python - yamnet - keras 和 tensorflow

Question

我正在研究这里提供的 Yamnet 代码 https://github.com/tensorflow/models/tree/master/research/audioset/yamnet

我正在遵循这些步骤，但是在运行 python3 yamnet_test.py 时出现下图中的错误。我正在使用树莓派 4 。[![在此处输入图片描述][1]][1]

ERROR: testZeros (__main__.YAMNetTest)
YAMNetTest.testZeros
----------------------------------------------------------------------
Traceback (most recent call last):
  File "yamnet_test.py", line 53, in testZeros
    expected_class_name='Silence')
  File "yamnet_test.py", line 40, in clip_test
    predictions, embeddings, log_mel_spectrogram = YAMNetTest._yamnet(waveform)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 891, in __call__
    outputs = self.call(cast_inputs, *args, **kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/network.py", line 708, in call
    convert_kwargs_to_constants=base_layer_utils.call_context().saving)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/network.py", line 860, in _run_internal_graph
    output_tensors = layer(computed_tensors, **kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 891, in __call__
    outputs = self.call(cast_inputs, *args, **kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 2483, in call
    return self._defun_call(inputs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py", line 1822, in __call__
    graph_function, args, kwargs = self._maybe_define_function(args, kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py", line 2150, in _maybe_define_function
    graph_function = self._create_graph_function(args, kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py", line 2041, in _create_graph_function
    capture_by_value=self._capture_by_value),
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/framework/func_graph.py", line 915, in func_graph_from_py_func
    func_outputs = python_func(*func_args, **func_kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py", line 2658, in bound_method_wrapper
    return wrapped_fn(*args, **kwargs)
  File "/home/pi/env/lib/python3.7/site-packages/tensorflow_core/python/framework/func_graph.py", line 905, in wrapper
    raise e.ag_error_metadata.to_exception(e)
ValueError: in converted code:
    relative to /home/pi/env/lib/python3.7/site-packages/tensorflow_core/python:

    keras/engine/base_layer.py:2529 _defun_call  *
        return self._make_op(inputs)
    keras/engine/base_layer.py:2507 _make_op
        c_op = ops._create_c_op(graph, node_def, inputs, control_inputs=[])
    framework/ops.py:1613 _create_c_op
        raise ValueError(str(e))

    ValueError: Inconsistent values for attr 'T' DT_DOUBLE vs. DT_FLOAT while building NodeDef 'Shape' using Op<name=Shape; signature=input:T -> output:out_type; attr=T:type; attr=out_type:type,default=DT_INT32,allowed=[DT_INT32, DT_INT64]>


----------------------------------------------------------------------
Ran 4 tests in 7.322s

FAILED (errors=3, skipped=1)

标签： pythontensorflowaudio