guillaume-chevalier / LSTM-Human-Activity-Recognition

Human Activity Recognition example using TensorFlow on smartphone sensors dataset and an LSTM RNN. Classifying the type of movement amongst six activity categories - Guillaume Chevalier
MIT License
3.33k stars 935 forks source link

Upgrade to the latest tensorflow ? #1

Closed StevenLOL closed 7 years ago

StevenLOL commented 7 years ago

The tensorflow used here is out-of-date, the latest tensorflow has something change in the rnn and rnn_cell. So it will be great if one can upgrade this project.

guillaume-chevalier commented 7 years ago

I updated it from version r0.8 to version r0.10 with this commit right here: https://github.com/guillaume-chevalier/LSTM-Human-Activity-Recognition/tree/3bacfe9b5a7e7d2b4dab3fa2f339408eeff3402b

Yet I did not test it with the newest official TensorFlow release r0.11 since I have other projects in parallel. If someone can confirm that it works in that version too it will be appreciated.

GDissard commented 7 years ago

I ran your code on the r0.11 and it worked :+1:

GDissard commented 7 years ago

however it doesnt work on the 0.12 version (on windows10). None of my LSTM scripts work.

Anybody knows what changed with this version ?