fix: correct model objective in High-Freq Tree Alpha158 workflow config#2246
Open
Whning0513 wants to merge 1 commit into
Open
fix: correct model objective in High-Freq Tree Alpha158 workflow config#2246Whning0513 wants to merge 1 commit into
Whning0513 wants to merge 1 commit into
Conversation
Fixes microsoft#1739 The label is a continuous regression value (Ref($close, -2) / Ref($close, -1) - 1) but the model objective was set to "binary". Changed to "l2" regression to match the label semantics. Also updated metric from binary_logloss/auc to mae/mse/rmse appropriate for regression.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The high-frequency Alpha158 example config
workflow_config_High_Freq_Tree_Alpha158.yamlhas a mismatch between the label and the model objective. The labelRef($close, -2) / Ref($close, -1) - 1produces continuous regression values (e.g., 0.023, -0.015), but the model was set to"binary"— so LightGBM silently trains a classifier on regression targets. I checked the other example configs for reference: the daily-frequency counterpartworkflow_config_lightgbm_Alpha158.yamlusesloss: "l2"with the same label pattern, confirming regression is the intended behavior.Changed the objective from
"binary"to"l2"and updated metrics frombinary_logloss/auctomae/mse/rmse.Would appreciate a review, thanks!