Failing tests

[FRidh]

Tests fail with onnxruntime 1.12.1 built from source. Seems like InferenceSession needds to be instantiated with providers.

======================================================================
ERROR: test_auto_mixed_precision (test_auto_mixed_precision.AutoFloat16Test)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_auto_mixed_precision.py", line 43, in test_auto_mixed_precision
    expected = transpose_n_matmul(m1)
  File "/build/source/onnxconverter_common/onnx_fx.py", line 274, in __call__
    return Graph.inference_runtime(self.oxml, kwargs)
  File "/build/source/tests/test_onnxfx.py", line 12, in _ort_inference
    sess = _ort.InferenceSession(mdl.SerializeToString())
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 347, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 375, in _create_inference_session
    raise ValueError(
ValueError: This ORT build has ['DnnlExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['DnnlExecutionProvider', 'CPUExecutionProvider'], ...)

======================================================================
ERROR: test_auto_mixed_precision_model_path (test_auto_mixed_precision_model_path.AutoFloat16Test)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_auto_mixed_precision_model_path.py", line 32, in test_auto_mixed_precision_model_path
    expected = _ort_inference(model32_path, {'modelInput': input_x})
  File "/build/source/tests/test_auto_mixed_precision_model_path.py", line 13, in _ort_inference
    sess = _ort.InferenceSession(model_path)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 347, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 375, in _create_inference_session
    raise ValueError(
ValueError: This ORT build has ['DnnlExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['DnnlExecutionProvider', 'CPUExecutionProvider'], ...)

======================================================================
ERROR: test_float16 (unittest.loader._FailedTest)
----------------------------------------------------------------------
ImportError: Failed to import test module: test_float16
Traceback (most recent call last):
  File "/nix/store/sz0j8k8ljh7y8qgyfxgqb3ws11bcy4gs-python3-3.10.6/lib/python3.10/unittest/loader.py", line 436, in _find_test_path
    module = self._get_module_from_name(name)
  File "/nix/store/sz0j8k8ljh7y8qgyfxgqb3ws11bcy4gs-python3-3.10.6/lib/python3.10/unittest/loader.py", line 377, in _get_module_from_name
    __import__(name)
  File "/build/source/tests/test_float16.py", line 7, in <module>
    import onnxmltools
ModuleNotFoundError: No module named 'onnxmltools'


======================================================================
ERROR: test_onnx2py (test_onnx2py.Onnx2PyTests)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_onnx2py.py", line 27, in test_onnx2py
    sess1 = _ort.InferenceSession(onnx_model.SerializeToString())
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 347, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 375, in _create_inference_session
    raise ValueError(
ValueError: This ORT build has ['DnnlExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['DnnlExecutionProvider', 'CPUExecutionProvider'], ...)

======================================================================
ERROR: test_onnx2py (test_onnx2py.Onnx2PyTests)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_onnx2py.py", line 15, in tearDown
    for f in os.listdir(tmp_path):
FileNotFoundError: [Errno 2] No such file or directory: '/build/source/tests/temp'

======================================================================
ERROR: test_core (test_onnxfx.ONNXFunctionTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_onnxfx.py", line 34, in test_core
    np.allclose(g([2.0], [-5.0]), np.array([2.0])))
  File "/build/source/onnxconverter_common/onnx_fx.py", line 274, in __call__
    return Graph.inference_runtime(self.oxml, kwargs)
  File "/build/source/tests/test_onnxfx.py", line 12, in _ort_inference
    sess = _ort.InferenceSession(mdl.SerializeToString())
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 347, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 375, in _create_inference_session
    raise ValueError(
ValueError: This ORT build has ['DnnlExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['DnnlExecutionProvider', 'CPUExecutionProvider'], ...)

======================================================================
ERROR: test_loop (test_onnxfx.ONNXFunctionTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_onnxfx.py", line 58, in test_loop
    loop_test(np.array([16], dtype=np.int64))[2][4], 3.0)
  File "/build/source/onnxconverter_common/onnx_fx.py", line 274, in __call__
    return Graph.inference_runtime(self.oxml, kwargs)
  File "/build/source/tests/test_onnxfx.py", line 12, in _ort_inference
    sess = _ort.InferenceSession(mdl.SerializeToString())
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 347, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 375, in _create_inference_session
    raise ValueError(
ValueError: This ORT build has ['DnnlExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['DnnlExecutionProvider', 'CPUExecutionProvider'], ...)

======================================================================
ERROR: test_matmul_opt (test_onnxfx.ONNXFunctionTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/build/source/tests/test_onnxfx.py", line 75, in test_matmul_opt
    expected = transpose_n_matmul(m1)
  File "/build/source/onnxconverter_common/onnx_fx.py", line 274, in __call__
    return Graph.inference_runtime(self.oxml, kwargs)
  File "/build/source/tests/test_onnxfx.py", line 12, in _ort_inference
    sess = _ort.InferenceSession(mdl.SerializeToString())
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 347, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "/nix/store/vy5fnk87f62mzqaaw3l208psi7lvpfns-onnxruntime-1.12.1-python/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 375, in _create_inference_session
    raise ValueError(
ValueError: This ORT build has ['DnnlExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['DnnlExecutionProvider', 'CPUExecutionProvider'], ...)

----------------------------------------------------------------------
Ran 27 tests in 1.060s

When needed I can provide the exact set of packages/build parameters as this was built using Nix.

[xiaowuhu]

@FRidh Hi, this issue still bother you? We don't have official testing on Nix.

[xiaowuhu]

if this still bother you, you can reopen it.

[tobim]

This is still an issue.