Replies: 1 comment
-
@ek-ak Could you help me with this? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
@ek-ak Could you help me with this? |
Beta Was this translation helpful? Give feedback.
-
catboost version: catboost 1.2
Operating System: macos 14.4.1
CPU: M1
GPU: no
In the paper, Algorithm 2 shows that the loss is obtained by calculating the cosine similarity of gradient and delta(leaf output). Please help me check if my understanding of calculating delta values is correct.
For example, I have a instance whose index is i, which is assigned to the left leaf node. So the prediction value for instance i is the average of the first i-1 instances whose allocated into left leaf node. We call the average value as j. Does the value of delta (i) is also the value j?
Here is the example for building process for regression task when split value for feature1 is 0.4 , does process I presented correct?
Beta Was this translation helpful? Give feedback.
All reactions