Reorganize repo (#8580)
* Put models in subfolders * Styling * Fix imports in tests * More fixes in test imports * Sneaky hidden imports * Fix imports in doc files * More sneaky imports * Finish fixing tests * Fix examples * Fix path for copies * More fixes for examples * Fix dummy files * More fixes for example * More model import fixes * Is this why you're unhappy GitHub? * Fix imports in conver command
This commit is contained in:
Sylvain Gugger
17
src/transformers/models/bert_generation/__init__.py
Normal file
17
src/transformers/models/bert_generation/__init__.py
Normal file
@@ -0,0 +1,17 @@
|
||||
# flake8: noqa
|
||||
# There's no way to ignore "F401 '...' imported but unused" warnings in this
|
||||
# module, but to preserve other warnings. So, don't check this module at all.
|
||||
|
||||
from ...file_utils import is_sentencepiece_available, is_torch_available
|
||||
from .configuration_bert_generation import BertGenerationConfig
|
||||
|
||||
|
||||
if is_sentencepiece_available():
|
||||
from .tokenization_bert_generation import BertGenerationTokenizer
|
||||
|
||||
if is_torch_available():
|
||||
from .modeling_bert_generation import (
|
||||
BertGenerationDecoder,
|
||||
BertGenerationEncoder,
|
||||
load_tf_weights_in_bert_generation,
|
||||
)
|
||||
@@ -0,0 +1,105 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2020 The Google AI Language Team Authors and The HuggingFace Inc. team.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
""" BertGeneration model configuration """
|
||||
|
||||
from ...configuration_utils import PretrainedConfig
|
||||
|
||||
|
||||
class BertGenerationConfig(PretrainedConfig):
|
||||
r"""
|
||||
This is the configuration class to store the configuration of a
|
||||
:class:`~transformers.BertGenerationPreTrainedModel`. It is used to instantiate a BertGeneration model according to
|
||||
the specified arguments, defining the model architecture.
|
||||
|
||||
Configuration objects inherit from :class:`~transformers.PretrainedConfig` and can be used to control the model
|
||||
outputs. Read the documentation from :class:`~transformers.PretrainedConfig` for more information.
|
||||
|
||||
Args:
|
||||
vocab_size (:obj:`int`, `optional`, defaults to 50358):
|
||||
Vocabulary size of the BERT model. Defines the number of different tokens that can be represented by the
|
||||
:obj:`inputs_ids` passed when calling :class:`~transformers.BertGeneration`.
|
||||
hidden_size (:obj:`int`, `optional`, defaults to 1024):
|
||||
Dimensionality of the encoder layers and the pooler layer.
|
||||
num_hidden_layers (:obj:`int`, `optional`, defaults to 24):
|
||||
Number of hidden layers in the Transformer encoder.
|
||||
num_attention_heads (:obj:`int`, `optional`, defaults to 16):
|
||||
Number of attention heads for each attention layer in the Transformer encoder.
|
||||
intermediate_size (:obj:`int`, `optional`, defaults to 3072):
|
||||
Dimensionality of the "intermediate" (often called feed-forward) layer in the Transformer encoder.
|
||||
hidden_act (:obj:`str` or :obj:`function`, `optional`, defaults to :obj:`"gelu"`):
|
||||
The non-linear activation function (function or string) in the encoder and pooler. If string,
|
||||
:obj:`"gelu"`, :obj:`"relu"`, :obj:`"silu"` and :obj:`"gelu_new"` are supported.
|
||||
hidden_dropout_prob (:obj:`float`, `optional`, defaults to 0.1):
|
||||
The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
|
||||
attention_probs_dropout_prob (:obj:`float`, `optional`, defaults to 0.1):
|
||||
The dropout ratio for the attention probabilities.
|
||||
max_position_embeddings (:obj:`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).
|
||||
initializer_range (:obj:`float`, `optional`, defaults to 0.02):
|
||||
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
|
||||
layer_norm_eps (:obj:`float`, `optional`, defaults to 1e-12):
|
||||
The epsilon used by the layer normalization layers.
|
||||
gradient_checkpointing (:obj:`bool`, `optional`, defaults to :obj:`False`):
|
||||
If :obj:`True`, use gradient checkpointing to save memory at the expense of slower backward pass.
