For the Lyrics Generator challenge, my submission returns the ‘errored’ message for 2000 and 2004 character long txt files.
Please Help!
For the Lyrics Generator challenge, my submission returns the ‘errored’ message for 2000 and 2004 character long txt files.
Please Help!
hey @Aayushhh ,
can you please share link to me of your submission file , so that i can have a look on it and understand the issue.
Thank You .
hey @Aayushhh ,
i was just checking your file.
There is some problem in server side , so your file couldn’t be read.
Although , i checked it manually and yes there are some wrong predictions in your file too.
So can you please share a link to your code too. I need to have a look and understand that is it actually an issue of yours or do we need to change something on our server side.
I didn’t know how to share the code itself. Here’s the link to the file.
hey @Aayushhh ,
for better communication and understanding of the code .
I would request you to kindly upload your code on google colab and share its link here , in future.
Oka sure. Here’s the link:
https://colab.research.google.com/drive/1KKtM9bnS8IAVC-onZTc4nacsYR08EKza?usp=sharing
hey @Aayushhh ,
I am confused a bit too, as this much big difference shouldn’t happen.
But for now , adding np.random.seed(11)
worked to generate the correct data.
Try it once.
And what is the main reason behind the working , i will get it confirmed from the mentor and clear that.
Hey.
I do think it is an issue with the
np.random.seed(11).
I got 76% accuracy once or twice, just by changing the location of this seed.
Yeah thats the only point.
that just specifying a particular seed value ,might be at a different place ,is having a major impact on the output.
So, where do I put the seed value? Or should I just keep doing this via trial and error to find which location the challenge expects?
just put the seed command as the first code line inside the generate function.
I guess then , the code will work fine with good results.
and also seed value to be 11. I have found that it works good.