def create_and_check_transfo_xl_model(self, config, input_ids_1,
input_ids_2, lm_labels):
model = TFTransfoXLModel(config)
hidden_states_1, mems_1 = model(input_ids_1).to_tuple()
inputs = {"input_ids": input_ids_2, "mems": mems_1}
hidden_states_2, mems_2 = model(inputs).to_tuple()
self.parent.assertEqual(
hidden_states_1.shape,
(self.batch_size, self.seq_length, self.hidden_size))
self.parent.assertEqual(
hidden_states_2.shape,
(self.batch_size, self.seq_length, self.hidden_size))
self.parent.assertListEqual(
[mem.shape for mem in mems_1],
[(self.mem_len, self.batch_size, self.hidden_size)] *
self.num_hidden_layers,
)
self.parent.assertListEqual(
[mem.shape for mem in mems_2],
[(self.mem_len, self.batch_size, self.hidden_size)] *
self.num_hidden_layers,
)
def create_and_check_transfo_xl_model(self, config, input_ids_1,
input_ids_2, lm_labels):
model = TFTransfoXLModel(config)
hidden_states_1, mems_1 = model(input_ids_1)
inputs = {"input_ids": input_ids_2, "mems": mems_1}
hidden_states_2, mems_2 = model(inputs)
result = {
"hidden_states_1": hidden_states_1.numpy(),
"mems_1": [mem.numpy() for mem in mems_1],
"hidden_states_2": hidden_states_2.numpy(),
"mems_2": [mem.numpy() for mem in mems_2],
}
self.parent.assertListEqual(
list(result["hidden_states_1"].shape),
[self.batch_size, self.seq_length, self.hidden_size])
self.parent.assertListEqual(
list(result["hidden_states_2"].shape),
[self.batch_size, self.seq_length, self.hidden_size])
self.parent.assertListEqual(
list(list(mem.shape) for mem in result["mems_1"]),
[[self.mem_len, self.batch_size, self.hidden_size]] *
self.num_hidden_layers,
)
self.parent.assertListEqual(
list(list(mem.shape) for mem in result["mems_2"]),
[[self.mem_len, self.batch_size, self.hidden_size]] *
self.num_hidden_layers,
)
def get_transformer(bert_model_type, output_hidden_states=False):
config = get_bert_config(bert_model_type, output_hidden_states)
if bert_model_type in [
'bert-base-uncased', 'bert-base-cased', 'bert-large-uncased',
'bert-large-uncased-whole-word-masking',
'bert-large-uncased-whole-word-masking-finetuned-squad'
]:
return TFBertModel.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config)
elif bert_model_type in [
'prod-bert-base-uncased', 'tune_bert-base-uncased_nsp'
]:
return TFBertModel.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config,
from_pt=True)
elif bert_model_type in [
'roberta-base', 'roberta-large', 'roberta-large-mnli',
'distilroberta-base'
]:
return TFRobertaModel.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config)
elif bert_model_type in ['prod-roberta-base-cased']:
return TFRobertaModel.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config,
from_pt=True)
elif bert_model_type in ['xlnet-base-cased']:
return TFXLNetModel.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config)
elif bert_model_type in [
'albert-base-v1', 'albert-large-v1', 'albert-xlarge-v1',
'albert-xxlarge-v1'
]:
return TFAlbertModel.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config)
elif bert_model_type in ['gpt2', 'gpt2-medium']:
return TFGPT2Model.from_pretrained(BERT_MODEL_FILE[bert_model_type],
config=config)
elif bert_model_type in ['transfo-xl']:
return TFTransfoXLModel.from_pretrained(
BERT_MODEL_FILE[bert_model_type], config=config)
elif bert_model_type in [
'distilbert-base-uncased',
'distilbert-base-uncased-distilled-squad'
]:
return TFDistilBertModel.from_pretrained(
BERT_MODEL_FILE[bert_model_type], config=config)
else:
raise ValueError(
f'`bert_model_type` not understood: {bert_model_type}')
def __init__(self,
pretrained_model_name_or_path='transfo-xl-wt103',
reduce_output='sum',
trainable=True,
**kwargs):
super().__init__()
try:
from transformers import TFTransfoXLModel
except ModuleNotFoundError:
logger.error(
' transformers is not installed. '
'In order to install all text feature dependencies run '
'pip install ludwig[text]')
sys.exit(-1)
self.transformer = TFTransfoXLModel.from_pretrained(
pretrained_model_name_or_path)
self.reduce_output = reduce_output
self.reduce_sequence = SequenceReducer(reduce_mode=reduce_output)
self.transformer.trainable = trainable
def test_model_from_pretrained(self):
for model_name in TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
model = TFTransfoXLModel.from_pretrained(model_name)
self.assertIsNotNone(model)
def test_model_from_pretrained(self):
for model_name in list(
TF_TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = TFTransfoXLModel.from_pretrained(model_name,
cache_dir=CACHE_DIR)
self.assertIsNotNone(model)
