跳转至

megatron_bert

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert

MindSpore MegatronBERT model.

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertEmbeddings

Bases: Module

Construct the embeddings from word, position and token_type embeddings.

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.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
 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
class MegatronBertEmbeddings(nn.Module):
    """Construct the embeddings from word, position and token_type embeddings."""

    def __init__(self, config):
        super().__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.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)

        # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
        # any TensorFlow checkpoint file

        # In Megatron, layer-norm is applied after the 1st dropout.
        # self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
        self.dropout = nn.Dropout(config.hidden_dropout_prob)

        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
        self.register_buffer(
            "position_ids", ops.arange(config.max_position_embeddings).broadcast_to((1, -1)), persistent=False
        )
        self.position_embedding_type = getattr(config, "position_embedding_type", "absolute")

    def forward(
        self,
        input_ids: Optional[mindspore.Tensor] = None,
        token_type_ids: Optional[mindspore.Tensor] = None,
        position_ids: Optional[mindspore.Tensor] = None,
        inputs_embeds: Optional[mindspore.Tensor] = None,
        past_key_values_length: int = 0,
    ) -> mindspore.Tensor:
        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[:, past_key_values_length : seq_length + past_key_values_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)
        token_type_embeddings = self.token_type_embeddings(token_type_ids)

        embeddings = inputs_embeds + token_type_embeddings
        if self.position_embedding_type == "absolute":
            position_embeddings = self.position_embeddings(position_ids)
            embeddings += position_embeddings

        # Megatron BERT moves that layer norm after the drop-out (and to each layer).
        # embeddings = self.LayerNorm(embeddings)
        embeddings = self.dropout(embeddings)
        return embeddings

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForCausalLM

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 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
1012
1013
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
1060
1061
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
class MegatronBertForCausalLM(MegatronBertPreTrainedModel):
    _tied_weights_keys = ["cls.predictions.decoder"]

    def __init__(self, config):
        super().__init__(config)

        if not config.is_decoder:
            logger.warning("If you want to use `MegatronBertForCausalLM` as a standalone, add `is_decoder=True.`")

        self.bert = MegatronBertModel(config, add_pooling_layer=False)
        self.cls = MegatronBertOnlyMLMHead(config)

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

    def get_output_embeddings(self):
        return self.cls.predictions.decoder

    def set_output_embeddings(self, new_embeddings):
        self.cls.predictions.decoder = new_embeddings
        self.cls.predictions.bias = new_embeddings.bias

    def forward(
        self,
        input_ids: 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,
        labels: Optional[mindspore.Tensor] = None,
        past_key_values: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, CausalLMOutputWithCrossAttentions]:
        r"""
        encoder_hidden_states  (`mindspore.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
            Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
            the model is configured as a decoder.
        encoder_attention_mask (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
            Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
            the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:

            - 1 for tokens that are **not masked**,
            - 0 for tokens that are **masked**.
        labels (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the left-to-right language modeling loss (next word prediction). 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 n `[0, ..., config.vocab_size]`
        past_key_values (`tuple(tuple(mindspore.Tensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
            Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
        use_cache (`bool`, *optional*):
            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
            `past_key_values`).

        Returns:

        Example:

        ```python
        >>> from transformers import AutoTokenizer, MegatronBertForCausalLM, MegatronBertConfig
        >>> import torch

        >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
        >>> model = MegatronBertForCausalLM.from_pretrained("nvidia/megatron-bert-cased-345m", is_decoder=True)

        >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
        >>> outputs = model(**inputs)

        >>> prediction_logits = outputs.logits
        ```"""

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
        if labels is not None:
            use_cache = False

        outputs = self.bert(
            input_ids,
            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,
            past_key_values=past_key_values,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

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

        lm_loss = None
        if labels is not None:
            # we are doing next-token prediction; shift prediction scores and input ids by one
            shifted_prediction_scores = prediction_scores[:, :-1, :]
            labels = labels[:, 1:]
            loss_fct = CrossEntropyLoss()
            lm_loss = loss_fct(shifted_prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))

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

        return CausalLMOutputWithCrossAttentions(
            loss=lm_loss,
            logits=prediction_scores,
            past_key_values=outputs.past_key_values,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
            cross_attentions=outputs.cross_attentions,
        )

    def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, **model_kwargs):
        input_shape = input_ids.shape
        # if model is used as a decoder in encoder-decoder model, the decoder attention mask is created on the fly
        if attention_mask is None:
            attention_mask = input_ids.new_ones(input_shape)

        # cut decoder_input_ids if past_key_values is used
        if past_key_values is not None:
            past_length = past_key_values[0][0].shape[2]

            # Some generation methods already pass only the last input ID
            if input_ids.shape[1] > past_length:
                remove_prefix_length = past_length
            else:
                # Default to old behavior: keep only final ID
                remove_prefix_length = input_ids.shape[1] - 1

            input_ids = input_ids[:, remove_prefix_length:]

        return {"input_ids": input_ids, "attention_mask": attention_mask, "past_key_values": past_key_values}

    def _reorder_cache(self, past_key_values, beam_idx):
        reordered_past = ()
        for layer_past in past_key_values:
            reordered_past += (
                tuple(past_state.index_select(0, beam_idx) for past_state in layer_past),
            )
        return reordered_past

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForCausalLM.forward(input_ids=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, labels=None, past_key_values=None, use_cache=None, output_attentions=None, output_hidden_states=None, return_dict=None)

encoder_hidden_states (mindspore.Tensor of shape (batch_size, sequence_length, hidden_size), optional): Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder. encoder_attention_mask (mindspore.Tensor of shape (batch_size, sequence_length), optional): Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in [0, 1]:

- 1 for tokens that are **not masked**,
- 0 for tokens that are **masked**.

labels (mindspore.Tensor of shape (batch_size, sequence_length), optional): Labels for computing the left-to-right language modeling loss (next word prediction). 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 n [0, ..., config.vocab_size] past_key_values (tuple(tuple(mindspore.Tensor)) of length config.n_layers with each tuple having 4 tensors of shape (batch_size, num_heads, sequence_length - 1, embed_size_per_head)): Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
`decoder_input_ids` of shape `(batch_size, sequence_length)`.

use_cache (bool, optional): If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).

Returns:

Example:

>>> from transformers import AutoTokenizer, MegatronBertForCausalLM, MegatronBertConfig
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
>>> model = MegatronBertForCausalLM.from_pretrained("nvidia/megatron-bert-cased-345m", is_decoder=True)

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs)

>>> prediction_logits = outputs.logits
Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
 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
1012
1013
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
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
def forward(
    self,
    input_ids: 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,
    labels: Optional[mindspore.Tensor] = None,
    past_key_values: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
    use_cache: Optional[bool] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, CausalLMOutputWithCrossAttentions]:
    r"""
    encoder_hidden_states  (`mindspore.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
        Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
        the model is configured as a decoder.
    encoder_attention_mask (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
        Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
        the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:

        - 1 for tokens that are **not masked**,
        - 0 for tokens that are **masked**.
    labels (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
        Labels for computing the left-to-right language modeling loss (next word prediction). 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 n `[0, ..., config.vocab_size]`
    past_key_values (`tuple(tuple(mindspore.Tensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
        Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

        If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
        don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
        `decoder_input_ids` of shape `(batch_size, sequence_length)`.
    use_cache (`bool`, *optional*):
        If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
        `past_key_values`).

