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

Generate: smaller TF serving test (#18840)

Browse Source
This commit is contained in:
Joao Gante
2022-09-01 10:53:39 +01:00
committed by GitHub
parent 563a8d58db
commit 6e016634f1
1 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
tests/test_modeling_tf_common.py
View File

@@ -75,7 +75,7 @@ if is_tf_available():
TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING, TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING,
BertConfig, BertConfig,
TFAutoModel, TFAutoModel,
TFAutoModelForSeq2SeqLM, TFAutoModelForCausalLM,
TFAutoModelForSequenceClassification, TFAutoModelForSequenceClassification,
TFBertModel, TFBertModel,
TFSharedEmbeddings, TFSharedEmbeddings,
@@ -2180,8 +2180,8 @@ class UtilsFunctionsTest(unittest.TestCase):
self.assertTrue(np.allclose(p1.numpy(), p2.numpy())) self.assertTrue(np.allclose(p1.numpy(), p2.numpy()))
def test_generate_tf_function_export(self): def test_generate_tf_function_export(self):
test_model = TFAutoModelForSeq2SeqLM.from_pretrained("hf-internal-testing/tiny-random-t5") test_model = TFAutoModelForCausalLM.from_pretrained("hf-internal-testing/tiny-random-gpt2")
max_length = 8 max_length = 2
class DummyModel(tf.Module): class DummyModel(tf.Module):
def __init__(self, model): def __init__(self, model):
@@ -2204,8 +2204,8 @@ class UtilsFunctionsTest(unittest.TestCase):
) )
return {"sequences": outputs["sequences"]} return {"sequences": outputs["sequences"]}
dummy_input_ids = [[2, 3, 4, 1, 0, 0, 0, 0], [102, 103, 104, 105, 1, 0, 0, 0]] dummy_input_ids = [[2, 0], [102, 103]]
dummy_attention_masks = [[1, 1, 1, 1, 0, 0, 0, 0], [1, 1, 1, 1, 1, 0, 0, 0]] dummy_attention_masks = [[1, 0], [1, 1]]
dummy_model = DummyModel(model=test_model) dummy_model = DummyModel(model=test_model)
with tempfile.TemporaryDirectory() as tmp_dir: with tempfile.TemporaryDirectory() as tmp_dir:
tf.saved_model.save(dummy_model, tmp_dir, signatures={"serving_default": dummy_model.serving}) tf.saved_model.save(dummy_model, tmp_dir, signatures={"serving_default": dummy_model.serving})