search

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

1.modified load function to be pithier. 2.rename some varibles to be … #5

Closed zhaoyu611 closed 7 years ago

zhaoyu611 commented 7 years ago

…more apropriate 3. finetuen hyperparameter to get better result 4. add relu activiation

guillaume-chevalier commented 7 years ago

I am not ready yet to accept the pull request. Some things must be improved, reverted to what it was before or else cleaned. In my opinion, cleaning things and organizing the code properly is something more important than shipping the code itself, especially for a tutorial project for others such as this one. At least, I like the fact that you improved some aspects such as adding the ReLU activation to the first layer before the LSTM and also removing the device placement logging.