def create_and_check_model_as_decoder(
self,
config,
input_ids,
token_type_ids,
input_mask,
sequence_labels,
token_labels,
choice_labels,
encoder_hidden_states,
encoder_attention_mask,
):
config.add_cross_attention = True
model = YosoModel(config)
model.to(torch_device)
model.eval()
result = model(
input_ids,
attention_mask=input_mask,
token_type_ids=token_type_ids,
encoder_hidden_states=encoder_hidden_states,
encoder_attention_mask=encoder_attention_mask,
)
result = model(
input_ids,
attention_mask=input_mask,
token_type_ids=token_type_ids,
encoder_hidden_states=encoder_hidden_states,
)
result = model(input_ids,
attention_mask=input_mask,
token_type_ids=token_type_ids)
self.parent.assertEqual(
result.last_hidden_state.shape,
(self.batch_size, self.seq_length, self.hidden_size))
def test_inference_no_head(self):
model = YosoModel.from_pretrained("uw-madison/yoso-4096")
input_ids = torch.tensor([[0, 1, 2, 3, 4, 5]])
with torch.no_grad():
output = model(input_ids)[0]
expected_shape = torch.Size((1, 6, 768))
self.assertEqual(output.shape, expected_shape)
expected_slice = torch.tensor([[[-0.0611, 0.1242, 0.0840],
[0.0280, -0.0048, 0.1125],
[0.0106, 0.0226, 0.0751]]])
self.assertTrue(
torch.allclose(output[:, :3, :3], expected_slice, atol=1e-4))
def create_and_check_model(self, config, input_ids, token_type_ids,
input_mask, sequence_labels, token_labels,
choice_labels):
model = YosoModel(config=config)
model.to(torch_device)
model.eval()
result = model(input_ids,
attention_mask=input_mask,
token_type_ids=token_type_ids)
result = model(input_ids, token_type_ids=token_type_ids)
result = model(input_ids)
self.parent.assertEqual(
result.last_hidden_state.shape,
(self.batch_size, self.seq_length, self.hidden_size))
def test_model_from_pretrained(self):
for model_name in YOSO_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
model = YosoModel.from_pretrained(model_name)
self.assertIsNotNone(model)