def test_forward(self):
with initialize_config_dir(config_dir=get_config_directory()):
data_folder, dataset_name = get_test_data_info()
config = compose("main",
overrides=[
f"data_folder={data_folder}",
f"dataset.name={dataset_name}"
])
dataset_folder = join(config.data_folder, config.dataset.name)
vocabulary = Vocabulary.load_vocabulary(
join(dataset_folder, config.vocabulary_name))
data_file_path = join(
dataset_folder,
f"{config.dataset.name}.{config.train_holdout}.c2s")
dataset = PathContextDataset(data_file_path, config, vocabulary, False)
batch = PathContextBatch(
[dataset[i] for i in range(config.hyper_parameters.batch_size)])
model = PathEncoder(
config.encoder,
config.decoder.decoder_size,
len(vocabulary.token_to_id),
vocabulary.token_to_id[PAD],
len(vocabulary.node_to_id),
vocabulary.node_to_id[PAD],
)
output = model(batch.contexts)
true_shape = (sum(batch.contexts_per_label),
config.decoder.decoder_size)
self.assertTupleEqual(true_shape, output.shape)
def test_forward(self):
config = EncoderConfig(self._hidden_size, self._hidden_size, True, 0.5, 1, 0.5)
buffered_path_contexts = BufferedPathContext.load(self._test_data_path)
batch = PathContextBatch([buffered_path_contexts[i] for i in range(self._batch_size)])
token_vocab_size = max(batch.context[FROM_TOKEN].max().item(), batch.context[TO_TOKEN].max().item())
type_vocab_size = batch.context[PATH_TYPES].max().item()
model = PathEncoder(config, self._hidden_size, token_vocab_size + 1, 0, type_vocab_size + 1, 0)
out = model(batch.context)
number_of_paths = sum(batch.contexts_per_label)
self.assertTupleEqual((number_of_paths, self._hidden_size), out.shape)
def __init__(self, config: DictConfig, vocabulary: Vocabulary):
super().__init__()
self._config = config
self.save_hyperparameters()
self.encoder = PathEncoder(
self._config.encoder,
self._config.classifier.classifier_input_size,
len(vocabulary.token_to_id),
vocabulary.token_to_id[PAD],
len(vocabulary.node_to_id),
vocabulary.node_to_id[PAD],
)
self.num_classes = len(vocabulary.label_to_id)
self.classifier = PathClassifier(self._config.classifier, self.num_classes)
def __init__(self, config: Code2ClassConfig, vocab: Vocabulary,
num_workers: int):
super().__init__(config.hyperparams, vocab, num_workers)
self.save_hyperparameters()
self.encoder = PathEncoder(
config.encoder_config,
config.classifier_config.classifier_input_size,
len(self.vocab.token_to_id),
self.vocab.token_to_id[PAD],
len(self.vocab.type_to_id),
self.vocab.type_to_id[PAD],
)
self.num_classes = len(self.vocab.label_to_id)
self.classifier = PathClassifier(config.classifier_config,
self.num_classes)
def __init__(self, config: Code2SeqConfig, vocab: Vocabulary, num_workers: int):
super().__init__(config.hyperparams, vocab, num_workers)
self.save_hyperparameters()
encoder_config = config.encoder_config
decoder_config = config.decoder_config
self.encoder = PathEncoder(
encoder_config,
decoder_config.decoder_size,
len(vocab.token_to_id),
vocab.token_to_id[PAD],
len(vocab.type_to_id),
vocab.type_to_id[PAD],
)
self.decoder = PathDecoder(
decoder_config, len(vocab.label_to_id), vocab.label_to_id[SOS], vocab.label_to_id[PAD]
)