fix(finetune): prevent data leakage in CustomKlineDataset normalization#263
Open
JasonOA888 wants to merge 1 commit intoshiyu-coder:masterfrom
Open
fix(finetune): prevent data leakage in CustomKlineDataset normalization#263JasonOA888 wants to merge 1 commit intoshiyu-coder:masterfrom
JasonOA888 wants to merge 1 commit intoshiyu-coder:masterfrom
Conversation
…t data leakage CustomKlineDataset was computing mean/std over the entire sliding window (lookback + predict + 1), leaking future data statistics into training. QlibDataset in finetune/dataset.py correctly uses only the lookback portion for this calculation.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Bug
CustomKlineDataset.__getitem__computesnp.mean(x)andnp.std(x)over the entire sliding window (lookback_window + predict_window + 1rows), which includes the prediction target period.This leaks future price statistics (mean and std of the prediction window) into the training features. The model can indirectly infer information about upcoming prices through the normalization parameters.
The sibling dataset
QlibDatasetinfinetune/dataset.pyhandles this correctly:Fix
Restrict normalization statistics to the lookback portion only (
x[:self.lookback_window]), matching the approach inQlibDataset.3 lines changed. No API changes.