代码调试问题记录

1.conv2d_v2() got an unexpected keyword argument ‘filter’

把filter改成filters

2.tf.placeholder() is not compatible with eager execution

在文件前面加：
tf.compat.v1.disable_eager_execution()

3.TypeError: dropout_v2() got an unexpected keyword argument ‘keep_prob’

将属性keep_prob改为rate

4.AttributeError: module ‘tensorflow._api.v2.train’ has no attribute ‘AdamOptimizer’

optimizer = tf.train.AdamOptimizer改为tf.optimizers.Adam()

5.ValueError: tape is required when a Tensor loss is passed.

问题意思是：传了loss参数，就得传tape，所以代码修改为：

6.RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device(‘cpu’) to map your storages to the CPU.

把出错的这一句
model_class.load_state_dict(torch.load(classes_weights_path))
改为
model.load_state_dict(torch.load(opt.weights_path,map_location=‘cpu’))