|
||||
|
||||
Examples::
|
||||
|
||||
>>> from transformers import BertGenerationConfig, BertGenerationEncoder
|
||||
|
||||
>>> # Initializing a BertGeneration config
|
||||
>>> configuration = BertGenerationConfig()
|
||||
|
||||
>>> # Initializing a model from the config
|
||||
>>> model = BertGenerationEncoder(configuration)
|
||||
|
||||
>>> # Accessing the model configuration
|
||||
>>> configuration = model.config
|
||||
"""
|
||||
model_type = "bert-generation"
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_size=50358,
|
||||
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,
|
||||
initializer_range=0.02,
|
||||
layer_norm_eps=1e-12,
|
||||
pad_token_id=0,
|
||||
bos_token_id=2,
|
||||
eos_token_id=1,
|
||||
gradient_checkpointing=False,
|
||||
**kwargs
|
||||
):
|
||||
super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_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.initializer_range = initializer_range
|
||||
self.layer_norm_eps = layer_norm_eps
|
||||
self.gradient_checkpointing = gradient_checkpointing
|
||||
518
src/transformers/models/bert_generation/modeling_bert_generation.py
Executable file
518
src/transformers/models/bert_generation/modeling_bert_generation.py
Executable file
@@ -0,0 +1,518 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2020 The Google AI Language Team Authors and The HuggingFace Inc. team.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
"""PyTorch BERT model specific for generation. """
|
||||
|
||||
|
||||
import torch
|
||||
import torch.utils.checkpoint
|
||||
from torch import nn
|
||||
from torch.nn import CrossEntropyLoss
|
||||
|
||||
from ...file_utils import (
|
||||
add_code_sample_docstrings,
|
||||
add_start_docstrings,
|
||||
add_start_docstrings_to_model_forward,
|
||||
replace_return_docstrings,
|
||||
)
|
||||
from ...modeling_outputs import BaseModelOutputWithCrossAttentions, CausalLMOutputWithCrossAttentions
|
||||
from ...modeling_utils import PreTrainedModel
|
||||
from ...utils import logging
|
||||
from ..bert.modeling_bert import BertEncoder
|
||||
from .configuration_bert_generation import BertGenerationConfig
|
||||
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
_CONFIG_FOR_DOC = "BertGenerationConfig"
|
||||
_TOKENIZER_FOR_DOC = "BertGenerationTokenizer"
|
||||
|
||||
|
||||
def load_tf_weights_in_bert_generation(
|
||||
model, tf_hub_path, model_class, is_encoder_named_decoder=False, is_encoder=False
|
||||
):
|
||||
try:
|
||||
import numpy as np
|
||||
import tensorflow.compat.v1 as tf
|
||||
|
||||
import tensorflow_hub as hub
|
||||
import tensorflow_text # noqa: F401
|
||||
|
||||
tf.disable_eager_execution()
|
||||
except ImportError:
|
||||
logger.error(
|
||||
"Loading a TensorFlow model in PyTorch, requires TensorFlow to be installed. Please see "
|
||||
"https://www.tensorflow.org/install/ for installation instructions."
|
||||
)
|
||||
raise
|
||||
tf_model = hub.Module(tf_hub_path)
|
||||
init = tf.global_variables_initializer()
|
||||
with tf.Session() as sess:
|
||||
init.run()
|
||||
all_variables = tf_model.variable_map
|
||||
keep_track_variables = all_variables.copy()
|
||||
for key in list(all_variables.keys()):
|
||||
if "global" in key:
|
||||
logger.info(f"Skipping {key}...")
|
||||
continue
|
||||
if not is_encoder:
|
||||
model_pointer = getattr(model, model_class)
|
||||
else:
|
||||
model_pointer = model
|
||||
is_embedding = False
|
||||
logger.info(f"Trying to match {key}...")
