Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

84% accuracy #2

Open
Steve1820 opened this issue May 3, 2018 · 3 comments
Open

84% accuracy #2

Steve1820 opened this issue May 3, 2018 · 3 comments

Comments

@Steve1820
Copy link

I can't seem to replicate the 84% accuracy number.

I changed the hidden layers to 20-10-6.

Any clues?

@minihat
Copy link
Owner

minihat commented May 3, 2018

Steve,
I've been working on final exams and haven't gotten the chance to add full documentation yet to replicate the results from my final project (see issue #1).

There is some code I need to add sitting on my local machine, but the primary thing you're missing is the algorithm I used to hand engineer features to use as inputs to the nn. The best accuracy I could get without the hand engineered features, on the raw data, was around 65%, so you can shoot for that until I get some better docs up.

Most of the hand engineered features were linear transforms of raw data; since these features are important to replicating my best result I'll get them posted when I can.

@Steve1820
Copy link
Author

Cool thanks!

Great work by the way, hope you upload it soon so I can learn from you.

Just getting started with Tensorflow - super interesting.

@Enochen
Copy link

Enochen commented Nov 11, 2018

Any updates on this? (on how you engineered the features)

Repository owner deleted a comment from 19890843006 Feb 23, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants