Tensorflow1. X and 2 How does x read those parameters saved in CKPT

about 1.x Of tf edition , There have been many introductions on the Internet , Here's the code :

import tensorflow as tf  
 
import os
from tensorflow.python import pywrap_tensorflow
 
model_dir='./model'# Set the folder where the model is located 
checkpoint_path = os.path.join(model_dir, "fineturing_model.ckpt")# location ckpt file 
#  from checkpoint Read out the data in the computer 
reader = pywrap_tensorflow.NewCheckpointReader(checkpoint_path)
# reader = tf.train.NewCheckpointReader(checkpoint_path) #  use tf.train Medium NewCheckpointReader Method 
var_to_shape_map = reader.get_variable_to_shape_map()
#  Output weight tensor Name and value 
for key in var_to_shape_map:
    print("tensor_name: ", key,reader.get_tensor(key).shape)
 

about tensorflow2.0 Come on , This code doesn't work properly , It's like :

module 'tensorflow.python.pywrap_tensorflow' has no attribute 'NewCheckpointReader'

Error of , This is the problem with the version , Some bloggers suggest reducing tf edition , But it's a little too expensive , Here's how tf2.0 The correct way to use ;

import tensorflow.compat.v1 as tf1

tf1.disable_v2_behavior()
checkpoint_path = 'xxx.ckpt'
# Read data from checkpoint file
reader = tf1.train.NewCheckpointReader(checkpoint_path)
var_to_shape_map = reader.get_variable_to_shape_map()
# Print tensor name and values
for key in var_to_shape_map:
    print("tensor_name: ", key)
    print(reader.get_tensor(key).shape)