版本问题
pip install tensorflow@1.14.0
那就用cpu 重新训练呗
I am guessing the instructor used TensorFlow 1.x but TensorFlow has released version 2. The function regarding how to load mnist has been changed.
I don't have pinyin input. Sorry about that.
是uint8,不是
不然你再安TensorFlow试试。。。
跑是都能跑起来,这个项目中比较耗费资源的是网络训练部分,配置高低只是影响训练时间,MNIST训练,8g内存足够,如果独显NVIDIA GPU加速几分钟搞定,如果只用amd cpu的话可能要几个小时。
你用谷歌浏览器打开看看
是的嘞
对应一下版本是不是不一样呀
你问题解决了吗 我也遇到同样的问题
卸载tensorflow
pip uninstall tensorflow
安装tensorflow 1.14
pip install tensorflow==1.14.0
同名 了吧,应该是
解决了吗? 我的输出一个矩阵, 结果显示不到web页面上
WARNING:tensorflow:From D:/Python/workspace/Pybasis/minist/Handwritten_digit_main.py:11: The name tf.train.Saver is deprecated. Please use tf.compat.v1.train.Saver instead.WARNING:tensorflow:From D:\Python\Anaconda3\lib\site-packages\tensorflow\python\training\saver.py:1276: checkpoint_exists (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.Instructions for updating:Use standard file APIs to check for files with this prefix.Traceback (most recent call last): File "D:/Python/workspace/Pybasis/minist/Handwritten_digit_main.py", line 12, in <module> saver.restore(sess,"data/regression.ckpt") File "D:\Python\Anaconda3\lib\site-packages\tensorflow\python\training\saver.py", line 1278, in restore compat.as_text(save_path))ValueError: The passed save_path is not a valid checkpoint: data/regression.ckptProcess finished with exit code 1