    Returns:

    Example:

    ```python
    >>> from transformers import AutoTokenizer, MegatronBertForCausalLM, MegatronBertConfig
    >>> import torch

    >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
    >>> model = MegatronBertForCausalLM.from_pretrained("nvidia/megatron-bert-cased-345m", is_decoder=True)

    >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
    >>> outputs = model(**inputs)

    >>> prediction_logits = outputs.logits
    ```"""

    return_dict = return_dict if return_dict is not None else self.config.use_return_dict
    if labels is not None:
        use_cache = False

    outputs = self.bert(
        input_ids,
        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,
        past_key_values=past_key_values,
        use_cache=use_cache,
        output_attentions=output_attentions,
        output_hidden_states=output_hidden_states,
        return_dict=return_dict,
    )

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

    lm_loss = None
    if labels is not None:
        # we are doing next-token prediction; shift prediction scores and input ids by one
        shifted_prediction_scores = prediction_scores[:, :-1, :]
        labels = labels[:, 1:]
        loss_fct = CrossEntropyLoss()
        lm_loss = loss_fct(shifted_prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))

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

    return CausalLMOutputWithCrossAttentions(
        loss=lm_loss,
        logits=prediction_scores,
        past_key_values=outputs.past_key_values,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
        cross_attentions=outputs.cross_attentions,
    )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForMaskedLM

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
1102
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
1131
1132
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
class MegatronBertForMaskedLM(MegatronBertPreTrainedModel):
    _tied_weights_keys = ["cls.predictions.decoder"]

    def __init__(self, config):
        super().__init__(config)

        if config.is_decoder:
            logger.warning(
                "If you want to use `MegatronBertForMaskedLM` make sure `config.is_decoder=False` for "
                "bi-directional self-attention."
            )

        self.bert = MegatronBertModel(config, add_pooling_layer=False)
        self.cls = MegatronBertOnlyMLMHead(config)

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

    def get_output_embeddings(self):
        return self.cls.predictions.decoder

    def set_output_embeddings(self, new_embeddings):
        self.cls.predictions.decoder = new_embeddings
        self.cls.predictions.bias = new_embeddings.bias

    def forward(
        self,
        input_ids: 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,
        labels: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, MaskedLMOutput]:
        r"""
        labels (`mindspore.Tensor` 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]`
        """

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.bert(
            input_ids,
            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:
            loss_fct = CrossEntropyLoss()  # -100 index = padding token
            masked_lm_loss = loss_fct(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,
        )

    def prepare_inputs_for_generation(self, input_ids, attention_mask=None, **model_kwargs):
        input_shape = input_ids.shape
        effective_batch_size = input_shape[0]

        #  add a dummy token
        if self.config.pad_token_id is None:
            raise ValueError("The PAD token should be defined for generation")
        attention_mask = ops.cat([attention_mask, attention_mask.new_zeros((attention_mask.shape[0], 1))], dim=-1)
        dummy_token = ops.full(
            (effective_batch_size, 1), self.config.pad_token_id, dtype=mindspore.int64
        )
        input_ids = ops.cat([input_ids, dummy_token], dim=1)

        return {"input_ids": input_ids, "attention_mask": attention_mask}

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForMaskedLM.forward(input_ids=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, labels=None, output_attentions=None, output_hidden_states=None, return_dict=None)

labels (mindspore.Tensor 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]

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
1127
1128
1129
1130
1131
1132
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
def forward(
    self,
    input_ids: 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,
    labels: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, MaskedLMOutput]:
    r"""
    labels (`mindspore.Tensor` 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]`
    """

    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.bert(
        input_ids,
        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:
        loss_fct = CrossEntropyLoss()  # -100 index = padding token
        masked_lm_loss = loss_fct(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.megatron_bert.modeling_megatron_bert.MegatronBertForMultipleChoice

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
class MegatronBertForMultipleChoice(MegatronBertPreTrainedModel):
    def __init__(self, config):
        super().__init__(config)

        self.bert = MegatronBertModel(config)
        self.dropout = nn.Dropout(config.hidden_dropout_prob)
        self.classifier = nn.Linear(config.hidden_size, 1)

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

    def forward(
        self,
        input_ids: 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, MultipleChoiceModelOutput]:
        r"""
        labels (`mindspore.Tensor` of shape `(batch_size,)`, *optional*):
            Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
            num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See
            `input_ids` above)
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
        num_choices = input_ids.shape[1] if input_ids is not None else inputs_embeds.shape[1]

        input_ids = input_ids.view(-1, input_ids.shape[-1]) if input_ids is not None else None
        attention_mask = attention_mask.view(-1, attention_mask.shape[-1]) if attention_mask is not None else None
        token_type_ids = token_type_ids.view(-1, token_type_ids.shape[-1]) if token_type_ids is not None else None
        position_ids = position_ids.view(-1, position_ids.shape[-1]) if position_ids is not None else None
        inputs_embeds = (
            inputs_embeds.view(-1, inputs_embeds.shape[-2], inputs_embeds.shape[-1])
            if inputs_embeds is not None
            else None
        )

        outputs = self.bert(
            input_ids,
            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)
        reshaped_logits = logits.view(-1, num_choices)

        loss = None
        if labels is not None:
            loss_fct = CrossEntropyLoss()
            loss = loss_fct(reshaped_logits, labels)

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

        return MultipleChoiceModelOutput(
            loss=loss,
            logits=reshaped_logits,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForMultipleChoice.forward(input_ids=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)

labels (mindspore.Tensor of shape (batch_size,), optional): Labels for computing the multiple choice classification loss. Indices should be in [0, ..., num_choices-1] where num_choices is the size of the second dimension of the input tensors. (See input_ids above)

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
def forward(
    self,
    input_ids: 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, MultipleChoiceModelOutput]:
    r"""
    labels (`mindspore.Tensor` of shape `(batch_size,)`, *optional*):
        Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
        num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See
        `input_ids` above)
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict
    num_choices = input_ids.shape[1] if input_ids is not None else inputs_embeds.shape[1]

    input_ids = input_ids.view(-1, input_ids.shape[-1]) if input_ids is not None else None
    attention_mask = attention_mask.view(-1, attention_mask.shape[-1]) if attention_mask is not None else None
    token_type_ids = token_type_ids.view(-1, token_type_ids.shape[-1]) if token_type_ids is not None else None
    position_ids = position_ids.view(-1, position_ids.shape[-1]) if position_ids is not None else None
    inputs_embeds = (
        inputs_embeds.view(-1, inputs_embeds.shape[-2], inputs_embeds.shape[-1])
        if inputs_embeds is not None
        else None
    )

    outputs = self.bert(
        input_ids,
        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)
    reshaped_logits = logits.view(-1, num_choices)

    loss = None
    if labels is not None:
        loss_fct = CrossEntropyLoss()
        loss = loss_fct(reshaped_logits, labels)

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

    return MultipleChoiceModelOutput(
        loss=loss,
        logits=reshaped_logits,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForNextSentencePrediction

Bases: MegatronBertPreTrainedModel

Represents a MegatronBert model for next sentence prediction.

