使用TensorFlow Keras子类API保存模型时出现NotFoundError

问题场景 #

使用TensorFlow Keras子类化API构建了WideAndDeepModel，完成训练、评估与预测后，调用model.save('my_model', save_format='tf')保存模型时触发NotFoundError，错误提示无法重命名临时变量文件，系统找不到指定路径。

模型代码 #

class WideAndDeepModel(tf.keras.Model):
    def __init__(self, units=30, activation="relu", **kwargs):
        super().__init__(**kwargs)  # needed to support naming the model
        self.norm_layer_wide = tf.keras.layers.Normalization()
        self.norm_layer_deep = tf.keras.layers.Normalization()
        self.hidden1 = tf.keras.layers.Dense(units, activation=activation)
        self.hidden2 = tf.keras.layers.Dense(units, activation=activation)
        self.main_output = tf.keras.layers.Dense(1)
        self.aux_output = tf.keras.layers.Dense(1)
        
    def call(self, inputs):
        input_wide, input_deep = inputs
        norm_wide = self.norm_layer_wide(input_wide)
        norm_deep = self.norm_layer_deep(input_deep)
        hidden1 = self.hidden1(norm_deep)
        hidden2 = self.hidden2(hidden1)
        concat = tf.keras.layers.concatenate([norm_wide, hidden2])
        output = self.main_output(concat)
        aux_output = self.aux_output(hidden2)
        return output, aux_output

tf.random.set_seed(42)  # extra code – just for reproducibility
model = WideAndDeepModel(30, activation="relu", name="my_cool_model")

optimizer = tf.keras.optimizers.Adam(learning_rate=1e-3)
model.compile(loss="mse", loss_weights=[0.9, 0.1], optimizer=optimizer,
              metrics=["RootMeanSquaredError"])
model.norm_layer_wide.adapt(X_train_wide)
model.norm_layer_deep.adapt(X_train_deep)
history = model.fit(
    (X_train_wide, X_train_deep), (y_train, y_train), epochs=10,
    validation_data=((X_valid_wide, X_valid_deep), (y_valid, y_valid)))
eval_results = model.evaluate((X_test_wide, X_test_deep), (y_test, y_test))
weighted_sum_of_losses, main_loss, aux_loss, main_rmse, aux_rmse = eval_results
y_pred_main, y_pred_aux = model.predict((X_new_wide, X_new_deep))

保存代码 #

model.save('my_model', save_format='tf')

错误信息 #

---------------------------------------------------------------------------
NotFoundError                             Traceback (most recent call last)
Cell In[94], line 1
----> 1 model.save('my_model', save_format='tf')

File ~\anaconda3\Lib\site-packages\keras\src\utils\traceback_utils.py:70, in filter_traceback.<locals>.error_handler(*args, **kwargs)
     67     filtered_tb = _process_traceback_frames(e.__traceback__)
     68     # To get the full stack trace, call:
     69     # `tf.debugging.disable_traceback_filtering()`
---> 70     raise e.with_traceback(filtered_tb) from None
     71 finally:
     72     del filtered_tb

File ~\anaconda3\Lib\site-packages\tensorflow\python\eager\execute.py:53, in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)
     51 try:
     52   ctx.ensure_initialized()
---> 53   tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
     54                                       inputs, attrs, num_outputs)
     55 except core._NotOkStatusException as e:
     56   if name is not None:

NotFoundError: {{function_node __wrapped__SaveV2_dtypes_43_device_/job:localhost/replica:0/task:0/device:CPU:0}} Failed to rename: my_model\variables\variables_temp/part-00000-of-00001.data-00000-of-00001.tempstate16756848698604293682 to: my_model\variables\variables_temp/part-00000-of-00001.data-00000-of-00001 : The system cannot find the path specified.
; No such process [Op:SaveV2]

解决方案 #

1. 提前创建保存所需的目录层级 #

错误提示找不到路径，可能是TensorFlow自动创建目录时出现异常，手动创建完整目录结构后再保存：

import os
# 创建完整的目录层级
os.makedirs('my_model/variables/variables_temp', exist_ok=True)
model.save('my_model', save_format='tf')

2. 使用绝对路径保存 #

相对路径可能因当前工作目录问题导致异常，换成绝对路径避免歧义：

import os
# 获取绝对路径
save_dir = os.path.abspath('my_model')
model.save(save_dir, save_format='tf')

3. 检查权限与进程占用 #

• 确认当前用户对目标保存目录有读写权限，如果是系统保护目录（如C:\Windows），换个非系统目录保存
• 关闭所有可能访问my_model目录的程序（如文件资源管理器、IDE的文件预览），避免目录被占用

4. 临时切换为HDF5格式保存 #

如果TF格式保存问题无法快速解决，可以先保存为HDF5格式，后续按需转换：

# 保存为HDF5
model.save('my_model.h5')

# 加载时需指定自定义模型类
loaded_model = tf.keras.models.load_model(
    'my_model.h5',
    custom_objects={'WideAndDeepModel': WideAndDeepModel}
)

5. 升级TensorFlow到最新稳定版 #

该错误可能是TensorFlow的已知bug，升级到最新版可修复部分路径相关的保存问题：

pip install --upgrade tensorflow

内容的提问来源于stack exchange，提问作者Alexander Peev