Open ovsen opened 7 years ago
Hi, I understand your frustration in this area, it took me a long time to get working. The approach I took was to get the inception model working on a device, https://github.com/tensorflow/tensorflow/tree/master/tensorflow/examples/android and then incrementally change the android code to use my model. Also I was working with tensorflow r0.9, and using ubunutu. Regs
Hi
Thanks for the quick reply. I actually got it to work by adding new input and out put variables that take an input of only one image.
Thanks Gautham
Sent from my iPhone
On 27 Nov 2016, at 20:34, Brad notifications@github.com<mailto:notifications@github.com> wrote:
Hi, I understand your frustration in this area, it took me a long time to get working. The approach I took was to get the inception model working on a device, https://github.com/tensorflow/tensorflow/tree/master/tensorflow/examples/android and then incrementally change the android code to use my model. Also I was working with tensorflow r0.9, and using ubunutu. Regs
- You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/bvstillwell/udacity-mlnd-capstone/issues/1#issuecomment-263145676, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMMbKeJgx43Ic4NouMzwkq5enunfWo45ks5rCelkgaJpZM4K9Hx4.
I am having trouble importing a graph that I made in python into the Android demo app and running it. I built a CNN for character recognition. I am having problems finding the input and output nodes so I kept trying random logical combinations. Also I froze the graph and ran it through inference. There seems to be no documentation at all for importing your own classifier into Android. Both the data set and the ipython notebook are at: https://drive.google.com/drive/folders/0By50gGWStkpFZ1ZsajI0WHpFN0E?usp=sharing
In the ipython noteboook, the CNN at the bottom of the file is what I am trying to import to Android, the one with Dropout and MaxPooling.