|
||||
# remove start_string = "module/bert/"
|
||||
sub_layers = key.split("/")[2:]
|
||||
if is_encoder_named_decoder and sub_layers[0] == "encoder":
|
||||
logger.info(f"Skipping encoder layer {key} for decoder")
|
||||
continue
|
||||
if is_encoder and sub_layers[0] == "decoder":
|
||||
logger.info(f"Skipping decoder layer {key} for encoder")
|
||||
continue
|
||||
for i, sub_layer in enumerate(sub_layers):
|
||||
if sub_layer == "embeddings":
|
||||
is_embedding = True
|
||||
elif sub_layer == "LayerNorm":
|
||||
is_embedding = False
|
||||
if "layer" in sub_layer:
|
||||
model_pointer = model_pointer.layer[int(sub_layer.split("_")[-1])]
|
||||
elif sub_layer in ["kernel", "gamma"]:
|
||||
model_pointer = model_pointer.weight
|
||||
elif sub_layer == "beta":
|
||||
model_pointer = model_pointer.bias
|
||||
elif sub_layer == "encdec":
|
||||
model_pointer = model_pointer.crossattention.self
|
||||
elif sub_layer == "encdec_output":
|
||||
model_pointer = model_pointer.crossattention.output
|
||||
elif is_encoder_named_decoder and sub_layer == "decoder":
|
||||
model_pointer = model_pointer.encoder
|
||||
else:
|
||||
if sub_layer == "attention" and "encdec" in sub_layers[i + 1]:
|
||||
continue
|
||||
try:
|
||||
model_pointer = getattr(model_pointer, sub_layer)
|
||||
except AttributeError:
|
||||
logger.info(f"Skipping to initialize {key} at {sub_layer}...")
|
||||
raise AttributeError
|
||||
|
||||
array = np.asarray(sess.run(all_variables[key]))
|
||||
if not is_embedding:
|
||||
logger.info("Transposing numpy weight of shape {} for {}".format(array.shape, key))
|
||||
array = np.transpose(array)
|
||||
else:
|
||||
model_pointer = model_pointer.weight
|
||||
|
||||
try:
|
||||
assert (
|
||||
model_pointer.shape == array.shape
|
||||
), f"Pointer shape {model_pointer.shape} and array shape {array.shape} mismatched"
|
||||
except AssertionError as e:
|
||||
e.args += (model_pointer.shape, array.shape)
|
||||
raise
|
||||
logger.info(f"Initialize PyTorch weight {key}")
|
||||
|
||||
model_pointer.data = torch.from_numpy(array.astype(np.float32))
|
||||
keep_track_variables.pop(key, None)
|
||||
|
||||
logger.info("Weights not copied to PyTorch model: {}".format(", ".join(keep_track_variables.keys())))
|
||||
return model
|
||||
|
||||
|
||||
class BertGenerationEmbeddings(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.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
|
||||
# any TensorFlow checkpoint file
|
||||
self.LayerNorm = torch.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", torch.arange(config.max_position_embeddings).expand((1, -1)))
|
||||
|
||||
def forward(self, input_ids=None, position_ids=None, inputs_embeds=None):
|
||||
if input_ids is not None:
|
||||
input_shape = input_ids.size()
|
||||
else:
|
||||
input_shape = inputs_embeds.size()[:-1]
|
||||
|
||||
seq_length = input_shape[1]
|
||||
|
||||
if position_ids is None:
|
||||
position_ids = self.position_ids[:, :seq_length]
|
||||
|
||||
if inputs_embeds is None:
|
||||
inputs_embeds = self.word_embeddings(input_ids)
|
||||
position_embeddings = self.position_embeddings(position_ids)
|
||||
|
||||
embeddings = inputs_embeds + position_embeddings
|
||||
embeddings = self.LayerNorm(embeddings)
|
||||
embeddings = self.dropout(embeddings)
|
||||
return embeddings
|
||||
|
||||
|
||||
class BertGenerationPreTrainedModel(PreTrainedModel):
|
||||
"""
|
||||
An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
|
||||
models.
|
||||
"""
|
||||
|
||||
config_class = BertGenerationConfig
|
||||
base_model_prefix = "bert"
|
||||
authorized_missing_keys = [r"position_ids"]
|
||||
|
||||
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
|
||||
module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
|
||||
elif isinstance(module, nn.LayerNorm):
|
||||
module.bias.data.zero_()
|
||||
module.weight.data.fill_(1.0)
|
||||
if isinstance(module, nn.Linear) and module.bias is not None:
|
||||
module.bias.data.zero_()
|
||||
|
||||
|
||||
BERT_GENERATION_START_DOCSTRING = r"""
|
||||
|
||||
This model inherits from :class:`~transformers.PreTrainedModel`. Check the superclass documentation for the generic
|
||||
methods the library implements for all its model (such as downloading or saving, resizing the input embeddings,
|
||||
pruning heads etc.)