This class inherits from the MegatronBertPreTrainedModel and provides next sentence prediction functionality using the Megatron BERT model.

The class forwardor initializes the MegatronBertForNextSentencePrediction model with the given configuration.

The forward method takes input tensors and computes the next sentence prediction loss. It returns the next sentence predictor output.

PARAMETER DESCRIPTION
input_ids

The input tensor containing the indices of input sequence tokens in the vocabulary. Defaults to None.

TYPE: Optional[Tensor]

attention_mask

The input tensor containing indices specifying which tokens should be attended to. Defaults to None.

TYPE: Optional[Tensor]

token_type_ids

The input tensor containing the segment token indices to differentiate the sequences. Defaults to None.

TYPE: Optional[Tensor]

position_ids

The input tensor containing the position indices of each input token. Defaults to None.

TYPE: Optional[Tensor]

head_mask

The input tensor containing the mask for the attention heads. Defaults to None.

TYPE: Optional[Tensor]

inputs_embeds

The input tensor containing the embedded inputs. Defaults to None.

TYPE: Optional[Tensor]

labels

The tensor containing the labels for computing the next sequence prediction loss. Defaults to None.

TYPE: Optional[Tensor]

output_attentions

Whether to return attentions. Defaults to None.

TYPE: Optional[bool]

output_hidden_states

Whether to return hidden states. Defaults to None.

TYPE: Optional[bool]

return_dict

Whether to return a dictionary. Defaults to None.

TYPE: Optional[bool]

RETURNS DESCRIPTION

Union[Tuple, NextSentencePredictorOutput]: A tuple containing the next sentence prediction loss and the next sentence predictor output.
RAISES DESCRIPTION
FutureWarning

If the next_sentence_label argument is used, as it is deprecated.
Example 
>>> from transformers import AutoTokenizer, MegatronBertForNextSentencePrediction
...
>>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
>>> model = MegatronBertForNextSentencePrediction.from_pretrained("nvidia/megatron-bert-cased-345m")
...
>>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
>>> next_sentence = "The sky is blue due to the shorter wavelength of blue light."
>>> encoding = tokenizer(prompt, next_sentence, return_tensors="ms")
...
>>> outputs = model(**encoding, labels=mindspore.Tensor([1]))
>>> logits = outputs.logits
>>> assert logits[0, 0] < logits[0, 1]  # next sentence was random
Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
class MegatronBertForNextSentencePrediction(MegatronBertPreTrainedModel):
    """
    Represents a MegatronBert model for next sentence prediction.

    This class inherits from the MegatronBertPreTrainedModel and provides next sentence prediction functionality
    using the Megatron BERT model.

    The class forwardor initializes the MegatronBertForNextSentencePrediction model with the given configuration.

    The `forward` method takes input tensors and computes the next sentence prediction loss.
    It returns the next sentence predictor output.

    Args:
        input_ids (Optional[mindspore.Tensor], optional): The input tensor containing the indices of input sequence
            tokens in the vocabulary. Defaults to None.
        attention_mask (Optional[mindspore.Tensor], optional): The input tensor containing indices specifying which
            tokens should be attended to. Defaults to None.
        token_type_ids (Optional[mindspore.Tensor], optional): The input tensor containing the segment token indices
            to differentiate the sequences. Defaults to None.
        position_ids (Optional[mindspore.Tensor], optional): The input tensor containing the position indices of
            each input token. Defaults to None.
        head_mask (Optional[mindspore.Tensor], optional): The input tensor containing the mask for the attention heads.
            Defaults to None.
        inputs_embeds (Optional[mindspore.Tensor], optional): The input tensor containing the embedded inputs.
            Defaults to None.
        labels (Optional[mindspore.Tensor], optional): The tensor containing the labels for computing the next sequence
            prediction loss. Defaults to None.
        output_attentions (Optional[bool], optional): Whether to return attentions. Defaults to None.
        output_hidden_states (Optional[bool], optional): Whether to return hidden states. Defaults to None.
        return_dict (Optional[bool], optional): Whether to return a dictionary. Defaults to None.

    Returns:
        Union[Tuple, NextSentencePredictorOutput]: A tuple containing the next sentence prediction loss and the
            next sentence predictor output.

    Raises:
        FutureWarning: If the `next_sentence_label` argument is used, as it is deprecated.

    Example:
        ```python
        >>> from transformers import AutoTokenizer, MegatronBertForNextSentencePrediction
        ...
        >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
        >>> model = MegatronBertForNextSentencePrediction.from_pretrained("nvidia/megatron-bert-cased-345m")
        ...
        >>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
        >>> next_sentence = "The sky is blue due to the shorter wavelength of blue light."
        >>> encoding = tokenizer(prompt, next_sentence, return_tensors="ms")
        ...
        >>> outputs = model(**encoding, labels=mindspore.Tensor([1]))
        >>> logits = outputs.logits
        >>> assert logits[0, 0] < logits[0, 1]  # next sentence was random
        ```
    """
    def __init__(self, config):
        super().__init__(config)

        self.bert = MegatronBertModel(config)
        self.cls = MegatronBertOnlyNSPHead(config)

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

    def forward(
        self,
        input_ids: 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,
        **kwargs,
    ) -> Union[Tuple, NextSentencePredictorOutput]:
        r"""
        labels (`mindspore.Tensor` of shape `(batch_size,)`, *optional*):
            Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair
            (see `input_ids` docstring). Indices should be in `[0, 1]`:

            - 0 indicates sequence B is a continuation of sequence A,
            - 1 indicates sequence B is a random sequence.

        Returns:

        Example:
        ```python
        >>> from transformers import AutoTokenizer, MegatronBertForNextSentencePrediction
        >>> import torch

        >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
        >>> model = MegatronBertForNextSentencePrediction.from_pretrained("nvidia/megatron-bert-cased-345m")

        >>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
        >>> next_sentence = "The sky is blue due to the shorter wavelength of blue light."
        >>> encoding = tokenizer(prompt, next_sentence, return_tensors="pt")

        >>> outputs = model(**encoding, labels=mindspore.Tensor([1]))
        >>> logits = outputs.logits
        >>> assert logits[0, 0] < logits[0, 1]  # next sentence was random
        ```"""

        if "next_sentence_label" in kwargs:
            warnings.warn(
                "The `next_sentence_label` argument is deprecated and will be removed in a future version, use"
                " `labels` instead.",
                FutureWarning,
            )
            labels = kwargs.pop("next_sentence_label")

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.bert(
            input_ids,
            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]

        seq_relationship_scores = self.cls(pooled_output)

        next_sentence_loss = None
        if labels is not None:
            loss_fct = CrossEntropyLoss()
            next_sentence_loss = loss_fct(seq_relationship_scores.view(-1, 2), labels.view(-1))

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

        return NextSentencePredictorOutput(
            loss=next_sentence_loss,
            logits=seq_relationship_scores,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForNextSentencePrediction.forward(input_ids=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, **kwargs)

labels (mindspore.Tensor of shape (batch_size,), optional): Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair (see input_ids docstring). Indices should be in [0, 1]:

- 0 indicates sequence B is a continuation of sequence A,
- 1 indicates sequence B is a random sequence.

