{"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/sgdr-stochastic-gradient-descent-with-warm","title":"SGDR: Stochastic Gradient Descent with Warm Restarts","arxiv_id":"1608.03983","date":"2016-08-13","proceeding":null,"authors":["Ilya Loshchilov","Frank Hutter"],"abstract":"Restart techniques are common in gradient-free optimization to deal with\nmultimodal functions. Partial warm restarts are also gaining popularity in\ngradient-based optimization to improve the rate of convergence in accelerated\ngradient schemes to deal with ill-conditioned functions. In this paper, we\npropose a simple warm restart technique for stochastic gradient descent to\nimprove its anytime performance when training deep neural networks. We\nempirically study its performance on the CIFAR-10 and CIFAR-100 datasets, where\nwe demonstrate new state-of-the-art results at 3.14% and 16.21%, respectively.\nWe also demonstrate its advantages on a dataset of EEG recordings and on a\ndownsampled version of the ImageNet dataset. Our source code is available at\nhttps://github.com/loshchil/SGDR","url_abs":"http://arxiv.org/abs/1608.03983v5","url_pdf":"http://arxiv.org/pdf/1608.03983v5.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":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/loshchil/SGDR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/Harshvardhan1/cyclic-learning-schedulers-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/Jmak12/Iris1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/PatriciaRodrigues1994/pytorch-facebook-udacity-challenge","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/SevenZhan/Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/abhuse/cyclic-cosine-decay","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/buptlwz/mabp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/jae1213/CyclicalLR_SGDR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/jolibrain/caffe","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/longjj/Caffe-SGDR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/rwightman/pytorch-image-models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/tmbluth/Distracted-Drivers-Detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/vinuni-vishc/transformer-gait-analysis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/yumatsuoka/check_cosine_annealing_lr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/1hb6s7t/SGDR-ms","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/HelloDuoLA/SGDR-Stochastic-Gradient-Descent-with-Warm-Restarts","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/MindCode-4/code-13/tree/main/SGDR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/MindCode-4/code-6/tree/main/Black-Box-Attacks-against","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/MindCode-4/code-9/tree/main/SGDR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"sgdr-stochastic-gradient-descent-with-warm","repo_url":"https://github.com/pwc-1/Paper-9/tree/main/3/SGDR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"eeg-1","task_name":"EEG"},{"task_slug":"eeg","task_name":"Electroencephalogram (EEG)"},{"task_slug":"stochastic-optimization","task_name":"Stochastic Optimization"}],"methods":[{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"}],"datasets_introduced":[],"methods_introduced":[{"slug":"cosine-annealing","name":"Cosine Annealing","full_name":"Cosine Annealing"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1608.03983","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1608.03983"}},"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/rwightman/pytorch-image-models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Harshvardhan1/cyclic-learning-schedulers-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vinuni-vishc/transformer-gait-analysis","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/PatriciaRodrigues1994/pytorch-facebook-udacity-challenge","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tmbluth/Distracted-Drivers-Detection","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pwc-1/Paper-9/tree/main/3/SGDR","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/loshchil/SGDR","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/abhuse/cyclic-cosine-decay","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jolibrain/caffe","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Jmak12/Iris1","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/1hb6s7t/SGDR-ms","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HelloDuoLA/SGDR-Stochastic-Gradient-Descent-with-Warm-Restarts","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindCode-4/code-13/tree/main/SGDR","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yumatsuoka/check_cosine_annealing_lr","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jae1213/CyclicalLR_SGDR","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/buptlwz/mabp","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindCode-4/code-6/tree/main/Black-Box-Attacks-against","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindCode-4/code-9/tree/main/SGDR","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/longjj/Caffe-SGDR","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/SevenZhan/Pytorch","reach":null}],"summary":{"ran":9,"ran_draft_wrong":1,"unverified":7},"by_repo_kind":{"listed":{"samples":17,"ran":10,"repositories":10}},"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":7,"samples":[{"code_sha256_prefix":"4f7499d2bb391700","entry":"CosineAnnealingLR_with_Restart","repo":"Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation","repo_kind":"listed","path":"config/res34_loss_naive_nodice_hp_scse_ds_border/loss/cyclic_lr.py","file_url":"https://github.com/Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation/blob/HEAD/config/res34_loss_naive_nodice_hp_scse_ds_border/loss/cyclic_lr.