def test_decode_infer(self):
r"""Tests decoding in inference mode."""
decoder = BasicRNNDecoder(input_size=self._emb_dim,
vocab_size=self._vocab_size,
hparams=self._hparams)
decoder.eval()
start_tokens = torch.tensor([self._vocab_size - 2] * self._batch_size)
helpers = []
for strategy in ['infer_greedy', 'infer_sample']:
helper = decoder.create_helper(
decoding_strategy=strategy,
embedding=self._embedding,
start_tokens=start_tokens,
end_token=self._vocab_size - 1)
helpers.append(helper)
for klass in ['TopKSampleEmbeddingHelper', 'SoftmaxEmbeddingHelper',
'GumbelSoftmaxEmbeddingHelper']:
helper = get_helper(
klass, embedding=self._embedding,
start_tokens=start_tokens, end_token=self._vocab_size - 1,
top_k=self._vocab_size // 2, tau=2.0,
straight_through=True)
helpers.append(helper)
for helper in helpers:
max_length = 100
outputs, final_state, sequence_lengths = decoder(
helper=helper, max_decoding_length=max_length)
self.assertLessEqual(max(sequence_lengths), max_length)
self._test_outputs(decoder, outputs, final_state, sequence_lengths,
test_mode=True, helper=helper)
def create_helper(self,
*,
decoding_strategy: Optional[str] = None,
embedding: Optional[Embedding] = None,
start_tokens: Optional[torch.LongTensor] = None,
end_token: Optional[int] = None,
softmax_temperature: Optional[float] = None,
infer_mode: Optional[bool] = None,
**kwargs) -> Helper:
r"""Create a helper instance for the decoder. This is a shared interface
for both :class:`~texar.modules.BasicRNNDecoder` and
:class:`~texar.modules.AttentionRNNDecoder`.
The function provides **3 ways** to specify the
decoding method, with varying flexibility:
1. The :attr:`decoding_strategy` argument: A string taking value of:
- **"train_greedy"**: decoding in teacher-forcing fashion (i.e.,
feeding `ground truth` to decode the next step), and each sample
is obtained by taking the `argmax` of the output logits.
Arguments :attr:`(inputs, sequence_length)`
are required for this strategy, and argument :attr:`embedding`
is optional.
- **"infer_greedy"**: decoding in inference fashion (i.e., feeding
the `generated` sample to decode the next step), and each sample
is obtained by taking the `argmax` of the output logits.
Arguments :attr:`(embedding, start_tokens, end_token)` are
required for this strategy, and argument
:attr:`max_decoding_length` is optional.
- **"infer_sample"**: decoding in inference fashion, and each
sample is obtained by `random sampling` from the RNN output
distribution. Arguments
:attr:`(embedding, start_tokens, end_token)` are
required for this strategy, and argument
:attr:`max_decoding_length` is optional.
This argument is used only when argument :attr:`helper` is `None`.
Example:
.. code-block:: python
embedder = WordEmbedder(vocab_size=data.vocab.size)
decoder = BasicRNNDecoder(vocab_size=data.vocab.size)
# Teacher-forcing decoding
outputs_1, _, _ = decoder(
decoding_strategy='train_greedy',
inputs=embedder(data_batch['text_ids']),
sequence_length=data_batch['length'] - 1)
# Random sample decoding. Gets 100 sequence samples
outputs_2, _, sequence_length = decoder(
decoding_strategy='infer_sample',
start_tokens=[data.vocab.bos_token_id] * 100,
end_token=data.vocab.eos.token_id,
embedding=embedder,
max_decoding_length=60)
2. The :attr:`helper` argument: An instance of subclass of
:class:`~texar.modules.decoders.rnn_decoder_helpers.Helper`. This
provides a superset of decoding strategies than above, for example:
- :class:`~texar.modules.TrainingHelper` corresponding to the
"train_greedy" strategy.
- :class:`~texar.modules.ScheduledEmbeddingTrainingHelper` and
:class:`~texar.modules.ScheduledOutputTrainingHelper` for
scheduled sampling.
- :class:`~texar.modules.SoftmaxEmbeddingHelper` and
:class:`~texar.modules.GumbelSoftmaxEmbeddingHelper` for
soft decoding and gradient backpropagation.
This means gives the maximal flexibility of configuring the decoding
strategy.
