Ingest TS2Vec, and cap the probe as the authors do - #24
Merged
Conversation
61 of 66 datasets. EmoPain is the archive-wide input-validation failure. The other four, AustraliaRainfall_disc, Locust2022, Tiselac and USCActivity, timed out at exactly the 60 hour limit using about 3 GB, so time rather than memory. The cause is a gap in this port. The authors' fit_svm subsamples to MAX_SAMPLES=10000 with stratification before the grid search and returns the estimator fitted on that subsample; fit_lr does the same at 100000. This wrapper reimplemented the probe and omitted the cap, so on the large collections it ran ten values of C over five folds of an RBF SVC on the whole training set, which is between quadratic and cubic in the number of cases. probability=True, needed for predict_proba, adds internal Platt folds on top. Added probe_max_samples, defaulting to the authors' values, and probe_cases_ records what the probe was actually fitted on. The four timeouts are rerunnable, and the missing_results rows say so rather than recording them as a property of the method. TS2Vec lands 20th of 26 on rank, mean accuracy 0.7212, which is above TimesURL, the fork of it already in the archive. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
61 of 66 datasets. EmoPain is the archive-wide input-validation failure. The other four, AustraliaRainfall_disc, Locust2022, Tiselac and USCActivity, timed out at exactly the 60 hour limit using about 3 GB, so time rather than memory.
The cause is a gap in this port. The authors' fit_svm subsamples to MAX_SAMPLES=10000 with stratification before the grid search and returns the estimator fitted on that subsample; fit_lr does the same at 100000. This wrapper reimplemented the probe and omitted the cap, so on the large collections it ran ten values of C over five folds of an RBF SVC on the whole training set, which is between quadratic and cubic in the number of cases. probability=True, needed for predict_proba, adds internal Platt folds on top.
Added probe_max_samples, defaulting to the authors' values, and probe_cases_ records what the probe was actually fitted on. The four timeouts are rerunnable, and the missing_results rows say so rather than recording them as a property of the method.
TS2Vec lands 20th of 26 on rank, mean accuracy 0.7212, which is above TimesURL, the fork of it already in the archive.
Summary
Checklist
results/schema.md.