我使用main_model.py训练了11个类(0 =基本+ 10个实际类)的对象检测模型。之后,我要使用export_inference_graph.py导出推理图。我正在使用与训练完全相同的pipeline_config_path,并且trained_checkpoint_prefix参数引用了我训练过的model.ckpt。我收到以下错误:
InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [44] rhs shape= [360]
[[Node: save/Assign_526 = Assign[T=DT_FLOAT, _class=["loc:@SecondStageBoxPredictor/BoxEncodingPredictor/biases"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/device:CPU:0"](SecondStageBoxPredictor/BoxEncodingPredictor/biases, save/RestoreV2:526)]]
labelmap:https://drive.google.com/file/d/1isVO81rbYRGNrSboUd_DQh03DOxWV5is/view?usp=sharing
配置文件:https://drive.google.com/file/d/1vFkKbU5cytWMJwyt7tztLPxAnQ_bVnNo/view?usp=sharing
Python:3.6.2 Tensorflow:1.3.0
答案 0 :(得分:0)
您的labelmap和配置不见了。请确保您的num_classes = 11而不是90。