|
||||
|
||||
This model is also a PyTorch `torch.nn.Module <https://pytorch.org/docs/stable/nn.html#torch.nn.Module>`__
|
||||
subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to
|
||||
general usage and behavior.
|
||||
|
||||
Parameters:
|
||||
config (:class:`~transformers.BertGenerationConfig`): Model configuration class with all the parameters of the model.
|
||||
Initializing with a config file does not load the weights associated with the model, only the
|
||||
configuration. Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model
|
||||
weights.
|
||||
"""
|
||||
|
||||
BERT_GENERATION_INPUTS_DOCSTRING = r"""
|
||||
Args:
|
||||
input_ids (:obj:`torch.LongTensor` of shape :obj:`({0})`):
|
||||
Indices of input sequence tokens in the vocabulary.
|
||||
|
||||
Indices can be obtained using :class:`~transformers.BertGenerationTokenizer`. See
|
||||
:meth:`transformers.PreTrainedTokenizer.__call__` and :meth:`transformers.PreTrainedTokenizer.encode` for
|
||||
details.
|
||||
|
||||
`What are input IDs? <../glossary.html#input-ids>`__
|
||||
attention_mask (:obj:`torch.FloatTensor` of shape :obj:`({0})`, `optional`):
|
||||
Mask to avoid performing attention on padding token indices. Mask values selected in ``[0, 1]``:
|
||||
|
||||
- 1 for tokens that are **not masked**,
|
||||
- 0 for tokens that are **masked**.
|
||||
|
||||
`What are attention masks? <../glossary.html#attention-mask>`__
|
||||
position_ids (:obj:`torch.LongTensor` of shape :obj:`({0})`, `optional`):
|
||||
Indices of positions of each input sequence tokens in the position embeddings. Selected in the range ``[0,
|
||||
config.max_position_embeddings - 1]``.
|
||||
|
||||
`What are position IDs? <../glossary.html#position-ids>`_
|
||||
head_mask (:obj:`torch.FloatTensor` of shape :obj:`(num_heads,)` or :obj:`(num_layers, num_heads)`, `optional`):
|
||||
Mask to nullify selected heads of the self-attention modules. Mask values selected in ``[0, 1]``:
|
||||
|
||||
- 1 indicates the head is **not masked**,
|
||||
- 0 indicates the head is **masked**.
|
||||
|
||||
inputs_embeds (:obj:`torch.FloatTensor` of shape :obj:`({0}, hidden_size)`, `optional`):
|
||||
Optionally, instead of passing :obj:`input_ids` you can choose to directly pass an embedded representation.
|
||||
This is useful if you want more control over how to convert :obj:`input_ids` indices into associated
|
||||
vectors than the model's internal embedding lookup matrix.
|
||||
output_attentions (:obj:`bool`, `optional`):
|
||||
Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under returned
|
||||
tensors for more detail.
|
||||
output_hidden_states (:obj:`bool`, `optional`):
|
||||
Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors for
|
||||
more detail.
|
||||
return_dict (:obj:`bool`, `optional`):
|
||||
Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
|
||||
"""
|
||||
|
||||
|
||||
@add_start_docstrings(
|
||||
"The bare BertGeneration model transformer outputting raw hidden-states without any specific head on top.",
|
||||
BERT_GENERATION_START_DOCSTRING,
|
||||
)
|
||||
class BertGenerationEncoder(BertGenerationPreTrainedModel):
|
||||
"""
|
||||
|
||||
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.
|
||||
|
||||
This model should be used when leveraging Bert or Roberta checkpoints for the
|
||||
:class:`~transformers.EncoderDecoderModel` class as described in `Leveraging Pre-trained Checkpoints for Sequence
|
||||
Generation Tasks <https://arxiv.org/abs/1907.12461>`__ by Sascha Rothe, Shashi Narayan, and Aliaksei Severyn.