Returns:

Example: 

>>> from transformers import AutoTokenizer, MegatronBertForNextSentencePrediction
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
>>> model = MegatronBertForNextSentencePrediction.from_pretrained("nvidia/megatron-bert-cased-345m")

>>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
>>> next_sentence = "The sky is blue due to the shorter wavelength of blue light."
>>> encoding = tokenizer(prompt, next_sentence, return_tensors="pt")

>>> outputs = model(**encoding, labels=mindspore.Tensor([1]))
>>> logits = outputs.logits
>>> assert logits[0, 0] < logits[0, 1]  # next sentence was random

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
def forward(
    self,
    input_ids: 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,
    **kwargs,
) -> Union[Tuple, NextSentencePredictorOutput]:
    r"""
    labels (`mindspore.Tensor` of shape `(batch_size,)`, *optional*):
        Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair
        (see `input_ids` docstring). Indices should be in `[0, 1]`:

        - 0 indicates sequence B is a continuation of sequence A,
        - 1 indicates sequence B is a random sequence.

    Returns:

    Example:
    ```python
    >>> from transformers import AutoTokenizer, MegatronBertForNextSentencePrediction
    >>> import torch

    >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
    >>> model = MegatronBertForNextSentencePrediction.from_pretrained("nvidia/megatron-bert-cased-345m")

    >>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
    >>> next_sentence = "The sky is blue due to the shorter wavelength of blue light."
    >>> encoding = tokenizer(prompt, next_sentence, return_tensors="pt")

    >>> outputs = model(**encoding, labels=mindspore.Tensor([1]))
    >>> logits = outputs.logits
    >>> assert logits[0, 0] < logits[0, 1]  # next sentence was random
    ```"""

    if "next_sentence_label" in kwargs:
        warnings.warn(
            "The `next_sentence_label` argument is deprecated and will be removed in a future version, use"
            " `labels` instead.",
            FutureWarning,
        )
        labels = kwargs.pop("next_sentence_label")

    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.bert(
        input_ids,
        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]

    seq_relationship_scores = self.cls(pooled_output)

    next_sentence_loss = None
    if labels is not None:
        loss_fct = CrossEntropyLoss()
        next_sentence_loss = loss_fct(seq_relationship_scores.view(-1, 2), labels.view(-1))

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

    return NextSentencePredictorOutput(
        loss=next_sentence_loss,
        logits=seq_relationship_scores,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForPreTraining

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
909
910
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
class MegatronBertForPreTraining(MegatronBertPreTrainedModel):
    _tied_weights_keys = ["cls.predictions.decoder"]

    def __init__(self, config, add_binary_head=True):
        super().__init__(config)

        self.bert = MegatronBertModel(config)
        self.cls = MegatronBertPreTrainingHeads(config)

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

    def get_output_embeddings(self):
        return self.cls.predictions.decoder

    def set_output_embeddings(self, new_embeddings):
        self.cls.predictions.decoder = new_embeddings
        self.cls.predictions.bias = new_embeddings.bias

    def forward(
        self,
        input_ids: 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,
        next_sentence_label: Optional[mindspore.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, MegatronBertForPreTrainingOutput]:
        r"""
        labels (`mindspore.Tensor` 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]`
        next_sentence_label (`mindspore.Tensor` of shape `(batch_size,)`, *optional*):
            Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair
            (see `input_ids` docstring) Indices should be in `[0, 1]`:

            - 0 indicates sequence B is a continuation of sequence A,
            - 1 indicates sequence B is a random sequence.
        kwargs (`Dict[str, any]`, *optional*, defaults to `{}`):
            Used to hide legacy arguments that have been deprecated.

        Returns:

        Example:

        ```python
        >>> from transformers import AutoTokenizer, MegatronBertForPreTraining
        >>> import torch

        >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
        >>> model = MegatronBertForPreTraining.from_pretrained("nvidia/megatron-bert-cased-345m")

        >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
        >>> outputs = model(**inputs)

        >>> prediction_logits = outputs.prediction_logits
        >>> seq_relationship_logits = outputs.seq_relationship_logits
        ```"""

        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.bert(
            input_ids,
            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, pooled_output = outputs[:2]
        prediction_scores, seq_relationship_score = self.cls(sequence_output, pooled_output)

        total_loss = None
        if labels is not None and next_sentence_label is not None:
            loss_fct = CrossEntropyLoss()
            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
            next_sentence_loss = loss_fct(seq_relationship_score.view(-1, 2), next_sentence_label.view(-1))
            total_loss = masked_lm_loss + next_sentence_loss

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

        return MegatronBertForPreTrainingOutput(
            loss=total_loss,
            prediction_logits=prediction_scores,
            seq_relationship_logits=seq_relationship_score,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForPreTraining.forward(input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, next_sentence_label=None, output_attentions=None, output_hidden_states=None, return_dict=None)

labels (mindspore.Tensor 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] next_sentence_label (mindspore.Tensor of shape (batch_size,), optional): Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair (see input_ids docstring) Indices should be in [0, 1]:

- 0 indicates sequence B is a continuation of sequence A,
- 1 indicates sequence B is a random sequence.

kwargs (Dict[str, any], optional, defaults to {}): Used to hide legacy arguments that have been deprecated.

Returns:

Example:

>>> from transformers import AutoTokenizer, MegatronBertForPreTraining
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
>>> model = MegatronBertForPreTraining.from_pretrained("nvidia/megatron-bert-cased-345m")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs)

>>> prediction_logits = outputs.prediction_logits
>>> seq_relationship_logits = outputs.seq_relationship_logits
Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
909
910
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
def forward(
    self,
    input_ids: 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,
    next_sentence_label: Optional[mindspore.Tensor] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, MegatronBertForPreTrainingOutput]:
    r"""
    labels (`mindspore.Tensor` 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]`
    next_sentence_label (`mindspore.Tensor` of shape `(batch_size,)`, *optional*):
        Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair
        (see `input_ids` docstring) Indices should be in `[0, 1]`:

        - 0 indicates sequence B is a continuation of sequence A,
        - 1 indicates sequence B is a random sequence.
    kwargs (`Dict[str, any]`, *optional*, defaults to `{}`):
        Used to hide legacy arguments that have been deprecated.