Example:
.. code-block:: python
embedder = WordEmbedder(vocab_size=data.vocab.size)
decoder = BasicRNNDecoder(vocab_size=data.vocab.size)
# Teacher-forcing decoding, same as above with
# `decoding_strategy='train_greedy'`
helper_1 = TrainingHelper(
inputs=embedders(data_batch['text_ids']),
sequence_length=data_batch['length'] - 1)
outputs_1, _, _ = decoder(helper=helper_1)
# Gumbel-softmax decoding
helper_2 = GumbelSoftmaxEmbeddingHelper(
embedding=embedder,
start_tokens=[data.vocab.bos_token_id] * 100,
end_token=data.vocab.eos_token_id,
tau=0.1)
outputs_2, _, sequence_length = decoder(
max_decoding_length=60, helper=helper_2)
3. ``hparams["helper_train"]`` and ``hparams["helper_infer"]``:
Specifying the helper through hyperparameters. Train and infer
strategy is toggled based on :attr:`mode`. Appropriate arguments
(e.g., :attr:`inputs`, :attr:`start_tokens`, etc) are selected to
construct the helper. Additional arguments for helper constructor
can be provided either through :attr:`**kwargs`, or through
``hparams["helper_train/infer"]["kwargs"]``.
This means is used only when both :attr:`decoding_strategy` and
:attr:`helper` are ``None``.
Example:
.. code-block:: python
h = {
"helper_infer": {
"type": "GumbelSoftmaxEmbeddingHelper",
"kwargs": { "tau": 0.1 }
}
}
embedder = WordEmbedder(vocab_size=data.vocab.size)
decoder = BasicRNNDecoder(vocab_size=data.vocab.size, hparams=h)
# Gumbel-softmax decoding
decoder.eval() # disable dropout
output, _, _ = decoder(
decoding_strategy=None, # Sets to None explicit
embedding=embedder,
start_tokens=[data.vocab.bos_token_id] * 100,
end_token=data.vocab.eos_token_id,
max_decoding_length=60)
Args:
decoding_strategy (str): A string specifying the decoding
strategy. Different arguments are required based on the
strategy.
Ignored if :attr:`helper` is given.
embedding (optional): A callable that returns embedding vectors
of `inputs` (e.g., an instance of subclass of
:class:`~texar.modules.EmbedderBase`), or the `params`
argument of
:tf_main:`tf.nn.embedding_lookup <nn/embedding_lookup>`.
If provided, `inputs` (if used) will be passed to
`embedding` to fetch the embedding vectors of the inputs.
Required when :attr:`decoding_strategy` is ``"infer_greedy"``
or ``"infer_sample"``; optional when
``decoding_strategy="train_greedy"``.
start_tokens (optional): A :tensor:`LongTensor` of shape
``[batch_size]``, the start tokens.
Used when :attr:`decoding_strategy` is ``"infer_greedy"`` or
``"infer_sample"``, or when `hparams`-configured
helper is used.
When used with the Texar data module, to get ``batch_size``
samples where ``batch_size`` is changing according to the data
module, this can be set as
``start_tokens=torch.full_like(batch['length'], bos_token_id)``.
end_token (optional): A integer or 0D :tensor:`LongTensor`, the
token that marks the end of decoding.
Used when :attr:`decoding_strategy` is ``"infer_greedy"`` or
``"infer_sample"``, or when `hparams`-configured helper is used.
softmax_temperature (float, optional): Value to divide the logits
by before computing the softmax. Larger values (above 1.0)
result in more random samples. Must be > 0. If ``None``, 1.0 is
used. Used when ``decoding_strategy="infer_sample"``.
infer_mode (optional): If not ``None``, overrides mode given by
:attr:`self.training`.
**kwargs: Other keyword arguments for constructing helpers
defined by ``hparams["helper_train"]`` or
``hparams["helper_infer"]``.
Returns:
The constructed helper instance.
"""
if decoding_strategy is not None:
if decoding_strategy == 'train_greedy':
helper: Helper = decoder_helpers.TrainingHelper(
embedding, self._input_time_major)
elif decoding_strategy in ['infer_greedy', 'infer_sample']:
if (embedding is None or start_tokens is None
or end_token is None):
raise ValueError(
f"When using '{decoding_strategy}' decoding strategy, "
f"'embedding', 'start_tokens', and 'end_token' must "
f"not be `None`.")
if decoding_strategy == 'infer_greedy':
helper = decoder_helpers.GreedyEmbeddingHelper(
embedding, start_tokens, end_token)
else:
helper = decoder_helpers.SampleEmbeddingHelper(
embedding, start_tokens, end_token,
softmax_temperature)
else:
raise ValueError(
f"Unknown decoding strategy: {decoding_strategy}")
else:
is_training = (not infer_mode
if infer_mode is not None else self.training)
if is_training:
kwargs_ = copy.copy(self._hparams.helper_train.kwargs.todict())
helper_type = self._hparams.helper_train.type
else:
kwargs_ = copy.copy(self._hparams.helper_infer.kwargs.todict())
helper_type = self._hparams.helper_infer.type
kwargs_.update({
'time_major': self._input_time_major,
'embedding': embedding,
'start_tokens': start_tokens,
'end_token': end_token,
'softmax_temperature': softmax_temperature
})
kwargs_.update(kwargs)
helper = decoder_helpers.get_helper(helper_type, **kwargs_)
return helper