|
||||
|
||||
To behave as an decoder the model needs to be initialized with the :obj:`is_decoder` argument of the configuration
|
||||
set to :obj:`True`. To be used in a Seq2Seq model, the model needs to initialized with both :obj:`is_decoder`
|
||||
argument and :obj:`add_cross_attention` set to :obj:`True`; an :obj:`encoder_hidden_states` is then expected as an
|
||||
input to the forward pass.
|
||||
"""
|
||||
|
||||
def __init__(self, config):
|
||||
super().__init__(config)
|
||||
self.config = config
|
||||
|
||||
self.embeddings = BertGenerationEmbeddings(config)
|
||||
self.encoder = BertEncoder(config)
|
||||
|
||||
self.init_weights()
|
||||
|
||||
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)
|
||||
|
||||
@add_start_docstrings_to_model_forward(BERT_GENERATION_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
|
||||
@add_code_sample_docstrings(
|
||||
tokenizer_class=_TOKENIZER_FOR_DOC,
|
||||
checkpoint="google/bert_for_seq_generation_L-24_bbc_encoder",
|
||||
output_type=BaseModelOutputWithCrossAttentions,
|
||||
config_class=_CONFIG_FOR_DOC,
|
||||
)
|
||||
def forward(
|
||||
self,
|
||||
input_ids=None,
|
||||
attention_mask=None,
|
||||
position_ids=None,
|
||||
head_mask=None,
|
||||
inputs_embeds=None,
|
||||
encoder_hidden_states=None,
|
||||
encoder_attention_mask=None,
|
||||
output_attentions=None,
|
||||
output_hidden_states=None,
|
||||
return_dict=None,
|
||||
):
|
||||
r"""
|
||||
encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(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 (:obj:`torch.FloatTensor` of shape :obj:`(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 MASKED tokens.
|
||||
"""
|
||||
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
|
||||
output_hidden_states = (
|
||||
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
|
||||
)
|
||||
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
||||
|
||||
if input_ids is not None and inputs_embeds is not None:
|
||||
raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
|
||||
elif input_ids is not None:
|
||||
input_shape = input_ids.size()
|
||||
elif inputs_embeds is not None:
|
||||
input_shape = inputs_embeds.size()[:-1]
|
||||
else:
|
||||
raise ValueError("You have to specify either input_ids or inputs_embeds")
|
||||
|
||||
device = input_ids.device if input_ids is not None else inputs_embeds.device
|
||||
|
||||
if attention_mask is None:
|
||||
attention_mask = torch.ones(input_shape, device=device)
|
||||
|
||||
# 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: torch.Tensor = self.get_extended_attention_mask(attention_mask, input_shape, device)
|
||||
|
||||
# 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.size()
|
||||
encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
|
||||
if encoder_attention_mask is None:
|
||||
encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
|
||||
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, inputs_embeds=inputs_embeds)
|
||||
|
||||
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,
|
||||
output_attentions=output_attentions,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict,
|
||||
)
|
||||
sequence_output = encoder_outputs[0]
|
||||
|
||||
if not return_dict:
|
||||
return (sequence_output,) + encoder_outputs[1:]
|
||||
|
||||
return BaseModelOutputWithCrossAttentions(
|
||||
last_hidden_state=sequence_output,
|
||||
hidden_states=encoder_outputs.hidden_states,
|
||||
attentions=encoder_outputs.attentions,
|
||||
cross_attentions=encoder_outputs.cross_attentions,
|
||||
)
|
||||
|
||||
|
||||
class BertGenerationOnlyLMHead(nn.Module):
|
||||
def __init__(self, config):
|
||||
super().__init__()
|
||||
self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
|
||||
self.bias = nn.Parameter(torch.zeros(config.vocab_size))
|
||||
|
||||
# Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
|
||||
self.decoder.bias = self.bias
|
||||
|
||||
def forward(self, hidden_states):
|
||||
logits = self.decoder(hidden_states)
|
||||
return logits
|
||||
|
||||
|
||||