    Returns:

    Example:

    ```python
    >>> from transformers import AutoTokenizer, MegatronBertForPreTraining
    >>> import torch

    >>> tokenizer = AutoTokenizer.from_pretrained("nvidia/megatron-bert-cased-345m")
    >>> model = MegatronBertForPreTraining.from_pretrained("nvidia/megatron-bert-cased-345m")

    >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
    >>> outputs = model(**inputs)

    >>> prediction_logits = outputs.prediction_logits
    >>> seq_relationship_logits = outputs.seq_relationship_logits
    ```"""

    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.bert(
        input_ids,
        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, pooled_output = outputs[:2]
    prediction_scores, seq_relationship_score = self.cls(sequence_output, pooled_output)

    total_loss = None
    if labels is not None and next_sentence_label is not None:
        loss_fct = CrossEntropyLoss()
        masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
        next_sentence_loss = loss_fct(seq_relationship_score.view(-1, 2), next_sentence_label.view(-1))
        total_loss = masked_lm_loss + next_sentence_loss

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

    return MegatronBertForPreTrainingOutput(
        loss=total_loss,
        prediction_logits=prediction_scores,
        seq_relationship_logits=seq_relationship_score,
        hidden_states=outputs.hidden_states,
        attentions=outputs.attentions,
    )

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForPreTrainingOutput dataclass

Bases: ModelOutput

Output type of [MegatronBertForPreTraining].

PARAMETER DESCRIPTION
loss

Total loss as the sum of the masked language modeling loss and the next sequence prediction (classification) loss.

TYPE: *optional*, returned when `labels` is provided, `mindspore.Tensor` of shape `(1,)` DEFAULT: None

prediction_logits

Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

TYPE: `mindspore.Tensor` of shape `(batch_size, sequence_length, config.vocab_size)` DEFAULT: None

seq_relationship_logits

Prediction scores of the next sequence prediction (classification) head (scores of True/False continuation before SoftMax).

TYPE: `mindspore.Tensor` of shape `(batch_size, 2)` DEFAULT: None

hidden_states

Tuple of mindspore.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

TYPE: `tuple(mindspore.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True` DEFAULT: None

attentions

Tuple of mindspore.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

TYPE: `tuple(mindspore.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True` DEFAULT: None

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
@dataclass
# Copied from transformers.models.bert.modeling_bert.BertForPreTrainingOutput with Bert->MegatronBert
class MegatronBertForPreTrainingOutput(ModelOutput):
    """
    Output type of [`MegatronBertForPreTraining`].

    Args:
        loss (*optional*, returned when `labels` is provided, `mindspore.Tensor` of shape `(1,)`):
            Total loss as the sum of the masked language modeling loss and the next sequence prediction
            (classification) loss.
        prediction_logits (`mindspore.Tensor` of shape `(batch_size, sequence_length, config.vocab_size)`):
            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
        seq_relationship_logits (`mindspore.Tensor` of shape `(batch_size, 2)`):
            Prediction scores of the next sequence prediction (classification) head (scores of True/False continuation
            before SoftMax).
        hidden_states (`tuple(mindspore.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
            Tuple of `mindspore.Tensor` (one for the output of the embeddings + one for the output of each layer) of
            shape `(batch_size, sequence_length, hidden_size)`.

            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
        attentions (`tuple(mindspore.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`):
            Tuple of `mindspore.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
            sequence_length)`.

            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
            heads.
    """

    loss: Optional[mindspore.Tensor] = None
    prediction_logits: mindspore.Tensor = None
    seq_relationship_logits: mindspore.Tensor = None
    hidden_states: Optional[Tuple[mindspore.Tensor]] = None
    attentions: Optional[Tuple[mindspore.Tensor]] = None

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertForQuestionAnswering

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
1653
1654
1655
1656
1657
class MegatronBertForQuestionAnswering(MegatronBertPreTrainedModel):
    def __init__(self, config):
        super().__init__(config)
        self.num_labels = config.num_labels

        self.bert = MegatronBertModel(config, add_pooling_layer=False)
        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)

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

    def forward(
        self,
        input_ids: 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"""
        start_positions (`mindspore.Tensor` 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 (`mindspore.Tensor` 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.
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.bert(
            input_ids,
            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 = ops.split(logits, 1, dim=-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)

            loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
            start_loss = loss_fct(start_logits, start_positions)
            end_loss = loss_fct(end_logits, end_positions)
            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.megatron_bert.modeling_megatron_bert.MegatronBertForQuestionAnswering.forward(input_ids=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)

start_positions (mindspore.Tensor 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 (mindspore.Tensor 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.

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
1653
1654
1655
1656
1657
def forward(
    self,
    input_ids: 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"""
    start_positions (`mindspore.Tensor` 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 (`mindspore.Tensor` 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.
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.bert(
        input_ids,
        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 = ops.split(logits, 1, dim=-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)

        loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
        start_loss = loss_fct(start_logits, start_positions)
        end_loss = loss_fct(end_logits, end_positions)
        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.megatron_bert.modeling_megatron_bert.MegatronBertForSequenceClassification

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
class MegatronBertForSequenceClassification(MegatronBertPreTrainedModel):
    def __init__(self, config):
        super().__init__(config)
        self.num_labels = config.num_labels

        self.bert = MegatronBertModel(config)
        self.dropout = nn.Dropout(config.hidden_dropout_prob)
        self.classifier = nn.Linear(config.hidden_size, config.num_labels)

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

    def forward(
        self,
        input_ids: 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"""
        labels (`mindspore.Tensor` 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).
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.bert(
            input_ids,
            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.int64, mindspore.int32):
                    self.config.problem_type = "single_label_classification"
                else:
                    self.config.problem_type = "multi_label_classification"

            if self.config.problem_type == "regression":
                loss_fct = MSELoss()
                if self.num_labels == 1:
                    loss = loss_fct(logits.squeeze(), labels.squeeze())
                else:
                    loss = loss_fct(logits, labels)
            elif self.config.problem_type == "single_label_classification":
                loss_fct = CrossEntropyLoss()
                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
            elif self.config.problem_type == "multi_label_classification":
                loss_fct = BCEWithLogitsLoss()
                loss = loss_fct(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.megatron_bert.modeling_megatron_bert.MegatronBertForSequenceClassification.forward(input_ids=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)

labels (mindspore.Tensor 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).

