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

Add BioGPTForSequenceClassification (#22253)

Browse Source 
* added BioGptForSequenceClassification

* added source of copied code

* typo

* Format code with black

* Update comments for copied code

* Remove code copy comment

* Apply suggestions from code review

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Fix failing tests

* Update code copied from comments

* Fix code quality

* Update src/transformers/models/biogpt/modeling_biogpt.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Apply suggestions from code review

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Fix lint error

* Update src/transformers/models/biogpt/modeling_biogpt.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Rename model to biogpt for consistency

* Add PipelineTesterMixin to test_modeling_biogpt.py

* Apply suggestions from code review

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* Resolve merge confict

---------

Co-authored-by: Guillem García Subies <37592763+GuillemGSubies@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
This commit is contained in:
Ashwin Mathur
2023-05-01 18:47:27 +05:30
committed by GitHub
parent 549e5f9f23
commit 487f132a6f
8 changed files with 198 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
tests/models/biogpt/test_modeling_biogpt.py
View File

@@ -29,7 +29,13 @@ from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import BioGptForCausalLM, BioGptForTokenClassification, BioGptModel, BioGptTokenizer
from transformers import (
BioGptForCausalLM,
BioGptForSequenceClassification,
BioGptForTokenClassification,
BioGptModel,
BioGptTokenizer,
)
from transformers.models.biogpt.modeling_biogpt import BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST
@@ -274,13 +280,18 @@ class BioGptModelTester:
@require_torch
class BioGptModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixin, unittest.TestCase):
all_model_classes = (BioGptModel, BioGptForCausalLM, BioGptForTokenClassification) if is_torch_available() else ()
all_model_classes = (
(BioGptModel, BioGptForCausalLM, BioGptForSequenceClassification, 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,
"token-classification": BioGptForTokenClassification,
"text-classification": BioGptForSequenceClassification,
}
if is_torch_available()
else {}
@@ -374,6 +385,35 @@ class BioGptModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMix
model = BioGptModel.from_pretrained(model_name)
self.assertIsNotNone(model)
# Copied from tests.models.opt.test_modeling_opt.OPTModelTest with OPT->BioGpt, prepare_config_and_inputs-> prepare_config_and_inputs_for_common
def test_biogpt_sequence_classification_model(self):
config, input_dict = self.model_tester.prepare_config_and_inputs_for_common()
config.num_labels = 3
input_ids = input_dict["input_ids"]
attention_mask = input_ids.ne(1).to(torch_device)
sequence_labels = ids_tensor([self.model_tester.batch_size], self.model_tester.type_sequence_label_size)
model = BioGptForSequenceClassification(config)
model.to(torch_device)
model.eval()
result = model(input_ids, attention_mask=attention_mask, labels=sequence_labels)
self.assertEqual(result.logits.shape, (self.model_tester.batch_size, self.model_tester.num_labels))
# Copied from tests.models.opt.test_modeling_opt.OPTModelTest with OPT->BioGpt, prepare_config_and_inputs-> prepare_config_and_inputs_for_common
def test_biogpt_sequence_classification_model_for_multi_label(self):
config, input_dict = self.model_tester.prepare_config_and_inputs_for_common()
config.num_labels = 3
config.problem_type = "multi_label_classification"
input_ids = input_dict["input_ids"]
attention_mask = input_ids.ne(1).to(torch_device)
sequence_labels = ids_tensor(
[self.model_tester.batch_size, config.num_labels], self.model_tester.type_sequence_label_size
).to(torch.float)
model = BioGptForSequenceClassification(config)
model.to(torch_device)
model.eval()
result = model(input_ids, attention_mask=attention_mask, labels=sequence_labels)
self.assertEqual(result.logits.shape, (self.model_tester.batch_size, self.model_tester.num_labels))
@require_torch
class BioGptModelIntegrationTest(unittest.TestCase):