defmodule Bumblebee.Text.Distilbert do
alias Bumblebee.Shared
options =
[
vocab_size: [
default: 30522,
doc: """
the vocabulary size of the token embedding. This corresponds to the number of distinct
tokens that can be represented in model input and output
"""
],
max_positions: [
default: 512,
doc: """
the vocabulary size of the position embedding. This corresponds to the maximum sequence
length that this model can process. Typically this is set to a large value just in case,
such as 512, 1024 or 2048
"""
],
hidden_size: [
default: 768,
doc: "the dimensionality of hidden layers"
],
num_blocks: [
default: 6,
doc: "the number of Transformer blocks in the encoder"
],
num_attention_heads: [
default: 12,
doc: "the number of attention heads for each attention layer in the encoder"
],
intermediate_size: [
default: 3072,
doc:
"the dimensionality of the intermediate layer in the transformer feed-forward network (FFN) in the encoder"
],
activation: [
default: :gelu,
doc: "the activation function"
],
dropout_rate: [
default: 0.1,
doc: "the dropout rate for embedding and encoder"
],
attention_dropout_rate: [
default: 0.1,
doc: "the dropout rate for attention weights"
],
classifier_dropout_rate: [
default: nil,
doc:
"the dropout rate for the classification head. If not specified, the value of `:dropout_rate` is used instead"
],
layer_norm_epsilon: [
default: 1.0e-12,
doc: "the epsilon used by the layer normalization layers"
],
initializer_scale: [
default: 0.02,
doc:
"the standard deviation of the normal initializer used for initializing kernel parameters"
]
] ++
Shared.common_options([
:use_cross_attention,
:output_hidden_states,
:output_attentions,
:num_labels,
:id_to_label
]) ++
Shared.token_options(pad_token_id: 0) ++ Shared.generation_options()
@moduledoc """
DistilBERT model family.
## Architectures
* `:base` - plain DistilBERT without any head on top
* `:for_masked_language_modeling` - DistilBERT with a language modeling
head. The head returns logits for each token in the original
sequence
* `:for_sequence_classification` - DistilBERT with a sequence
classification head. The head returns logits corresponding to
possible classes
* `:for_token_classification` - DistilBERT with a token classification
head. The head returns logits for each token in the original
sequence
* `:for_question_answering` - DistilBERT with a span classification head.
The head returns logits for the span start and end positions
* `:for_multiple_choice` - DistilBERT with a multiple choice prediction
head. Each input in the batch consists of several sequences to
choose from and the model returns logits corresponding to those
choices
## Inputs
* `"input_ids"` - `{batch_size, sequence_length}`
Indices of input sequence tokens in the vocabulary.
* `"attention_mask"` - `{batch_size, sequence_length}`
Mask indicating which tokens to attend to. This is used to ignore
padding tokens, which are added when processing a batch of sequences
with different length.
Mask distinguishing groups in the input sequence. This is used
in when the input sequence is a semantically a pair of sequences.
* `"position_ids"` - `{batch_size, sequence_length}`
Indices of positions of each input sequence tokens in the position
embeddings.
* `"attention_head_mask"` - `{num_blocks, num_attention_heads}`
Mask to nullify selected heads of the self-attention blocks in
the encoder.
### Exceptions
The `:for_multiple_choice` model accepts groups of sequences, so the
expected sequence shape is `{batch_size, num_choices, sequence_length}`.
## Configuration
#{Shared.options_doc(options)}
## References
* [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108)
"""
defstruct [architecture: :base] ++ Shared.option_defaults(options)
@behaviour Bumblebee.ModelSpec
@behaviour Bumblebee.Configurable
import Bumblebee.Utils.Model, only: [join: 2]
alias Bumblebee.Layers
@impl true
def architectures(),
do: [
:base,
:for_masked_language_modeling,
:for_sequence_classification,
:for_token_classification,
:for_question_answering,
:for_multiple_choice
]
@impl true
def config(spec, opts \\ []) do
spec
|> Shared.put_config_attrs(opts)
|> Shared.validate_label_options()
end
@impl true
def input_template(%{architecture: :for_multiple_choice}) do
%{"input_ids" => Nx.template({1, 1, 1}, :s64)}
end
def input_template(_spec) do
%{"input_ids" => Nx.template({1, 1}, :s64)}
end
@impl true
