You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for your code! And I have some questions, glad for your help.
With the Ubuntu 18.04 and the tensorflow 0.12.0, when I run python mnist_cnn_test.py --model-dir model/model01_99.61 --batch-size 5000 --use-ensemble False
It don't work and return NotFoundError (see above for traceback): Key fc3/BatchNorm/moving_mean not found in checkpoint [[Node: save/RestoreV2_9 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_save/Const_0, save/RestoreV2_9/tensor_names, save/RestoreV2_9/shape_and_slices)]]
And when I run 'python mnist_cnn_test.py --model-dir model/model01_99.61 --batch-size 5000 --use-ensemble True'
The "accuracy“ only 0.092, All models are like this.
Looking forward to your reply.
The text was updated successfully, but these errors were encountered:
Thanks for your code! And I have some questions, glad for your help.
With the Ubuntu 18.04 and the tensorflow 0.12.0, when I run
python mnist_cnn_test.py --model-dir model/model01_99.61 --batch-size 5000 --use-ensemble False
It don't work and return
NotFoundError (see above for traceback): Key fc3/BatchNorm/moving_mean not found in checkpoint [[Node: save/RestoreV2_9 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_save/Const_0, save/RestoreV2_9/tensor_names, save/RestoreV2_9/shape_and_slices)]]
And when I run 'python mnist_cnn_test.py --model-dir model/model01_99.61 --batch-size 5000 --use-ensemble True'
The "accuracy“ only 0.092, All models are like this.
Looking forward to your reply.
The text was updated successfully, but these errors were encountered: