{"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/tganv2-efficient-training-of-large-models-for","title":"Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GAN","arxiv_id":"1811.09245","date":"2018-11-22","proceeding":null,"authors":["Masaki Saito","Shunta Saito","Masanori Koyama","Sosuke Kobayashi"],"abstract":"Training of Generative Adversarial Network (GAN) on a video dataset is a challenge because of the sheer size of the dataset and the complexity of each observation. In general, the computational cost of training GAN scales exponentially with the resolution. In this study, we present a novel memory efficient method of unsupervised learning of high-resolution video dataset whose computational cost scales only linearly with the resolution. We achieve this by designing the generator model as a stack of small sub-generators and training the model in a specific way. We train each sub-generator with its own specific discriminator. At the time of the training, we introduce between each pair of consecutive sub-generators an auxiliary subsampling layer that reduces the frame-rate by a certain ratio. This procedure can allow each sub-generator to learn the distribution of the video at different levels of resolution. We also need only a few GPUs to train a highly complex generator that far outperforms the predecessor in terms of inception scores.","url_abs":"https://arxiv.org/abs/1811.09245v2","url_pdf":"https://arxiv.org/pdf/1811.09245v2.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":"tganv2-efficient-training-of-large-models-for","repo_url":"https://github.com/pfnet-research/tgan2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"tganv2-efficient-training-of-large-models-for","repo_url":"https://github.com/Zasder3/Tganv2-PyTorch-Train-Sparsely--Generate-Densely","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"video-generation","task_name":"Video Generation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-generation-on-ucf-101-16-frames-128x128","task":"Video Generation","dataset":"UCF-101 16 frames, 128x128, Unconditional","model":"TGANv2 (2020)","rank_in_archive_order":1,"of":6,"metrics":{"Inception Score":"28.87"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-ucf-101-16-frames-128x128","task":"Video Generation","dataset":"UCF-101 16 frames, 128x128, Unconditional","model":"TGANv2","rank_in_archive_order":4,"of":6,"metrics":{"Inception Score":"24.34"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-ucf-101-16-frames","task":"Video Generation","dataset":"UCF-101 16 frames, Unconditional, Single GPU","model":"TGANv2","rank_in_archive_order":2,"of":7,"metrics":{"Inception Score":"21.45"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.09245","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1811.09245"}},"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/pfnet-research/tgan2","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Zasder3/Tganv2-PyTorch-Train-Sparsely--Generate-Densely","reach":{"status":"ok"}}],"summary":{"unverified":6},"by_repo_kind":{"official":{"samples":6,"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":"38401635d6be718f","entry":"init_logger","repo":"pfnet-research/tgan2","repo_kind":"official","path":"tgan2/datasets/motion_jpeg.py","file_url":"https://github.com/pfnet-research/tgan2/blob/HEAD/tgan2/datasets/motion_jpeg.py","link_basis":"first_harvest_node","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":"38401635d6be718f"}},{"code_sha256_prefix":"550ecc473bd99ad0","entry":"make_config","repo":"pfnet-research/tgan2","repo_kind":"official","path":"tgan2/utils.py","file_url":"https://github.com/pfnet-research/tgan2/blob/HEAD/tgan2/utils.py","link_basis":"first_harvest_node","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":"550ecc473bd99ad0"}},{"code_sha256_prefix":"66377a4582ab9255","entry":"make_frame","repo":"pfnet-research/tgan2","repo_kind":"official","path":"tgan2/datasets/motion_jpeg.py","file_url":"https://github.com/pfnet-research/tgan2/blob/HEAD/tgan2/datasets/motion_jpeg.py","link_basis":"first_harvest_node","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":"66377a4582ab9255"}},{"code_sha256_prefix":"cd696a1b5338b469","entry":"make_instance","repo":"pfnet-research/tgan2","repo_kind":"official","path":"tgan2/utils.py","file_url":"https://github.com/pfnet-research/tgan2/blob/HEAD/tgan2/utils.py","link_basis":"first_harvest_node","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":"cd696a1b5338b469"}},{"code_sha256_prefix":"8a91291b41bbb98e","entry":"pooling","repo":"pfnet-research/tgan2","repo_kind":"official","path":"tgan2/datasets/multi_level.py","file_url":"https://github.com/pfnet-research/tgan2/blob/HEAD/tgan2/datasets/multi_level.py","link_basis":"first_harvest_node","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":"8a91291b41bbb98e"}},{"code_sha256_prefix":"932dd8314fd7500b","entry":"subsample","repo":"pfnet-research/tgan2","repo_kind":"official","path":"tgan2/datasets/multi_level.py","file_url":"https://github.com/pfnet-research/tgan2/blob/HEAD/tgan2/datasets/multi_level.py","link_basis":"first_harvest_node","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":"932dd8314fd7500b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}