.
            position_embedding_type (str, optional): The type of position embedding. Defaults to 'absolute'.
            use_cache (bool, optional): Whether to use cache for the model. Defaults to True.
            max_2d_position_embeddings (int, optional): The maximum 2D sequence length that this model 
                might ever be used with. Defaults to 1024.

        Returns:
            None.

        Raises:
            None.
        """
        super().__init__(pad_token_id=pad_token_id, **kwargs)
        self.vocab_size = vocab_size
        self.hidden_size = hidden_size
        self.num_hidden_layers = num_hidden_layers
        self.num_attention_heads = num_attention_heads
        self.hidden_act = hidden_act
        self.intermediate_size = intermediate_size
        self.hidden_dropout_prob = hidden_dropout_prob
        self.attention_probs_dropout_prob = attention_probs_dropout_prob
        self.max_position_embeddings = max_position_embeddings
        self.type_vocab_size = type_vocab_size
        self.initializer_range = initializer_range
        self.layer_norm_eps = layer_norm_eps
        self.position_embedding_type = position_embedding_type
        self.use_cache = use_cache
        self.max_2d_position_embeddings = max_2d_position_embeddings

mindnlp.transformers.models.layoutlm.configuration_layoutlm.LayoutLMConfig.__init__(vocab_size=30522, hidden_size=768, num_hidden_layers=12, num_attention_heads=12, intermediate_size=3072, hidden_act='gelu', hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, max_position_embeddings=512, type_vocab_size=2, initializer_range=0.02, layer_norm_eps=1e-12, pad_token_id=0, position_embedding_type='absolute', use_cache=True, max_2d_position_embeddings=1024, **kwargs)

Initializes a LayoutLMConfig object.

PARAMETER DESCRIPTION
self

The instance of the class.
vocab_size

The size of the vocabulary. Defaults to 30522.

TYPE: int DEFAULT: 30522

hidden_size

The size of the hidden layers. Defaults to 768.

TYPE: int DEFAULT: 768

num_hidden_layers

The number of hidden layers. Defaults to 12.

TYPE: int DEFAULT: 12

num_attention_heads

The number of attention heads. Defaults to 12.

TYPE: int DEFAULT: 12

intermediate_size

The size of the intermediate layer in the transformer encoder. Defaults to 3072.

TYPE: int DEFAULT: 3072

hidden_act

The activation function for the hidden layers. Defaults to 'gelu'.

TYPE: str DEFAULT: 'gelu'

hidden_dropout_prob

The dropout probability for the hidden layers. Defaults to 0.1.

TYPE: float DEFAULT: 0.1

attention_probs_dropout_prob

The dropout probability for the attention probabilities. Defaults to 0.1.

TYPE: float DEFAULT: 0.1

max_position_embeddings

The maximum sequence length that this model might ever be used with. Defaults to 512.

TYPE: int DEFAULT: 512

type_vocab_size

The size of the token type vocabulary. Defaults to 2.

TYPE: int DEFAULT: 2

initializer_range

The standard deviation of the truncated_normal_initializer for initializing all weight matrices. Defaults to 0.02.

TYPE: float DEFAULT: 0.02

layer_norm_eps

The epsilon value to use in LayerNorm layers. Defaults to 1e-12.

TYPE: float DEFAULT: 1e-12

pad_token_id

The id of the padding token. Defaults to 0.

TYPE: int DEFAULT: 0

position_embedding_type

The type of position embedding. Defaults to 'absolute'.

TYPE: str DEFAULT: 'absolute'

use_cache

Whether to use cache for the model. Defaults to True.

TYPE: bool DEFAULT: True

max_2d_position_embeddings

The maximum 2D sequence length that this model might ever be used with. Defaults to 1024.

TYPE: int DEFAULT: 1024

RETURNS DESCRIPTION

None.
Source code in mindnlp/transformers/models/layoutlm/configuration_layoutlm.py 
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
def __init__(
    self,
    vocab_size=30522,
    hidden_size=768,
    num_hidden_layers=12,
    num_attention_heads=12,
    intermediate_size=3072,
    hidden_act="gelu",
    hidden_dropout_prob=0.1,
    attention_probs_dropout_prob=0.1,
    max_position_embeddings=512,
    type_vocab_size=2,
    initializer_range=0.02,
    layer_norm_eps=1e-12,
    pad_token_id=0,
    position_embedding_type="absolute",
    use_cache=True,
    max_2d_position_embeddings=1024,
    **kwargs,
):
    """
    Initializes a LayoutLMConfig object.

    Args:
        self: The instance of the class.
        vocab_size (int, optional): The size of the vocabulary. Defaults to 30522.
        hidden_size (int, optional): The size of the hidden layers. Defaults to 768.
        num_hidden_layers (int, optional): The number of hidden layers. Defaults to 12.
        num_attention_heads (int, optional): The number of attention heads. Defaults to 12.
        intermediate_size (int, optional): The size of the intermediate layer in the transformer encoder. 
            Defaults to 3072.
        hidden_act (str, optional): The activation function for the hidden layers. Defaults to 'gelu'.
        hidden_dropout_prob (float, optional): The dropout probability for the hidden layers. Defaults to 0.1.
        attention_probs_dropout_prob (float, optional): The dropout probability for the attention probabilities. 
            Defaults to 0.1.
        max_position_embeddings (int, optional): The maximum sequence length that this model might ever be used with. 
            Defaults to 512.
        type_vocab_size (int, optional): The size of the token type vocabulary. Defaults to 2.
        initializer_range (float, optional): The standard deviation of the truncated_normal_initializer for 
            initializing all weight matrices. Defaults to 0.02.
        layer_norm_eps (float, optional): The epsilon value to use in LayerNorm layers. Defaults to 1e-12.
        pad_token_id (int, optional): The id of the padding token. Defaults to 0.
        position_embedding_type (str, optional): The type of position embedding. Defaults to 'absolute'.
        use_cache (bool, optional): Whether to use cache for the model. Defaults to True.
        max_2d_position_embeddings (int, optional): The maximum 2D sequence length that this model 
            might ever be used with. Defaults to 1024.

    Returns:
        None.

    Raises:
        None.
    """
    super().__init__(pad_token_id=pad_token_id, **kwargs)
    self.vocab_size = vocab_size
    self.hidden_size = hidden_size
    self.num_hidden_layers = num_hidden_layers
    self.num_attention_heads = num_attention_heads
    self.hidden_act = hidden_act
    self.intermediate_size = intermediate_size
    self.hidden_dropout_prob = hidden_dropout_prob
    self.attention_probs_dropout_prob = attention_probs_dropout_prob
    self.max_position_embeddings = max_position_embeddings
    self.type_vocab_size = type_vocab_size
    self.initializer_range = initializer_range
    self.layer_norm_eps = layer_norm_eps
    self.position_embedding_type = position_embedding_type
    self.use_cache = use_cache
    self.max_2d_position_embeddings = max_2d_position_embeddings