def model(%__MODULE__{architecture: :base} = spec) do
inputs = inputs(spec)
inputs
|> core(spec)
|> Layers.output()
end
def model(%__MODULE__{architecture: :for_masked_language_modeling} = spec) do
inputs = inputs(spec)
outputs = core(inputs, spec)
logits = language_modeling_head(outputs.hidden_state, spec, name: "language_modeling_head")
Layers.output(%{
logits: logits,
hidden_states: outputs.hidden_states,
attentions: outputs.attentions
})
end
def model(%__MODULE__{architecture: :for_sequence_classification} = spec) do
inputs = inputs(spec)
outputs = core(inputs, spec)
logits =
outputs.pooled_state
|> Axon.dropout(
rate: classifier_dropout_rate(spec),
name: "sequence_classification_head.dropout"
)
|> Axon.dense(spec.num_labels,
kernel_initializer: kernel_initializer(spec),
name: "sequence_classification_head.output"
)
Layers.output(%{
logits: logits,
hidden_states: outputs.hidden_states,
attentions: outputs.attentions
})
end
def model(%__MODULE__{architecture: :for_token_classification} = spec) do
inputs = inputs(spec)
outputs = core(inputs, spec)
logits =
outputs.hidden_state
|> Axon.dropout(
rate: classifier_dropout_rate(spec),
name: "token_classification_head.dropout"
)
|> Axon.dense(spec.num_labels,
kernel_initializer: kernel_initializer(spec),
name: "token_classification_head.output"
)
Layers.output(%{
logits: logits,
hidden_states: outputs.hidden_states,
attentions: outputs.attentions
})
end
def model(%__MODULE__{architecture: :for_question_answering} = spec) do
inputs = inputs(spec)
outputs = core(inputs, spec)
logits =
outputs.hidden_state
|> Axon.dropout(
rate: classifier_dropout_rate(spec),
name: "question_answering_head.dropout"
)
|> Axon.dense(2,
kernel_initializer: kernel_initializer(spec),
name: "question_answering_head.output"
)
{start_logits, end_logits} = Layers.split_pair(logits)
Layers.output(%{
start_logits: start_logits,
end_logits: end_logits,
hidden_states: outputs.hidden_states,
attentions: outputs.attentions
})
end
def model(%__MODULE__{architecture: :for_multiple_choice} = spec) do
inputs = inputs(spec, shape: {nil, nil, nil})
group_inputs = ["input_ids", "attention_mask", "position_ids"]
flat_inputs =
Enum.reduce(group_inputs, inputs, fn name, inputs ->
Map.update!(inputs, name, &Layers.flatten_leading/1)
end)
outputs = core(flat_inputs, spec)
logits =
outputs.pooled_state
|> Axon.dropout(rate: classifier_dropout_rate(spec), name: "multiple_choice_head.dropout")
|> Axon.dense(1,
kernel_initializer: kernel_initializer(spec),
name: "multiple_choice_head.output"
)
# The final shape depends on the dynamic batch size and number
# of choices, so we do a reshape based on the input shape
logits =
Axon.layer(
fn logits, input_ids, _opts ->
num_choices = Nx.axis_size(input_ids, 1)
Nx.reshape(logits, {:auto, num_choices})
end,
[logits, inputs["input_ids"]]
)
Layers.output(%{
logits: logits,
hidden_states: outputs.hidden_states,
attentions: outputs.attentions
})
end
defp inputs(spec, opts \\ []) do
shape = Keyword.get(opts, :shape, {nil, nil})
attention_head_mask_shape = {spec.num_blocks, spec.num_attention_heads}
inputs =
Bumblebee.Utils.Model.inputs_to_map([
Axon.input("input_ids", shape: shape),
Axon.input("attention_mask", optional: true, shape: shape),
Axon.input("position_ids", optional: true, shape: shape),
Axon.input("attention_head_mask", optional: true, shape: attention_head_mask_shape)
])
inputs
end
defp core(inputs, spec) do
embeddings = embedder(inputs["input_ids"], inputs["position_ids"], spec, name: "embedder")
encoder_outputs =
encoder(
embeddings,
inputs["attention_mask"],
inputs["attention_head_mask"],
spec,
name: "encoder"
)
pooled_state = pooler(encoder_outputs.hidden_state, spec, name: "pooler")
%{
hidden_state: encoder_outputs.hidden_state,
pooled_state: pooled_state,
hidden_states: encoder_outputs.hidden_states,
attentions: encoder_outputs.attentions
}
end
defp embedder(input_ids, position_ids, spec, opts) do
name = opts[:name]
position_ids =
Layers.default position_ids do
Layers.default_position_ids(input_ids)
end
inputs_embeddings =
Axon.embedding(input_ids, spec.vocab_size, spec.hidden_size,
kernel_initializer: kernel_initializer(spec),
name: join(name, "token_embedding")
)
position_embeddings =
Axon.embedding(position_ids, spec.max_positions, spec.hidden_size,
kernel_initializer: kernel_initializer(spec),
name: join(name, "position_embedding")
)
Axon.add([inputs_embeddings, position_embeddings])
