You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If I pick the first 6 items in areas array, this is the result I get
This plots were made by using matplotlib, not the plot_knee method. Just in case.
But, the weird thing I found comes when I add the next element into the array. So, for the first 7 items of the same array, the results are
There you can see the knee point changing from 1 to 2. But if we check it visually, let's say, the optimal point keeps on 1 where at least the blue plot shows a knee.
My first approach was changing the online/offline parameter. But the behavior remains the same.
I really appreciate your work!
Thanks in advance
PS, I wonder if the normalization applied to the orange curve is changing the result in these cases. I'm going to check that.
This discussion was converted from issue #79 on February 13, 2021 01:23.
Heading
Bold
Italic
Quote
Code
Link
Numbered list
Unordered list
Task list
Attach files
Mention
Reference
Menu
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hi!
I'm looking for your insight because I have been facing a situation I can't explain.
Let's suppose we have this original array:
Trying to find the knee point this way:
If I pick the first 6 items in areas array, this is the result I get
This plots were made by using matplotlib, not the plot_knee method. Just in case.
But, the weird thing I found comes when I add the next element into the array. So, for the first 7 items of the same array, the results are
There you can see the knee point changing from 1 to 2. But if we check it visually, let's say, the optimal point keeps on 1 where at least the blue plot shows a knee.
My first approach was changing the online/offline parameter. But the behavior remains the same.
I really appreciate your work!
Thanks in advance
PS, I wonder if the normalization applied to the orange curve is changing the result in these cases. I'm going to check that.
Beta Was this translation helpful? Give feedback.
All reactions