ValueError: None values not supported.

[1044197988]

Traceback (most recent call last): File "<pyshell#5>", line 1, in model = mask_rcnn.MaskRCNN(mode="inference", config=configs, model_dir=os.getcwd()) File "C:\Users\Admin\Desktop\666\Project\mask_rcnn\model.py", line 1907, in init self.keras_model = self.build(mode=mode, config=config) File "C:\Users\Admin\Desktop\666\Project\mask_rcnn\model.py", line 2114, in build [rpn_rois, mrcnn_class, mrcnn_bbox, input_image_meta]) File "C:\Users\Admin\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\keras\engine\base_layer.py", line 662, in call outputs = call_fn(inputs, *args, **kwargs) File "C:\Users\Admin\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\autograph\impl\api.py", line 169, in wrapper raise e.ag_error_metadata.to_exception(type(e)) ValueError: in converted code: relative to C:\Users\Admin:

Desktop\666\Project\mask_rcnn\model.py:880 call  *
    detections_batch = utils.batch_slice(
Desktop\666\Project\mask_rcnn\utils.py:835 batch_slice  *
    output_slice = graph_fn(*inputs_slice)
AppData\Local\Temp\tmphkl59he4.py:15 None  *
    detections_batch = ag__.converted_call('batch_slice', utils, ag__.ConversionOptions(recursive=True, force_conversion=False, optional_features=(), internal_convert_user_code=True), ([rois, mrcnn_class, mrcnn_bbox, window], lambda x, y, w, z: ag__.converted_call(refine_detections_graph, None, ag__.ConversionOptions(recursive=True, force_conversion=False, optional_features=(), internal_convert_user_code=True), (x, y, w, z, self.config), None), self.config.IMAGES_PER_GPU), None)
Desktop\666\Project\mask_rcnn\model.py:772 refine_detections_graph  *
    indices = tf.stack([tf.range(probs.shape[0]), class_ids], axis=1)
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\math_ops.py:1305 range
    limit = ops.convert_to_tensor(limit, dtype=dtype, name="limit")
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\ops.py:1100 convert_to_tensor
    return convert_to_tensor_v2(value, dtype, preferred_dtype, name)
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\ops.py:1158 convert_to_tensor_v2
    as_ref=False)
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\ops.py:1237 internal_convert_to_tensor
    ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\constant_op.py:305 _constant_tensor_conversion_function
    return constant(v, dtype=dtype, name=name)
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\constant_op.py:246 constant
    allow_broadcast=True)
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\constant_op.py:284 _constant_impl
    allow_broadcast=allow_broadcast))
AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\tensor_util.py:455 make_tensor_proto
    raise ValueError("None values not supported.")

ValueError: None values not supported.

你好，感谢您的代码 我在运行代码maskrcnn时候报错了 我的环境 win10 tensorflow-gpu-beta1 代码： model = mask_rcnn.MaskRCNN(mode="inference", config=configs,model_dir=os.getcwd()) 就报错了 因为这是一个巨大的项目，代码量很大，我很难检查出来

[Vishal-V]

Hi, I am currently working on fixing this issue. This is actually adapted from a Mask R-CNN implementation that was built for TF 1.8. Since TF 2.0 is eager by default, the batch utils cannot slice an eager tensor that has None in the shape and hence this error.

Thank you for checking it out. Will provide an update soon.

[1044197988]

Thank you and look forward to your next update I'm going to learn about this model