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4f7499d2bb391700"}},{"code_sha256_prefix":"f4a8879aa783ad15","entry":"CosineAnnealingWithRestartsLR","repo":"gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR","repo_kind":"listed","path":"CosineAnnealingWithRestartsLR.py","file_url":"https://github.com/gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR/blob/HEAD/CosineAnnealingWithRestartsLR.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f4a8879aa783ad15"}},{"code_sha256_prefix":"fe5e79b70f387bec","entry":"CosineLR","repo":"SevenZhan/Pytorch","repo_kind":"listed","path":"schedulers.py","file_url":"https://github.com/SevenZhan/Pytorch/blob/HEAD/schedulers.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fe5e79b70f387bec"}},{"code_sha256_prefix":"8a9fabc6ccb89932","entry":"CyclicCosAnnealingLR","repo":"Harshvardhan1/cyclic-learning-schedulers-pytorch","repo_kind":"listed","path":"cyclicLR.py","file_url":"https://github.com/Harshvardhan1/cyclic-learning-schedulers-pytorch/blob/HEAD/cyclicLR.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8a9fabc6ccb89932"}},{"code_sha256_prefix":"43331e0456fed368","entry":"CyclicCosAnnealingLR","repo":"bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch","repo_kind":"listed","path":"cyclicLR.py","file_url":"https://github.com/bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch/blob/HEAD/cyclicLR.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"43331e0456fed368"}},{"code_sha256_prefix":"926fab0aa32bd97a","entry":"CyclicCosineDecayLR","repo":"abhuse/cyclic-cosine-decay","repo_kind":"listed","path":"scheduler.py","file_url":"https://github.com/abhuse/cyclic-cosine-decay/blob/HEAD/scheduler.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"926fab0aa32bd97a"}},{"code_sha256_prefix":"e2d73df88181d0fb","entry":"LearningRateScheduler","repo":"MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts","repo_kind":"listed","path":"lib/Training/learning_rate_scheduling.py","file_url":"https://github.com/MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts/blob/HEAD/lib/Training/learning_rate_scheduling.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"e2d73df88181d0fb"}},{"code_sha256_prefix":"899d6ddaba2af2a7","entry":"PolyLR","repo":"Jmak12/Iris1","repo_kind":"listed","path":"optimizers/lr_scheduler.py","file_url":"https://github.com/Jmak12/Iris1/blob/HEAD/optimizers/lr_scheduler.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"899d6ddaba2af2a7"}},{"code_sha256_prefix":"6e9ad0df07eed6da","entry":"SGDRLR","repo":"jae1213/CyclicalLR_SGDR","repo_kind":"listed","path":"LRscheduler.py","file_url":"https://github.com/jae1213/CyclicalLR_SGDR/blob/HEAD/LRscheduler.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6e9ad0df07eed6da"}},{"code_sha256_prefix":"61bf152e6a42a184","entry":"window_reverse","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/modeling/backbone/swin.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/modeling/backbone/swin.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"61bf152e6a42a184"}},{"code_sha256_prefix":"803f499019d43ed0","entry":"CustomLR","repo":"jae1213/CyclicalLR_SGDR","repo_kind":"listed","path":"LRscheduler.py","file_url":"https://github.com/jae1213/CyclicalLR_SGDR/blob/HEAD/LRscheduler.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"803f499019d43ed0"}},{"code_sha256_prefix":"745bf09fefeb50c3","entry":"computeIoU","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/evaluation/refer_evaluation.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/evaluation/refer_evaluation.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"745bf09fefeb50c3"}},{"code_sha256_prefix":"58cc9ff3bf75e753","entry":"nested_tensor_from_tensor_list","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/utils/misc.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/utils/misc.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"58cc9ff3bf75e753"}},{"code_sha256_prefix":"755b208f128fbce6","entry":"prepare_data","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/data_prepare.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/data_prepare.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"755b208f128fbce6"}},{"code_sha256_prefix":"67dae4afd1febb5e","entry":"refer_bce_loss","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/modeling/criterion-so.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/modeling/criterion-so.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"67dae4afd1febb5e"}},{"code_sha256_prefix":"7b44eda6b63c9a3a","entry":"refer_ce_loss","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/modeling/criterion-so.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/modeling/criterion-so.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7b44eda6b63c9a3a"}},{"code_sha256_prefix":"2f4f1157ededdc6b","entry":"window_partition","repo":"buptlwz/mabp","repo_kind":"listed","path":"gres_model/modeling/criterion-so.py","file_url":"https://github.com/buptlwz/mabp/blob/HEAD/gres_model/modeling/criterion-so.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2f4f1157ededdc6b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}