tensorflow读取模型

两种方法：
直接读取模型，以下两种方式都是不可以的
错误方法一：
checkpoint_exclude_scopes = "voxel_decoder,detail"
if checkpoint_exclude_scopes:
exclusions = [scope.strip()
for scope in checkpoint_exclude_scopes.split(',')]

variables_to_restore = []
for var in slim.get_model_variables():
    excluded = False
    for exclusion in exclusions:
        if var.op.name.startswith(exclusion):
            excluded = True
            break
    if not excluded:
        variables_to_restore.append(var)
save = tf.train.Saver(variables_to_restore)
save.restore(sess, save_path)

错误方法2
checkpoint_include_scopes = "gaussian_MLP_encoder,bernoulli_MLP_decoder"
if checkpoint_include_scopes:
inclusions = [scope.strip()
for scope in checkpoint_include_scopes.split(',')]
variables_to_restore = []
for var in slim.get_model_variables():
for inclusion in inclusions:
if var.op.name.startswith(inclusion):
variables_to_restore.append(var)
save = tf.train.Saver(variables_to_restore)
save.restore(sess, save_path)

正确方案1：

inclusion = ['is_training','encoder','decoder']
inception_except_logits = slim.get_variables_to_restore(include=inclusion)
init_fn = slim.assign_from_checkpoint_fn(part_save_path, inception_except_logits, ignore_missing_vars=True)
init_fn(sess)

正确方案2：

exclusion = ['is_training','encoder','decoder']
inception_except_logits = slim.get_variables_to_restore(exclude=exclusion)
init_fn = slim.assign_from_checkpoint_fn(part_save_path, inception_except_logits, ignore_missing_vars=True)
init_fn(sess)