make tensorflow be optional for unittest (#394)

* make tensorflow be optional for unitest.

* typo

pyproject.toml

@@ -4,3 +4,6 @@ requires = ["setuptools", "wheel", "numpy>=1.18.5", "cmake"]  # PEP 508 specific
 
 [tool.black]
 line-length = 120
+
+[tool.ruff]
+line-length = 120

test/test_audio_codec.py

@@ -43,8 +43,9 @@ def test_mp3_decoder(self):
         pcm_tensor = self.decoder(np.expand_dims(np.asarray(blob), axis=(0,)))
         self.assertTrue(pcm_tensor.shape[1] > len(blob))
         # lossy compression, so we can only check the range
-        np.testing.assert_allclose(np.asarray([np.max(pcm_tensor), np.average(pcm_tensor), np.min(pcm_tensor)]),
-                                   np.asarray([np.max(self.raw_data), np.average(self.raw_data), np.min(self.raw_data)]), atol=1e-01)
+        np.testing.assert_allclose(
+            np.asarray([np.max(pcm_tensor), np.average(pcm_tensor), np.min(pcm_tensor)]),
+            np.asarray([np.max(self.raw_data), np.average(self.raw_data), np.min(self.raw_data)]), atol=1e-01)
 
 
 if __name__ == "__main__":

test/test_sentencepiece_ops.py

@@ -13,8 +13,15 @@
     OrtPyFunction,
     make_onnx_model,
     get_library_path as _get_library_path)
-import tensorflow as tf
-from tensorflow_text import SentencepieceTokenizer
+
+
+_is_tensorflow_avaliable = False
+try:
+    import tensorflow as tf
+    from tensorflow_text import SentencepieceTokenizer
+    _is_tensorflow_avaliable = True
+except ImportError:
+    pass
 
 
 def load_piece(name):
@@ -235,6 +242,7 @@ def _create_test_model_ragged_to_dense(
     return model
 
 
+@unittest.skipIf(not _is_tensorflow_avaliable, "tensorflow/tensorflow-text is unavailable")
 class TestPythonOpSentencePiece(unittest.TestCase):
 
     @classmethod
@@ -430,6 +438,8 @@ def test_string_sentencepiece_tokenizer_bin(self):
             assert_almost_equal(exp[i], py_txout[i])
             assert_almost_equal(exp[i], cc_txout[i])
 
+
+class TestOrtXSentencePiece(unittest.TestCase):
     def test_external_pretrained_model(self):
         fullname = util.get_test_data_file('data', 'en.wiki.bpe.vs100000.model')
         ofunc = OrtPyFunction.from_customop('SentencepieceTokenizer', model=open(fullname, 'rb').read())

tutorials/whisper_e2e.py

@@ -160,7 +160,8 @@ def postprocessing(token_ids, hf_processor):
     model_name = "openai/whisper-base.en"
     onnx_model_name = "whisper-base.en_beamsearch.onnx"
     if not Path(onnx_model_name).is_file():
-        raise RuntimeError("Please run the script from where Whisper ONNX model was exported. like */onnx_models/openai")
+        raise RuntimeError(
+            "Please run the script from where Whisper ONNX model was exported. like */onnx_models/openai")
 
     _processor = WhisperProcessor.from_pretrained(model_name)
     if USE_ONNX_COREMODEL: