def main():
reader = TatoebaSentenceReader()
train_set = reader.read('data/mt/sentences.top10langs.train.tsv')
dev_set = reader.read('data/mt/sentences.top10langs.dev.tsv')
vocab = Vocabulary.from_instances(train_set,
min_count={'tokens': 3})
token_embedding = Embedding(num_embeddings=vocab.get_vocab_size('tokens'),
embedding_dim=EMBEDDING_DIM)
word_embeddings = BasicTextFieldEmbedder({"tokens": token_embedding})
encoder = PytorchSeq2VecWrapper(
torch.nn.LSTM(EMBEDDING_DIM, HIDDEN_DIM, batch_first=True))
positive_label = vocab.get_token_index('eng', namespace='labels')
model = LstmClassifier(word_embeddings, encoder, vocab, positive_label=positive_label)
optimizer = optim.Adam(model.parameters())
iterator = BucketIterator(batch_size=32, sorting_keys=[("tokens", "num_tokens")])
iterator.index_with(vocab)
trainer = Trainer(model=model,
optimizer=optimizer,
iterator=iterator,
train_dataset=train_set,
validation_dataset=dev_set,
num_epochs=3)
trainer.train()
def main():
reader = TatoebaSentenceReader()
train_set = reader.read('data/tatoeba/sentences.top10langs.train.tsv')
dev_set = reader.read('data/tatoeba/sentences.top10langs.dev.tsv')
vocab = Vocabulary.from_instances(train_set, min_count={'tokens': 3})
token_embedding = Embedding(num_embeddings=vocab.get_vocab_size('tokens'),
embedding_dim=EMBEDDING_DIM)
word_embeddings = BasicTextFieldEmbedder({"tokens": token_embedding})
encoder = PytorchSeq2VecWrapper(
torch.nn.LSTM(EMBEDDING_DIM, HIDDEN_DIM, batch_first=True))
positive_label = vocab.get_token_index('eng', namespace='labels')
model = LstmClassifier(word_embeddings,
encoder,
vocab,
positive_label=positive_label)
optimizer = optim.Adam(model.parameters())
iterator = BucketIterator(batch_size=32,
sorting_keys=[("tokens", "num_tokens")])
iterator.index_with(vocab)
trainer = Trainer(model=model,
optimizer=optimizer,
iterator=iterator,
train_dataset=train_set,
validation_dataset=dev_set,
num_epochs=10)
trainer.train()
classify('Take your raincoat in case it rains.', model)
classify('Tu me recuerdas a mi padre.', model)
classify('Wie organisierst du das Essen am Mittag?', model)
classify("Il est des cas où cette règle ne s'applique pas.", model)
classify('Estou fazendo um passeio em um parque.', model)
classify('Ve, postmorgaŭ jam estas la limdato.', model)
classify('Credevo che sarebbe venuto.', model)
classify('Nem tudja, hogy én egy macska vagyok.', model)
classify('Nella ur nli qrib acemma deg tenwalt.', model)
classify('Kurşun kalemin yok, değil mi?', model)
def classify(text: str, model: LstmClassifier):
tokenizer = CharacterTokenizer()
token_indexers = {'tokens': SingleIdTokenIndexer()}
tokens = tokenizer.tokenize(text)
instance = Instance({'tokens': TextField(tokens, token_indexers)})
logits = model.forward_on_instances([instance])[0]['logits']
label_id = np.argmax(logits)
label = model.vocab.get_token_from_index(label_id, 'labels')
print('text: {}, label: {}'.format(text, label))
def main():
# In order to use ELMo, each word in a sentence needs to be indexed with
# an array of character IDs.
elmo_token_indexer = ELMoTokenCharactersIndexer()
reader = StanfordSentimentTreeBankDatasetReader(
token_indexers={'tokens': elmo_token_indexer})
train_dataset = reader.read('data/stanfordSentimentTreebank/trees/train.txt')
dev_dataset = reader.read('data/stanfordSentimentTreebank/trees/dev.txt')
# Initialize the ELMo-based token embedder using a pre-trained file.
# This takes a while if you run this script for the first time
# Original
# options_file = "https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x4096_512_2048cnn_2xhighway/elmo_2x4096_512_2048cnn_2xhighway_options.json"
# weight_file = "https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x4096_512_2048cnn_2xhighway/elmo_2x4096_512_2048cnn_2xhighway_weights.hdf5"
# Medium
# options_file = "https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x2048_256_2048cnn_1xhighway/elmo_2x2048_256_2048cnn_1xhighway_options.json"
# weight_file = "https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x2048_256_2048cnn_1xhighway/elmo_2x2048_256_2048cnn_1xhighway_weights.hdf5"
# Use the 'Small' pre-trained model
options_file = ('https://s3-us-west-2.amazonaws.com/allennlp/models/elmo'
'/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_options.json')
weight_file = ('https://s3-us-west-2.amazonaws.com/allennlp/models/elmo'
'/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5')
elmo_embedder = ElmoTokenEmbedder(options_file, weight_file)
vocab = Vocabulary.from_instances(train_dataset + dev_dataset,
min_count={'tokens': 3})
# Pass in the ElmoTokenEmbedder instance instead
word_embeddings = BasicTextFieldEmbedder({"tokens": elmo_embedder})
# The dimension of the ELMo embedding will be 2 x [size of LSTM hidden states]
elmo_embedding_dim = 256
lstm = PytorchSeq2VecWrapper(
torch.nn.LSTM(elmo_embedding_dim, HIDDEN_DIM, batch_first=True))
model = LstmClassifier(word_embeddings, lstm, vocab)
optimizer = optim.Adam(model.parameters())
iterator = BucketIterator(batch_size=32, sorting_keys=[("tokens", "num_tokens")])
iterator.index_with(vocab)
trainer = Trainer(model=model,
optimizer=optimizer,
iterator=iterator,
train_dataset=train_dataset,
validation_dataset=dev_dataset,
patience=10,
num_epochs=20)
trainer.train()
tokens = ['This', 'is', 'the', 'best', 'movie', 'ever', '!']
predictor = SentenceClassifierPredictor(model, dataset_reader=reader)
logits = predictor.predict(tokens)['logits']
label_id = np.argmax(logits)
print(model.vocab.get_token_from_index(label_id, 'labels'))
# '/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5')
elmo_embedder = ElmoTokenEmbedder(options_file, weight_file)
vocab = Vocabulary.from_instances(train_dataset + dev_dataset,
min_count={'tokens': 3})
# Pass in the ElmoTokenEmbedder instance instead
word_embeddings = BasicTextFieldEmbedder({"tokens": elmo_embedder})
# The dimension of the ELMo embedding will be 2 x [size of LSTM hidden states]
elmo_embedding_dim = 1024
lstm = PytorchSeq2VecWrapper(
torch.nn.LSTM(elmo_embedding_dim, HIDDEN_DIM, batch_first=True))
model = LstmClassifier(word_embeddings, lstm, vocab)
optimizer = optim.AdamW(model.parameters())
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)
iterator = BucketIterator(batch_size=32,
sorting_keys=[("tokens", "num_tokens")])
iterator.index_with(vocab)
trainer = Trainer(model=model,
optimizer=optimizer,
iterator=iterator,
train_dataset=train_dataset,
validation_dataset=dev_dataset,