在對訓練好的模型進行加載測驗時，出現了這樣的錯誤
tensorflow.python.framework.errors_impl.NotFoundError: Restoring from checkpoint failed. This is most likely due to a Variable name or other graph key that is missing from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:
Key pose_net/Stage2/unit_1/depthwise/depthwise_weights not found in checkpoint
[[node save/RestoreV2 (defined at test_kitti_pose.py:65) ]]
在網路結構里我用到一個tf.nn.depthwise_conv2d
這個函式的權重我已經定義了變數
W = tf.get_variable(
'depthwise_weights',
[kernel, kernel, in_channels, 1], dtype=tf.float32,
initializer=weights_initializer
)
開始訓練時后列印的可訓練變數中有這個引數pose_net/Stage2/unit_1/depthwise/depthwise_weights，但是后面查看節點檔案中卻沒有，這是什么原因。有沒有大神解答一下啊，很感謝！！
轉載請註明出處，本文鏈接：https://www.uj5u.com/qita/83940.html
標籤：機器視覺