python遇到的错误

在运行understand_facenet时，由于TensorFlow版本的问题遇到以下错误：
1、AttributeError: module ‘tensorflow’ has no attribute ‘GPUOptions’
原本：gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=gpu_memory_fraction)
改为：gpu_options = tf.compat.v1.GPUOptions(per_process_gpu_memory_fraction=gpu_memory_fraction)
2、AttributeError: module ‘tensorflow’ has no attribute ‘Session’. Did you mean: ‘version’?
原本：sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
改成：sess = tf.compat.v1.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
3、AttributeError: module ‘tensorflow’ has no attribute ‘ConfigProto’
原本：sess = tf.compat.v1.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
改为：sess = tf.compat.v1.Session(config=tf.compat.v1.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
还有其他类似的，都加上compat.v1,不过挺多的可以重新安装一下版本。
4、ValueError: Object arrays cannot be loaded when allow_pickle=False
原本：allow_pickle=allow_pickle
改成：allow_pickle=True
5、AttributeError: ‘int’ object has no attribute ‘value’
原本：feed_in, dim = (inp, input_shape[-1].value)
改为：feed_in, dim = (inp, input_shape[-1])
6、AttributeError: scipy.misc is deprecated and has no attribute imread.
解决：要不降级，要不改用
import imageio
content_image = imageio.imread
7、SyntaxError: from __ future__ imports must occur at the beginning of the file
__ future__ 必须在导入库的开头