|> Axon.layer_norm(epsilon: spec.layer_norm_epsilon, name: join(name, "norm"))
|> Axon.dropout(rate: spec.dropout_rate, name: join(name, "dropout"))
end
defp encoder(
hidden_state,
attention_mask,
attention_head_mask,
spec,
opts
) do
name = opts[:name]
Layers.Transformer.blocks(
hidden_state,
attention_mask: attention_mask,
attention_head_mask: attention_head_mask,
num_blocks: spec.num_blocks,
num_attention_heads: spec.num_attention_heads,
hidden_size: spec.hidden_size,
kernel_initializer: kernel_initializer(spec),
dropout_rate: spec.dropout_rate,
attention_dropout_rate: spec.attention_dropout_rate,
layer_norm: [
epsilon: spec.layer_norm_epsilon
],
ffn: [
intermediate_size: spec.intermediate_size,
activation: spec.activation
],
output_hidden_states: spec.output_hidden_states,
output_attentions: spec.output_attentions,
name: join(name, "blocks")
)
end
defp pooler(hidden_state, spec, opts) do
name = opts[:name]
hidden_state
|> Layers.take_token(index: 0, axis: 1)
|> Axon.dense(spec.hidden_size,
kernel_initializer: kernel_initializer(spec),
name: join(name, "output")
)
|> Axon.relu()
end
defp language_modeling_head(hidden_state, spec, opts) do
name = opts[:name]
# TODO: use a shared parameter with embeddings.word_embeddings.kernel
# if spec.tie_word_embeddings is true (relevant for training)
hidden_state
|> Axon.dense(spec.hidden_size,
kernel_initializer: kernel_initializer(spec),
name: join(name, "dense")
)
|> Layers.activation(spec.activation, name: join(name, "activation"))
|> Axon.layer_norm(epsilon: spec.layer_norm_epsilon, name: join(name, "norm"))
# We reuse the kernel of input embeddings and add bias for each token
|> Layers.dense_transposed(spec.vocab_size,
kernel_initializer: kernel_initializer(spec),
name: join(name, "output")
)
|> Axon.bias(name: join(name, "bias"))
end
defp classifier_dropout_rate(spec) do
spec.classifier_dropout_rate || spec.dropout_rate
end
defp kernel_initializer(spec) do
Axon.Initializers.normal(scale: spec.initializer_scale)
end
defimpl Bumblebee.HuggingFace.Transformers.Config do
def load(spec, data) do
import Shared.Converters
opts =
convert!(data,
vocab_size: {"vocab_size", number()},
max_positions: {"max_position_embeddings", number()},
hidden_size: {"hidden_size", number()},
num_blocks: {"num_hidden_layers", number()},
num_attention_heads: {"num_attention_heads", number()},
intermediate_size: {"intermediate_size", number()},
activation: {"hidden_act", atom()},
dropout_rate: {"hidden_dropout_prob", number()},
attention_dropout_rate: {"attention_probs_dropout_prob", number()},
classifier_dropout_rate: {"classifier_dropout", optional(number())},
layer_norm_epsilon: {"layer_norm_eps", number()},
initializer_scale: {"initializer_range", number()}
) ++ Shared.common_options_from_transformers(data, spec)
@for.config(spec, opts)
end
end
defimpl Bumblebee.HuggingFace.Transformers.Model do
def params_mapping(_spec) do
%{
"embedder.token_embedding" => "distilbert.embeddings.word_embeddings",
"embedder.position_embedding" => "distilbert.embeddings.position_embeddings",
"embedder.norm" => "distilbert.embeddings.LayerNorm",
"encoder.blocks.{n}.self_attention.query" =>
"distilbert.transformer.layer.{n}.attention.q_lin",
"encoder.blocks.{n}.self_attention.key" =>
"distilbert.transformer.layer.{n}.attention.k_lin",
"encoder.blocks.{n}.self_attention.value" =>
"distilbert.transformer.layer.{n}.attention.v_lin",
"encoder.blocks.{n}.self_attention.output" =>
"distilbert.transformer.layer.{n}.attention.out_lin",
"encoder.blocks.{n}.self_attention_norm" =>
"distilbert.transformer.layer.{n}.sa_layer_norm",
"encoder.blocks.{n}.ffn.intermediate" => "distilbert.transformer.layer.{n}.ffn.lin1",
"encoder.blocks.{n}.ffn.output" => "distilbert.transformer.layer.{n}.ffn.lin2",
"encoder.blocks.{n}.output_norm" => "distilbert.transformer.layer.{n}.output_layer_norm",
"pooler.output" => "pre_classifier",
"language_modeling_head.dense" => "vocab_transform",
"language_modeling_head.norm" => "vocab_layer_norm",
"language_modeling_head.output" => "vocab_projector",
"language_modeling_head.bias" => "vocab_projector",
"next_sentence_prediction_head.output" => "cls.seq_relationship",
"sequence_classification_head.output" => "classifier",
"token_classification_head.output" => "classifier",
"multiple_choice_head.output" => "classifier",
"question_answering_head.output" => "qa_outputs"
}
end
end
end