@add_start_docstrings(
|
||||
"""BertGeneration Model with a `language modeling` head on top for CLM fine-tuning. """,
|
||||
BERT_GENERATION_START_DOCSTRING,
|
||||
)
|
||||
class BertGenerationDecoder(BertGenerationPreTrainedModel):
|
||||
def __init__(self, config):
|
||||
super().__init__(config)
|
||||
|
||||
if not config.is_decoder:
|
||||
logger.warn("If you want to use `BertGenerationDecoder` as a standalone, add `is_decoder=True.`")
|
||||
|
||||
self.bert = BertGenerationEncoder(config)
|
||||
self.lm_head = BertGenerationOnlyLMHead(config)
|
||||
|
||||
self.init_weights()
|
||||
|
||||
def get_output_embeddings(self):
|
||||
return self.lm_head.decoder
|
||||
|
||||
@add_start_docstrings_to_model_forward(BERT_GENERATION_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
|
||||
@replace_return_docstrings(output_type=CausalLMOutputWithCrossAttentions, config_class=_CONFIG_FOR_DOC)
|
||||
def forward(
|
||||
self,
|
||||
input_ids=None,
|
||||
attention_mask=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,
|
||||
):
|
||||
r"""
|
||||
encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(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 (:obj:`torch.FloatTensor` of shape :obj:`(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 (:obj:`torch.LongTensor` of shape :obj:`(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 in ``[0, ..., config.vocab_size]``
|
||||
|
||||
Returns:
|
||||
|
||||
Example::
|
||||
|
||||
>>> from transformers import BertGenerationTokenizer, BertGenerationDecoder, BertGenerationConfig
|
||||
>>> import torch
|
||||
|
||||
>>> tokenizer = BertGenerationTokenizer.from_pretrained('google/bert_for_seq_generation_L-24_bbc_encoder')
|
||||
>>> config = BertGenerationConfig.from_pretrained("google/bert_for_seq_generation_L-24_bbc_encoder")
|
||||
>>> config.is_decoder = True
|
||||
>>> model = BertGenerationDecoder.from_pretrained('google/bert_for_seq_generation_L-24_bbc_encoder', config=config)
|
||||
|
||||
>>> 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
|
||||
|
||||
outputs = self.bert(
|
||||
input_ids,
|
||||
attention_mask=attention_mask,
|
||||
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.lm_head(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, :].contiguous()
|
||||
labels = labels[:, 1:].contiguous()
|
||||
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[1:]
|
||||
return ((lm_loss,) + output) if lm_loss is not None else output
|
||||
|
||||
return CausalLMOutputWithCrossAttentions(
|
||||
loss=lm_loss,
|
||||
logits=prediction_scores,
|
||||
hidden_states=outputs.hidden_states,
|
||||
attentions=outputs.attentions,
|
||||
cross_attentions=outputs.cross_attentions,
|
||||
)
|
||||
|
||||
def prepare_inputs_for_generation(self, input_ids, 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)
|
||||
|
||||
return {"input_ids": input_ids, "attention_mask": attention_mask}
|
||||
@@ -0,0 +1,139 @@
|
||||
# coding=utf-8
|
||||
# Copyright (c) 2020, NVIDIA CORPORATION. All rights reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
""" Tokenization class for model BertGeneration."""
|
||||
|
||||
|
||||
import os
|
||||
from shutil import copyfile
|
||||
from typing import List, Optional, Tuple
|
||||
|
||||
import sentencepiece as spm
|
||||
|
||||
from ...tokenization_utils import PreTrainedTokenizer
|
||||
from ...utils import logging
|
||||
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
VOCAB_FILES_NAMES = {"vocab_file": "spiece.model"}
|
||||
|
||||
tokenizer_url = "https://huggingface.co/google/bert_for_seq_generation_L-24_bbc_encoder/resolve/main/spiece.model"
|
||||
|
||||
|
||||
class BertGenerationTokenizer(PreTrainedTokenizer):
|
||||
"""
|
||||
Construct a BertGeneration tokenizer. Based on `SentencePiece <https://github.com/google/sentencepiece>`__.
|
||||
|
||||
This tokenizer inherits from :class:`~transformers.PreTrainedTokenizer` which contains most of the main methods.
|
||||
Users should refer to this superclass for more information regarding those methods.
