I have tried many times and with many different parameters but accuracy always comes out to be negative during submission. Please tell me what am i doing wrong.
Here is my code: https://colab.research.google.com/drive/1k4rcMN1RJQrfFL1er59R15Wg7oJ9SRDt?usp=sharing
hey @Naman-Taneja-2420178691408660 ,
your code isn’t wrong , it works the way you made it.
Things to focus on and can be improved :
Preprocessing the data, you are normalizing X data , but i guess normalizing the output Y data can also be useful , you can try that too.
lag time that you are using , currently its 19 , but you can try other values too. Or even prepare data with multiple lags , like 7,14,21,35 and pass all them to model.
Preparing model architecture , the most important thing in Deep learning. You need to work a lot and improve it. You can try different combinations of those layers , adding a dropout after a LSTM is not always useful and using large number is also not useful sometimes , so thing and implement accordingly.
Metric is R2 , not accuracy. So prepare a model that works for this metric the best.
I guess these points can be a lot useful for you in this task.
I hope this helps.
Thank You 
.
OK, I will try it again.
Thanks for your help☺
hey @Naman-Taneja-2420178691408660 ,
Kindly mark this doubt as resolved in your course’s doubt platform.
and if in case you get a doubt then you can raise a new doubt on the portal.
Thank You .
I Tried all the things you told me still I was able to get a max accuracy of 29%.
Since I had never done such kind of a problem before this. I have researched about similar problems on the internet too and tried to apply those things into my code, still there was no improvement in accuraacy. It would be a great help if you could provide me with the solution of this problem. I can assure you that I won’t submit it until I understand it completely. The solution will be of a great help to me. Thank You
hey @Naman-Taneja-2420178691408660 ,
I respect your spirit but sorry i can’t share you the solution like this way.
Else , i can only say that , just try as much as you can, Learn and understand it.
You will get better score.
Alright, Thnx for your help
hey @Naman-Taneja-2420178691408660 ,
I request you to kindly mark this doubt as resolved in your course’s doubt section.
Thank You .
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.