-
Notifications
You must be signed in to change notification settings - Fork 18
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
all output are nan?! #1
Comments
Hi Hualin, thanks for your message. I haven't looked at the code in a while, but I'll check this out when I get a chance. The code could do definitely do with some cleaning up. The functionality was based on a Matlab function, which I remember would return NaNs in the following circumstances:
Do either of these issues apply here? i.e. please could you try removing (1) all NaNs from your input array and (2) any duplicate data points from your input array and let me know if this helps? |
Cool, thanks for your fast reply. I guess sum of E_new or E is NaN because there was a value NaN in the array of E. Perhaps, delta = D-d. I believe I should do some cleanup routine on distance matrix for D and d, it should fix something. I will be back here. |
Hi there. I believe I got it fixed. Though I drop something:
and somehow. those changes make error gone! I didn't touch the optimisation process. cheers, |
Thanks Hualin - please could you put the suggested changes into a pull request? |
This should have been addressed in the merged pull request. |
hey, thanks for you code. but I run the sammontest.py, both raw and distance output are nan, because E-E_new array is all nan.
at first glance i am not sure where to fix it, since there Dinv but you catch the inf error and replace it with 0.
The text was updated successfully, but these errors were encountered: