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

i have a question #4

Open
unbizsw opened this issue Feb 5, 2021 · 1 comment
Open

i have a question #4

unbizsw opened this issue Feb 5, 2021 · 1 comment

Comments

@unbizsw
Copy link

unbizsw commented Feb 5, 2021

I tried y your code
and I did what you write and then ...
how to work is it and how can is show that video you posted .
do i have to make views? or just run that project ?

@anaramirli
Copy link
Owner

Hey! GIF example on the repository page, is just the visualization of our results (comparison of the target and predicted outcome), in order to see how it works.

All you need is the raw data form of the structure that is given in the feature_generation.ipynb (take a look there). Using feature_generation.ipynb you can create new data points for prediction. Later fitting this data in the pre-trained model, you can get the outcome.

for that, you will need the pre-trained model along with the normalization values (so you can normalize your new data points). Right now, I'm working on an update, as soon as I'm done (gonna take a few weeks), I'll provide the pre-trained model along with the normalization values. So, you u can run it. However, if you want to train your own model from scratch, right now we can provide the dataset. We will do to after we publish our paper.

Hope I answered your questions!

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

2 participants