zhaokanghuiasd
commented
3 years ago 您好~请问您解决了吗？我也遇到了这个问题
Open padluo opened 3 years ago
zhaokanghuiasd
commented
3 years ago 您好~请问您解决了吗？我也遇到了这个问题
padluo
commented
3 years ago 您好~请问您解决了吗？我也遇到了这个问题
还未解决
kerlomz
commented
3 years ago 修改config.py的这部分内容如下：
RESIZE_MAP = { LossFunction.CTC: lambda x, y: [x, y], LossFunction.CrossEntropy: lambda x, y: [x, y] }
出现这个没什么关系，模型已经生成了其实，这是tf2onnx的模型不支持不定宽输出
模型已经训练完毕，在保存模型时，报错如下： Traceback (most recent call last): File "trains.py", line 336, in
tf.compat.v1.app.run()
File "/usr/local/anaconda3/lib/python3.7/site-packages/tensorflow/python/platform/app.py", line 40, in run
_run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
File "/usr/local/anaconda3/lib/python3.7/site-packages/absl_py-0.7.1-py3.7.egg/absl/app.py", line 300, in run
_run_main(main, args)
File "/usr/local/anaconda3/lib/python3.7/site-packages/absl_py-0.7.1-py3.7.egg/absl/app.py", line 251, in _run_main
sys.exit(main(argv))
File "trains.py", line 329, in main
Trains(model_conf).train_process()
File "trains.py", line 318, in train_process
self.compile_graph(accuracy)
File "trains.py", line 117, in compile_graph
self.compile_tflite(last_compile_model_path)
File "trains.py", line 63, in compile_tflite
tflite_model = converter.convert()
File "/usr/local/anaconda3/lib/python3.7/site-packages/tensorflow/lite/python/lite.py", line 811, in convert
_get_tensor_name(tensor), shape_list))
ValueError: None is only supported in the 1st dimension. Tensor 'input' has invalid shape '[None, None, 60, 1]'.