1.
error:
File "/Users/jasonjia/miniforge3/envs/pri_tf/lib/python3.8/site-packages/tensorflow/python/framework/constant_op.py", line 301, in _constant_eager_impl
t = convert_to_eager_tensor(value, ctx, dtype)
File "/Users/jasonjia/miniforge3/envs/pri_tf/lib/python3.8/site-packages/tensorflow/python/framework/constant_op.py", line 98, in convert_to_eager_tensor
return ops.EagerTensor(value, ctx.device_name, dtype)
ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type list).
reason:
y_train数据没有padding成相同的长度，因此一直报这个错误，这个搞了好久。
2.
error_info: input_spec.assert_input_compatibility(self.input_spec, inputs, self.name)
File "/Users/jasonjia/miniforge3/envs/pri_tf/lib/python3.8/site-packages/tensorflow/python/keras/engine/input_spec.py", line 271, in assert_input_compatibility
raise ValueError('Input ' + str(input_index) +
ValueError: Input 0 is incompatible with layer model_1: expected shape=(None, 150, 50, 1), found shape=(None, 7168)
reason: x_train 是图像数据，读取成一维的，reshape成二维的就ok了。
x_train = X.reshape(X.shape[0], height, width, 1)
3.
error_info:
File "/Users/jasonjia/miniforge3/envs/pri_tf/lib/python3.8/site-packages/tensorflow/python/keras/engine/input_spec.py", line 271, in assert_input_compatibility
raise ValueError('Input ' + str(input_index) +
ValueError: Input 1 is incompatible with layer model_1: expected shape=(None, 9), found shape=(None, 8, 1)
reason:
输入y_train格式错误 [[1],[2],[3],[4],[5],[6],[7],[8]] --> [1,2,3,4,5,6,7,8] 然后ok了
4:
error：
tensorflow.python.framework.errors_impl.InvalidArgumentError: Saw a non-null label (index >= num_classes - 1) following a null label, batch: 1 num_classes: 11 labels: 2,10,8,0,0,0,0,0 labels seen so far: 2
[[node model_1/ctc/CTCLoss (defined at test.py:174) ]] [Op:__inference_train_function_11652]
reason：
有人说要加占位符，也有人说直接“nclass = len(char_set)+1“
我查看了一下原来自己漏了一个类，识别0-10是个大写汉字，共11个，然后+1 nclass=12，我直接写成了11，修改之后，跑起来了，再看看有没有别的错误吧。
5. model跑起来之后出现了告警，不过不影响跑模型。
warning:
WARNING:tensorflow:AutoGraph could not transform
解决方法参考：https://stackoverflow.com/questions/66986540/warningtensorflowautograph-could-not-transform-function-preprocess-at-0x7f8f
查看了一下自己的gast是0.5.0，按照文中所说的，降到0.3.3告警消失