|
||||
|
||||
Args:
|
||||
vocab_file (:obj:`str`):
|
||||
`SentencePiece <https://github.com/google/sentencepiece>`__ file (generally has a `.spm` extension) that
|
||||
contains the vocabulary necessary to instantiate a tokenizer.
|
||||
eos_token (:obj:`str`, `optional`, defaults to :obj:`"</s>"`):
|
||||
The end of sequence token.
|
||||
bos_token (:obj:`str`, `optional`, defaults to :obj:`"<s>"`):
|
||||
The begin of sequence token.
|
||||
unk_token (:obj:`str`, `optional`, defaults to :obj:`"<unk>"`):
|
||||
The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
|
||||
token instead.
|
||||
pad_token (:obj:`str`, `optional`, defaults to :obj:`"<pad>"`):
|
||||
The token used for padding, for example when batching sequences of different lengths.
|
||||
"""
|
||||
|
||||
vocab_files_names = VOCAB_FILES_NAMES
|
||||
pretrained_vocab_files_map = {"vocab_file": {"bert_for_seq_generation": tokenizer_url}}
|
||||
max_model_input_sizes = {"bert_for_seq_generation": 512}
|
||||
prefix_tokens: List[int] = []
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_file,
|
||||
bos_token="<s>",
|
||||
eos_token="</s>",
|
||||
unk_token="<unk>",
|
||||
pad_token="<pad>",
|
||||
sep_token="<::::>",
|
||||
**kwargs
|
||||
):
|
||||
# Add extra_ids to the special token list
|
||||
super().__init__(
|
||||
bos_token=bos_token,
|
||||
eos_token=eos_token,
|
||||
unk_token=unk_token,
|
||||
pad_token=pad_token,
|
||||
sep_token=sep_token,
|
||||
**kwargs,
|
||||
)
|
||||
|
||||
self.vocab_file = vocab_file
|
||||
|
||||
self.sp_model = spm.SentencePieceProcessor()
|
||||
self.sp_model.Load(vocab_file)
|
||||
|
||||
@property
|
||||
def vocab_size(self):
|
||||
return self.sp_model.get_piece_size()
|
||||
|
||||
def get_vocab(self):
|
||||
vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
|
||||
vocab.update(self.added_tokens_encoder)
|
||||
return vocab
|
||||
|
||||
def __getstate__(self):
|
||||
state = self.__dict__.copy()
|
||||
state["sp_model"] = None
|
||||
return state
|
||||
|
||||
def __setstate__(self, d):
|
||||
self.__dict__ = d
|
||||
self.sp_model = spm.SentencePieceProcessor()
|
||||
self.sp_model.Load(self.vocab_file)
|
||||
|
||||
def _tokenize(self, text, sample=False):
|
||||
"""Take as input a string and return a list of strings (tokens) for words/sub-words"""
|
||||
if not sample:
|
||||
pieces = self.sp_model.EncodeAsPieces(text)
|
||||
else:
|
||||
pieces = self.sp_model.SampleEncodeAsPieces(text, 64, 0.1)
|
||||
return pieces
|
||||
|
||||
def _convert_token_to_id(self, token):
|
||||
""" Converts a token (str) in an id using the vocab. """
|
||||
return self.sp_model.piece_to_id(token)
|
||||
|
||||
def _convert_id_to_token(self, index):
|
||||
"""Converts an index (integer) in a token (str) using the vocab."""
|
||||
token = self.sp_model.IdToPiece(index)
|
||||
return token
|
||||
|
||||
def convert_tokens_to_string(self, tokens):
|
||||
""" Converts a sequence of tokens (string) in a single string. """
|
||||
out_string = self.sp_model.decode_pieces(tokens)
|
||||
return out_string
|
||||
|
||||
def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
|
||||
if not os.path.isdir(save_directory):
|
||||
logger.error("Vocabulary path ({}) should be a directory".format(save_directory))
|
||||
return
|
||||
out_vocab_file = os.path.join(
|
||||
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
|
||||
)
|
||||
|
||||
if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
|
||||
copyfile(self.vocab_file, out_vocab_file)
|
||||
|
||||
return (out_vocab_file,)
|
||||
Reference in New Issue
Block a user