SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

added biogpt token classifier (#22447)

Browse Source 
* added biogpt token classifier

* fix reviews

* Updated modeling_biogpt.py

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
This commit is contained in:
Mohammed Jabir
2023-04-03 18:50:02 +05:30
committed by GitHub
parent 1194c3e315
commit 7d25c9c81e
8 changed files with 138 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
tests/models/biogpt/test_modeling_biogpt.py
View File

@@ -29,7 +29,7 @@ from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import BioGptForCausalLM, BioGptModel, BioGptTokenizer
from transformers import BioGptForCausalLM, BioGptForTokenClassification, BioGptModel, BioGptTokenizer
from transformers.models.biogpt.modeling_biogpt import BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST
@@ -247,6 +247,16 @@ class BioGptModelTester:
self.parent.assertLessEqual(abs(torch.std(model.state_dict()[key]) - model_std), 0.001)
self.parent.assertLessEqual(abs(torch.mean(model.state_dict()[key]) - 0.0), 0.01)
def create_and_check_biogpt_for_token_classification(
self, config, input_ids, input_mask, head_mask, token_type_ids, *args
):
config.num_labels = self.num_labels
model = BioGptForTokenClassification(config)
model.to(torch_device)
model.eval()
result = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
self.parent.assertEqual(result.logits.shape, (self.batch_size, self.seq_length, self.num_labels))
def prepare_config_and_inputs_for_common(self):
config_and_inputs = self.prepare_config_and_inputs()
(
@@ -264,10 +274,16 @@ class BioGptModelTester:
@require_torch
class BioGptModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixin, unittest.TestCase):
all_model_classes = (BioGptModel, BioGptForCausalLM) if is_torch_available() else ()
all_model_classes = (BioGptModel, BioGptForCausalLM, BioGptForTokenClassification) if is_torch_available() else ()
all_generative_model_classes = (BioGptForCausalLM,) if is_torch_available() else ()
pipeline_model_mapping = (
{"feature-extraction": BioGptModel, "text-generation": BioGptForCausalLM} if is_torch_available() else {}
{
"feature-extraction": BioGptModel,
"text-generation": BioGptForCausalLM,
"token-classification": BioGptForTokenClassification,
}
if is_torch_available()
else {}
)
test_pruning = False
@@ -304,6 +320,10 @@ class BioGptModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMix
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_biogpt_weight_initialization(*config_and_inputs)
def test_biogpt_token_classification_model(self):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_biogpt_for_token_classification(*config_and_inputs)
@slow
def test_batch_generation(self):
model = BioGptForCausalLM.from_pretrained("microsoft/biogpt")