Jul 10, 2017 · 1 min read
Hi,
I am trying to follow your example, but it is not converged if i use your pure source code.
You said the accuracy is reached up to 95% when you trained during 3 days, but it was not for me.
Would you please give me any advice or tip for me?
Now i am trying to train with higher layers (5 for input layers and 5 for output layers).