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
def forward(
    self,
    input_ids: 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"""
    labels (`mindspore.Tensor` 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).
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.bert(
        input_ids,
        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.int64, mindspore.int32):
                self.config.problem_type = "single_label_classification"
            else:
                self.config.problem_type = "multi_label_classification"

        if self.config.problem_type == "regression":
            loss_fct = MSELoss()
            if self.num_labels == 1:
                loss = loss_fct(logits.squeeze(), labels.squeeze())
            else:
                loss = loss_fct(logits, labels)
        elif self.config.problem_type == "single_label_classification":
            loss_fct = CrossEntropyLoss()
            loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
        elif self.config.problem_type == "multi_label_classification":
            loss_fct = BCEWithLogitsLoss()
            loss = loss_fct(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.megatron_bert.modeling_megatron_bert.MegatronBertForTokenClassification

Bases: MegatronBertPreTrainedModel

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
class MegatronBertForTokenClassification(MegatronBertPreTrainedModel):
    def __init__(self, config):
        super().__init__(config)
        self.num_labels = config.num_labels

        self.bert = MegatronBertModel(config, add_pooling_layer=False)
        self.dropout = nn.Dropout(config.hidden_dropout_prob)
        self.classifier = nn.Linear(config.hidden_size, config.num_labels)

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

    def forward(
        self,
        input_ids: 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"""
        labels (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.bert(
            input_ids,
            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_fct = CrossEntropyLoss()
            loss = loss_fct(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.megatron_bert.modeling_megatron_bert.MegatronBertForTokenClassification.forward(input_ids=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)

labels (mindspore.Tensor of shape (batch_size, sequence_length), optional): Labels for computing the token classification loss. Indices should be in [0, ..., config.num_labels - 1].

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
def forward(
    self,
    input_ids: 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"""
    labels (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
        Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
    """
    return_dict = return_dict if return_dict is not None else self.config.use_return_dict

    outputs = self.bert(
        input_ids,
        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_fct = CrossEntropyLoss()
        loss = loss_fct(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.megatron_bert.modeling_megatron_bert.MegatronBertModel

Bases: MegatronBertPreTrainedModel

The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of cross-attention is added between the self-attention layers, following the architecture described in Attention is all you need by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.

To behave as an decoder the model needs to be initialized with the is_decoder argument of the configuration set to True. To be used in a Seq2Seq model, the model needs to initialized with both is_decoder argument and add_cross_attention set to True; an encoder_hidden_states is then expected as an input to the forward pass.

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
790
791
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
class MegatronBertModel(MegatronBertPreTrainedModel):
    """

    The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of
    cross-attention is added between the self-attention layers, following the architecture described in [Attention is
    all you need](https://arxiv.org/abs/1706.03762) by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit,
    Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.

    To behave as an decoder the model needs to be initialized with the `is_decoder` argument of the configuration set
    to `True`. To be used in a Seq2Seq model, the model needs to initialized with both `is_decoder` argument and
    `add_cross_attention` set to `True`; an `encoder_hidden_states` is then expected as an input to the forward pass.
    """

    def __init__(self, config, add_pooling_layer=True):
        super().__init__(config)
        self.config = config

        self.embeddings = MegatronBertEmbeddings(config)
        self.encoder = MegatronBertEncoder(config)

        self.pooler = MegatronBertPooler(config) if add_pooling_layer else None

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

    def get_input_embeddings(self):
        return self.embeddings.word_embeddings

    def set_input_embeddings(self, value):
        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,
        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,
        past_key_values: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]:
        r"""
        encoder_hidden_states  (`mindspore.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
            Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
            the model is configured as a decoder.
        encoder_attention_mask (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
            Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
            the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:

            - 1 for tokens that are **not masked**,
            - 0 for tokens that are **masked**.
        past_key_values (`tuple(tuple(mindspore.Tensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
            Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
        use_cache (`bool`, *optional*):
            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
            `past_key_values`).
        """
        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 self.config.is_decoder:
            use_cache = use_cache if use_cache is not None else self.config.use_cache
        else:
            use_cache = False

        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")

        batch_size, seq_length = input_shape

        # past_key_values_length
        past_key_values_length = past_key_values[0][0].shape[2] if past_key_values is not None else 0

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

        # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
        # ourselves in which case we just need to make it broadcastable to all heads.
        extended_attention_mask: mindspore.Tensor = self.get_extended_attention_mask(attention_mask, input_shape)

        # If a 2D or 3D attention mask is provided for the cross-attention
        # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
        if self.config.is_decoder and encoder_hidden_states is not None:
            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.shape
            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
            if encoder_attention_mask is None:
                encoder_attention_mask = ops.ones(encoder_hidden_shape)
            encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
        else:
            encoder_extended_attention_mask = None

        # Prepare head mask if needed
        # 1.0 in head_mask indicate we keep the head
        # attention_probs has shape bsz x n_heads x N x N
        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
        head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)

        embedding_output = self.embeddings(
            input_ids=input_ids,
            position_ids=position_ids,
            token_type_ids=token_type_ids,
            inputs_embeds=inputs_embeds,
            past_key_values_length=past_key_values_length,
        )
        encoder_outputs = self.encoder(
            embedding_output,
            attention_mask=extended_attention_mask,
            head_mask=head_mask,
            encoder_hidden_states=encoder_hidden_states,
            encoder_attention_mask=encoder_extended_attention_mask,
            past_key_values=past_key_values,
            use_cache=use_cache,
            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 self.pooler is not None else None

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

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

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertModel.forward(input_ids=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, past_key_values=None, use_cache=None, output_attentions=None, output_hidden_states=None, return_dict=None)

encoder_hidden_states (mindspore.Tensor of shape (batch_size, sequence_length, hidden_size), optional): Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder. encoder_attention_mask (mindspore.Tensor of shape (batch_size, sequence_length), optional): Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in [0, 1]:

- 1 for tokens that are **not masked**,
- 0 for tokens that are **masked**.

past_key_values (tuple(tuple(mindspore.Tensor)) of length config.n_layers with each tuple having 4 tensors of shape (batch_size, num_heads, sequence_length - 1, embed_size_per_head)): Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
`decoder_input_ids` of shape `(batch_size, sequence_length)`.

use_cache (bool, optional): If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).

Source code in mindnlp\transformers\models\megatron_bert\modeling_megatron_bert.py 
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
790
791
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
def forward(
    self,
    input_ids: 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,
    past_key_values: Optional[Tuple[Tuple[mindspore.Tensor]]] = None,
    use_cache: Optional[bool] = None,
    output_attentions: Optional[bool] = None,
    output_hidden_states: Optional[bool] = None,
    return_dict: Optional[bool] = None,
) -> Union[Tuple, BaseModelOutputWithPoolingAndCrossAttentions]:
    r"""
    encoder_hidden_states  (`mindspore.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
        Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
        the model is configured as a decoder.
    encoder_attention_mask (`mindspore.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
        Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
        the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:

        - 1 for tokens that are **not masked**,
        - 0 for tokens that are **masked**.
    past_key_values (`tuple(tuple(mindspore.Tensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
        Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

        If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
        don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
        `decoder_input_ids` of shape `(batch_size, sequence_length)`.
    use_cache (`bool`, *optional*):
        If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
        `past_key_values`).
    """
    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 self.config.is_decoder:
        use_cache = use_cache if use_cache is not None else self.config.use_cache
    else:
        use_cache = False

    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")

    batch_size, seq_length = input_shape

    # past_key_values_length
    past_key_values_length = past_key_values[0][0].shape[2] if past_key_values is not None else 0

