{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/performance-prediction-under-dataset-shift","title":"Performance Prediction Under Dataset Shift","arxiv_id":"2206.10697","date":"2022-06-21","proceeding":null,"authors":["Simona Maggio","Victor Bouvier","Léo Dreyfus-Schmidt"],"abstract":"ML models deployed in production often have to face unknown domain changes, fundamentally different from their training settings. Performance prediction models carry out the crucial task of measuring the impact of these changes on model performance. We study the generalization capabilities of various performance prediction models to new domains by learning on generated synthetic perturbations. Empirical validation on a benchmark of ten tabular datasets shows that models based upon state-of-the-art shift detection metrics are not expressive enough to generalize to unseen domains, while Error Predictors bring a consistent improvement in performance prediction under shift. We additionally propose a natural and effortless uncertainty estimation of the predicted accuracy that ensures reliable use of performance predictors. Our implementation is available at https: //github.com/dataiku-research/performance_prediction_under_shift.","url_abs":"https://arxiv.org/abs/2206.10697v1","url_pdf":"https://arxiv.org/pdf/2206.10697v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"performance-prediction-under-dataset-shift","repo_url":"https://github.com/dataiku-research/performance_prediction_under_shift","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2206.10697","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2206.10697"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dataiku-research/performance_prediction_under_shift","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":16},"by_repo_kind":{"official":{"samples":16,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"005d05c5fb2709d2","entry":"any_other_label","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/perturbation_shared_utils.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/perturbation_shared_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"005d05c5fb2709d2"}},{"code_sha256_prefix":"2b7902ac234fdd50","entry":"get_X_y","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac_experiments/perf_drop_experiment_utils.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac_experiments/perf_drop_experiment_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2b7902ac234fdd50"}},{"code_sha256_prefix":"d13b27c3bade1017","entry":"is_integer_feature","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/features_utils.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/features_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d13b27c3bade1017"}},{"code_sha256_prefix":"8dcd025f2071ff67","entry":"knockout_shift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/prior_shift.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/prior_shift.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8dcd025f2071ff67"}},{"code_sha256_prefix":"d81e69bf33734e70","entry":"only_one_shift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/prior_shift.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/prior_shift.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d81e69bf33734e70"}},{"code_sha256_prefix":"ae1884f61848e0f1","entry":"prediction_drift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/drift_metrics.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/drift_metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ae1884f61848e0f1"}},{"code_sha256_prefix":"f70c0142ed4deeba","entry":"rebalance_shift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/prior_shift.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/prior_shift.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f70c0142ed4deeba"}},{"code_sha256_prefix":"0a325b447708425d","entry":"resize_dataset","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac_experiments/multi_domain_performance_predictor.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac_experiments/multi_domain_performance_predictor.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0a325b447708425d"}},{"code_sha256_prefix":"cab47f390a020275","entry":"retrieve_domain_df","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac_experiments/perf_drop_experiment_utils.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac_experiments/perf_drop_experiment_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cab47f390a020275"}},{"code_sha256_prefix":"c945e0892ad1b1e4","entry":"sample_random_indices","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/perturbation_shared_utils.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/perturbation_shared_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c945e0892ad1b1e4"}},{"code_sha256_prefix":"d18b5cfe530340f2","entry":"split_into_domains","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac_experiments/perf_drop_experiment_utils.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac_experiments/perf_drop_experiment_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d18b5cfe530340f2"}},{"code_sha256_prefix":"b32f1cda2af5ba27","entry":"subsample_feature_shift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/covariate_shift.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/covariate_shift.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b32f1cda2af5ba27"}},{"code_sha256_prefix":"41370631ad5ffc8a","entry":"subsample_joint_shift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/covariate_shift.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/covariate_shift.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"41370631ad5ffc8a"}},{"code_sha256_prefix":"bbe9076675db4cbe","entry":"subsample_one_feature_shift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/covariate_shift.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/covariate_shift.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bbe9076675db4cbe"}},{"code_sha256_prefix":"372e7aa065fc8cf5","entry":"test_chi2","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/drift_metrics.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/drift_metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"372e7aa065fc8cf5"}},{"code_sha256_prefix":"8b37b6c06ce6a1bf","entry":"univariate_drift","repo":"dataiku-research/performance_prediction_under_shift","repo_kind":"official","path":"drift_dac/drift_metrics.py","file_url":"https://github.com/dataiku-research/performance_prediction_under_shift/blob/HEAD/drift_dac/drift_metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8b37b6c06ce6a1bf"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}