You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/12/20 21:00:03 UTC

[GitHub] [incubator-mxnet] marcoabreu commented on a change in pull request #19697: [v1.x] Add more CV models to onnxruntime inference test, add bert model test.

marcoabreu commented on a change in pull request #19697:
URL: https://github.com/apache/incubator-mxnet/pull/19697#discussion_r546435689



##########
File path: tests/python-pytest/onnx/test_onnxruntime.py
##########
@@ -118,4 +121,52 @@ def download_test_images(tmpdir):
     shutil.rmtree(tmp_path)
 
 
+@pytest.mark.parametrize('model', ['bert_12_768_12'])
+def test_bert_inference_onnxruntime(tmp_path, model):
+    import gluonnlp as nlp
+    dataset = 'book_corpus_wiki_en_uncased'
+    ctx = mx.cpu(0)
+    model, vocab = nlp.model.get_model(
+        name=model,
+        ctx=ctx,
+        dataset_name=dataset,
+        pretrained=False,
+        use_pooler=True,
+        use_decoder=False,
+        use_classifier=False)
+    model.initialize(ctx=ctx)
+    model.hybridize(static_alloc=True)
+
+    batch = 5
+    seq_length = 16
+    # create synthetic test data
+    inputs = mx.nd.random.uniform(0, 30522, shape=(batch, seq_length), dtype='float32')
+    token_types = mx.nd.random.uniform(0, 2, shape=(batch, seq_length), dtype='float32')
+    valid_length = mx.nd.array([seq_length] * batch, dtype='float32')
+
+    seq_encoding, cls_encoding = model(inputs, token_types, valid_length)
+
+    prefix = "%s/bert" % tmp_path
+    model.export(prefix)
+    sym_file = "%s-symbol.json" % prefix
+    params_file = "%s-0000.params" % prefix
+    onnx_file = "%s.onnx" % prefix
+
+
+    input_shapes = [(batch, seq_length), (batch, seq_length), (batch,)]
+    converted_model_path = mx.contrib.onnx.export_model(sym_file, params_file, input_shapes, np.float32, onnx_file)
+
+
+    # create onnxruntime session using the generated onnx file
+    ses_opt = onnxruntime.SessionOptions()
+    ses_opt.log_severity_level = 3
+    session = onnxruntime.InferenceSession(onnx_file, ses_opt)
+    onnx_inputs = [inputs, token_types, valid_length]
+    input_dict = dict((session.get_inputs()[i].name, onnx_inputs[i].asnumpy()) for i in range(len(onnx_inputs)))
+    pred_onx = session.run(None, input_dict)[0]
+
+    assert_almost_equal(seq_encoding, pred_onx)
+
+    shutil.rmtree(tmp_path)

Review comment:
       Use try finally




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org