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

    # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
    # ourselves in which case we just need to make it broadcastable to all heads.
    extended_attention_mask: mindspore.Tensor = self.get_extended_attention_mask(attention_mask, input_shape)

    # If a 2D or 3D attention mask is provided for the cross-attention
    # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
    if self.config.is_decoder and encoder_hidden_states is not None:
        encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.shape
        encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
        if encoder_attention_mask is None:
            encoder_attention_mask = ops.ones(encoder_hidden_shape)
        encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
    else:
        encoder_extended_attention_mask = None

    # Prepare head mask if needed
    # 1.0 in head_mask indicate we keep the head
    # attention_probs has shape bsz x n_heads x N x N
    # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
    # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
    head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)

    embedding_output = self.embeddings(
        input_ids=input_ids,
        position_ids=position_ids,
        token_type_ids=token_type_ids,
        inputs_embeds=inputs_embeds,
        past_key_values_length=past_key_values_length,
    )
    encoder_outputs = self.encoder(
        embedding_output,
        attention_mask=extended_attention_mask,
        head_mask=head_mask,
        encoder_hidden_states=encoder_hidden_states,
        encoder_attention_mask=encoder_extended_attention_mask,
        past_key_values=past_key_values,
        use_cache=use_cache,
        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 self.pooler is not None else None

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

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

mindnlp.transformers.models.megatron_bert.modeling_megatron_bert.MegatronBertPreTrainedModel

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\megatron_bert\modeling_megatron_bert.py 
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
class MegatronBertPreTrainedModel(PreTrainedModel):
    """
    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
    models.
    """

    config_class = MegatronBertConfig
    base_model_prefix = "bert"
    supports_gradient_checkpointing = True

    def _init_weights(self, module):
        """Initialize the weights"""
        if isinstance(module, (nn.Linear, nn.Embedding)):
            # Slightly different from the TF version which uses truncated_normal for initialization
            # cf https://github.com/pytorch/pytorch/pull/5617
            nn.init.normal_(module.weight, mean=0.0, std=self.config.initializer_range)
        elif isinstance(module, nn.LayerNorm):
            nn.init.zeros_(module.bias)
            nn.init.ones_(module.weight)
        if isinstance(module, nn.Linear) and module.bias is not None:
            nn.init.zeros_(module.bias)

mindnlp.transformers.models.megatron_bert.configuration_megatron_bert

MEGATRON_BERT model configuration

mindnlp.transformers.models.megatron_bert.configuration_megatron_bert.MegatronBertConfig

Bases: PretrainedConfig

This is the configuration class to store the configuration of a [MegatronBertModel]. It is used to instantiate a MEGATRON_BERT model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the MEGATRON_BERT nvidia/megatron-bert-uncased-345m architecture.

Configuration objects inherit from [PretrainedConfig] and can be used to control the model outputs. Read the documentation from [PretrainedConfig] for more information.

PARAMETER DESCRIPTION
vocab_size

Vocabulary size of the MEGATRON_BERT model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling [MegatronBertModel].

TYPE: `int`, *optional*, defaults to 29056 DEFAULT: 29056

hidden_size

Dimensionality of the encoder layers and the pooler layer.

TYPE: `int`, *optional*, defaults to 1024 DEFAULT: 1024

num_hidden_layers

Number of hidden layers in the Transformer encoder.

TYPE: `int`, *optional*, defaults to 24 DEFAULT: 24

num_attention_heads

Number of attention heads for each attention layer in the Transformer encoder.

TYPE: `int`, *optional*, defaults to 16 DEFAULT: 16

intermediate_size

Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.

TYPE: `int`, *optional*, defaults to 4096 DEFAULT: 4096

hidden_act

The non-linear activation function (function or string) in the encoder and pooler. If string, "gelu", "relu", "silu" and "gelu_new" are supported.

TYPE: `str` or `Callable`, *optional*, defaults to `"gelu"` DEFAULT: 'gelu'

hidden_dropout_prob

The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.

TYPE: `float`, *optional*, defaults to 0.1 DEFAULT: 0.1

attention_probs_dropout_prob

The dropout ratio for the attention probabilities.

TYPE: `float`, *optional*, defaults to 0.1 DEFAULT: 0.1

max_position_embeddings

The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048).

TYPE: `int`, *optional*, defaults to 512 DEFAULT: 512

type_vocab_size

The vocabulary size of the token_type_ids passed when calling [MegatronBertModel].

TYPE: `int`, *optional*, defaults to 2 DEFAULT: 2

initializer_range

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

TYPE: `float`, *optional*, defaults to 0.02 DEFAULT: 0.02

layer_norm_eps

The epsilon used by the layer normalization layers.

TYPE: `float`, *optional*, defaults to 1e-12 DEFAULT: 1e-12

position_embedding_type

Type of position embedding. Choose one of "absolute", "relative_key", "relative_key_query". For positional embeddings use "absolute". For more information on "relative_key", please refer to Self-Attention with Relative Position Representations (Shaw et al.). For more information on "relative_key_query", please refer to Method 4 in Improve Transformer Models with Better Relative Position Embeddings (Huang et al.).

TYPE: `str`, *optional*, defaults to `"absolute"` DEFAULT: 'absolute'

is_decoder

Whether the model is used as a decoder or not. If False, the model is used as an encoder.

TYPE: `bool`, *optional*, defaults to `False`

use_cache

Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if config.is_decoder=True.

TYPE: `bool`, *optional*, defaults to `True` DEFAULT: True

Example 
>>> from transformers import MegatronBertConfig, MegatronBertModel
...
>>> # Initializing a MEGATRON_BERT bert-base-uncased style configuration
>>> configuration = MegatronBertConfig()
...
>>> # Initializing a model (with random weights) from the bert-base-uncased style configuration
>>> model = MegatronBertModel(configuration)
...
>>> # Accessing the model configuration
>>> configuration = model.config
Source code in mindnlp\transformers\models\megatron_bert\configuration_megatron_bert.py 
 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
 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 MegatronBertConfig(PretrainedConfig):
    r"""
    This is the configuration class to store the configuration of a [`MegatronBertModel`]. It is used to instantiate a
    MEGATRON_BERT model according to the specified arguments, defining the model architecture. Instantiating a
    configuration with the defaults will yield a similar configuration to that of the MEGATRON_BERT
    [nvidia/megatron-bert-uncased-345m](https://hf-mirror.com/nvidia/megatron-bert-uncased-345m) architecture.

    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
    documentation from [`PretrainedConfig`] for more information.


    Args:
        vocab_size (`int`, *optional*, defaults to 29056):
            Vocabulary size of the MEGATRON_BERT model. Defines the number of different tokens that can be represented
            by the `inputs_ids` passed when calling [`MegatronBertModel`].
        hidden_size (`int`, *optional*, defaults to 1024):
            Dimensionality of the encoder layers and the pooler layer.
        num_hidden_layers (`int`, *optional*, defaults to 24):
            Number of hidden layers in the Transformer encoder.
        num_attention_heads (`int`, *optional*, defaults to 16):
            Number of attention heads for each attention layer in the Transformer encoder.
        intermediate_size (`int`, *optional*, defaults to 4096):
            Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
        hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`):
            The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
            `"relu"`, `"silu"` and `"gelu_new"` are supported.
        hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
            The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
        attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
            The dropout ratio for the attention probabilities.
        max_position_embeddings (`int`, *optional*, defaults to 512):
            The maximum sequence length that this model might ever be used with. Typically set this to something large
            just in case (e.g., 512 or 1024 or 2048).
        type_vocab_size (`int`, *optional*, defaults to 2):
            The vocabulary size of the `token_type_ids` passed when calling [`MegatronBertModel`].
        initializer_range (`float`, *optional*, defaults to 0.02):
            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
        layer_norm_eps (`float`, *optional*, defaults to 1e-12):
            The epsilon used by the layer normalization layers.
        position_embedding_type (`str`, *optional*, defaults to `"absolute"`):
            Type of position embedding. Choose one of `"absolute"`, `"relative_key"`, `"relative_key_query"`. For
            positional embeddings use `"absolute"`. For more information on `"relative_key"`, please refer to
            [Self-Attention with Relative Position Representations (Shaw et al.)](https://arxiv.org/abs/1803.02155).
            For more information on `"relative_key_query"`, please refer to *Method 4* in [Improve Transformer Models
            with Better Relative Position Embeddings (Huang et al.)](https://arxiv.org/abs/2009.13658).
        is_decoder (`bool`, *optional*, defaults to `False`):
            Whether the model is used as a decoder or not. If `False`, the model is used as an encoder.
        use_cache (`bool`, *optional*, defaults to `True`):
            Whether or not the model should return the last key/values attentions (not used by all models). Only
            relevant if `config.is_decoder=True`.

    Example:
        ```python
        >>> from transformers import MegatronBertConfig, MegatronBertModel
        ...
        >>> # Initializing a MEGATRON_BERT bert-base-uncased style configuration
        >>> configuration = MegatronBertConfig()
        ...
        >>> # Initializing a model (with random weights) from the bert-base-uncased style configuration
        >>> model = MegatronBertModel(configuration)
        ...
        >>> # Accessing the model configuration
        >>> configuration = model.config
        ```
    """
    model_type = "megatron-bert"

    def __init__(
        self,
        vocab_size=29056,
        hidden_size=1024,
        num_hidden_layers=24,
        num_attention_heads=16,
        intermediate_size=4096,
        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,
        **kwargs,
    ):
        """
        Initialize a MegatronBertConfig object with the provided parameters.

        Args:
            vocab_size (int): The size of the vocabulary used for tokenization.
            hidden_size (int): The size of the hidden layers in the model.
            num_hidden_layers (int): The number of hidden layers in the model.
            num_attention_heads (int): The number of attention heads in the model.
            intermediate_size (int): The size of the intermediate (feed-forward) layer.
            hidden_act (str): The activation function used in the hidden layers.
            hidden_dropout_prob (float): The dropout probability for the hidden layers.
            attention_probs_dropout_prob (float): The dropout probability for attention probabilities.
            max_position_embeddings (int): The maximum length of input sequences.
            type_vocab_size (int): The size of the token type embeddings.
            initializer_range (float): The range for parameter initializations.
            layer_norm_eps (float): The epsilon value for layer normalization.
            pad_token_id (int): The ID of the padding token.
            position_embedding_type (str): The type of position embeddings used.
            use_cache (bool): Whether to use caching during inference.

        Returns:
            None.

        Raises:
            ValueError: If any argument is invalid or out of range.
        """
        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

mindnlp.transformers.models.megatron_bert.configuration_megatron_bert.MegatronBertConfig.__init__(vocab_size=29056, hidden_size=1024, num_hidden_layers=24, num_attention_heads=16, intermediate_size=4096, 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, **kwargs)

Initialize a MegatronBertConfig object with the provided parameters.

PARAMETER DESCRIPTION
vocab_size

The size of the vocabulary used for tokenization.

TYPE: int DEFAULT: 29056

hidden_size

The size of the hidden layers in the model.

TYPE: int DEFAULT: 1024

num_hidden_layers

The number of hidden layers in the model.

TYPE: int DEFAULT: 24

num_attention_heads

The number of attention heads in the model.

TYPE: int DEFAULT: 16

intermediate_size

The size of the intermediate (feed-forward) layer.

TYPE: int DEFAULT: 4096

hidden_act

The activation function used in the hidden layers.

TYPE: str DEFAULT: 'gelu'

hidden_dropout_prob

The dropout probability for the hidden layers.

TYPE: float DEFAULT: 0.1

attention_probs_dropout_prob

The dropout probability for attention probabilities.

TYPE: float DEFAULT: 0.1

max_position_embeddings

The maximum length of input sequences.

TYPE: int DEFAULT: 512

type_vocab_size

The size of the token type embeddings.

TYPE: int DEFAULT: 2

initializer_range

The range for parameter initializations.

TYPE: float DEFAULT: 0.02

layer_norm_eps

The epsilon value for layer normalization.

TYPE: float DEFAULT: 1e-12

pad_token_id

The ID of the padding token.

TYPE: int DEFAULT: 0

position_embedding_type

The type of position embeddings used.

TYPE: str DEFAULT: 'absolute'

use_cache

Whether to use caching during inference.

TYPE: bool DEFAULT: True

RETURNS DESCRIPTION

None.
RAISES DESCRIPTION
ValueError

If any argument is invalid or out of range.
Source code in mindnlp\transformers\models\megatron_bert\configuration_megatron_bert.py 
 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 __init__(
    self,
    vocab_size=29056,
    hidden_size=1024,
    num_hidden_layers=24,
    num_attention_heads=16,
    intermediate_size=4096,
    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,
    **kwargs,
):
    """
    Initialize a MegatronBertConfig object with the provided parameters.

    Args:
        vocab_size (int): The size of the vocabulary used for tokenization.
        hidden_size (int): The size of the hidden layers in the model.
        num_hidden_layers (int): The number of hidden layers in the model.
        num_attention_heads (int): The number of attention heads in the model.
        intermediate_size (int): The size of the intermediate (feed-forward) layer.
        hidden_act (str): The activation function used in the hidden layers.
        hidden_dropout_prob (float): The dropout probability for the hidden layers.
        attention_probs_dropout_prob (float): The dropout probability for attention probabilities.
        max_position_embeddings (int): The maximum length of input sequences.
        type_vocab_size (int): The size of the token type embeddings.
        initializer_range (float): The range for parameter initializations.
        layer_norm_eps (float): The epsilon value for layer normalization.
        pad_token_id (int): The ID of the padding token.
        position_embedding_type (str): The type of position embeddings used.
        use_cache (bool): Whether to use caching during inference.

    Returns:
        None.

    Raises:
        ValueError: If any argument is invalid or out of range.
    """
    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