{"id":13546201,"url":"https://github.com/pfnet-research/tgan","last_synced_at":"2025-04-13T10:27:07.798Z","repository":{"id":86583229,"uuid":"105860031","full_name":"pfnet-research/tgan","owner":"pfnet-research","description":"The implementation of Temporal Generative Adversarial Nets with Singular Value Clipping","archived":false,"fork":false,"pushed_at":"2020-07-07T07:24:54.000Z","size":17096,"stargazers_count":79,"open_issues_count":7,"forks_count":34,"subscribers_count":10,"default_branch":"master","last_synced_at":"2025-03-27T01:51:26.293Z","etag":null,"topics":["deep-learning","generative-adversarial-network","machine-learning","neural-networks","python"],"latest_commit_sha":null,"homepage":"https://pfnet-research.github.io/tgan","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/pfnet-research.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2017-10-05T07:14:35.000Z","updated_at":"2025-03-02T14:57:03.000Z","dependencies_parsed_at":"2023-03-17T05:15:40.812Z","dependency_job_id":null,"html_url":"https://github.com/pfnet-research/tgan","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pfnet-research%2Ftgan","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pfnet-research%2Ftgan/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pfnet-research%2Ftgan/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pfnet-research%2Ftgan/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pfnet-research","download_url":"https://codeload.github.com/pfnet-research/tgan/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248697387,"owners_count":21147320,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deep-learning","generative-adversarial-network","machine-learning","neural-networks","python"],"created_at":"2024-08-01T12:00:33.853Z","updated_at":"2025-04-13T10:27:07.774Z","avatar_url":"https://github.com/pfnet-research.png","language":"Python","funding_links":[],"categories":["Preferred Networks Research"],"sub_categories":["Services using Chainer"],"readme":"Temporal Generative Adversarial Nets\n====================================\n\n**The new version of TGAN has been published and the code is available: [TGANv2](https://github.com/pfnet-research/tgan2).**\n\nThis repository contains a collection of scripts used in the experiments of\n[Temporal Generative Adversarial Nets with Singular Value Clipping](https://arxiv.org/abs/1611.06624).\n\nDisclaimer: PFN provides no warranty or support for this implementation. Use it at your own risk. See [license](LICENSE.md) for details.\n\n## Results\n\n![](https://raw.githubusercontent.com/wiki/pfnet-research/tgan/images/ucf_cond_scaled.gif)\n\n## Requirements\n\nThese scripts require the following python libraries.\n\n- Chainer 2.0.0+\n- h5py\n- numpy\n- pandas\n- PIL\n- PyYAML\n- matplotlib\n\nNote that they also require ffmpeg to produce a video from a set of images.\n\n## Usage\n\n### Datasets\n\nIn order to run our scripts, you need to prepare MovingMNIST and UCF-101 datasets as follows.\n\n#### MovingMNIST\n\n1. Download `mnist_test_seq.npy` from [here](http://www.cs.toronto.edu/~nitish/unsupervised_video/).\n2. Put it on `path-to-tgans/data/mnist_test_seq.npy`.\n\n#### UCF-101\n\nThere are two ways to create an UCF-101 dataset for this script.\n\n1. Transforms all the videos in the UCF-101 dataset to the images.\n2. Resizes these images to the appropriate resolution, and concatenate\n   them into as single hdf5 format represented as (time, channel, rows, cols).\n   In this transformation we used ``make_ucf101.py`` in this repository.\n   Note that this script also produces a config file that describes videos and\n   these corresponding label information.\n3. puts them on `path-to-tgans/data`.\n\nAnother way is to simply download these files; please download them from\n[this url](https://www.dropbox.com/sh/j9fsakeuvicpeo8/AAD6BVhbZRyi7NXaMfn6TO4da?dl=0),\nand put them on the same directory.\n\n### Training\n\n#### TGAN with WGAN and Singular Value Clipping\n\n```\npython train.py --config_path configs/moving_mnist/mnist_wgan_svd_zdim-100_no-beta-all_init-uniform-all.yml --gpu 0\npython train.py --config_path configs/ucf101/ucf101_wgan_svd_zdim-100_no-beta.yml --gpu 0\n```\n\n#### TGAN (WGAN and weight clipping)\n\n```\npython train.py --config_path configs/moving_mnist/mnist_wgan_clip_zdim-100_no-beta-all_init-uniform-all.yml --gpu 0\npython train.py --config_path configs/ucf101/ucf101_wgan_clip_zdim-100_no-beta.yml --gpu 0\n```\n\n#### TGAN (vanilla GAN)\n\n```\npython train.py --config_path configs/ucf101/ucf101_vanilla_zdim-100_no-beta.yml --gpu 0\n```\n\n## Quantitative evaluation on UCF101 (2019/08/20)\n\nWe have uploaded ``mean2.npz`` on GitHub because there are many inquiries about the mean file in the UCF101.\nIf you want to perform a quantitative evaluation, please download it from\n[this url](https://github.com/pfnet-research/tgan/releases/download/v1.0.0/mean2.npz).\n\n## Citation\n\nPlease cite the paper if you are interested in:\n\n```\n@inproceedings{TGAN2017,\n    author = {Saito, Masaki and Matsumoto, Eiichi and Saito, Shunta},\n    title = {Temporal Generative Adversarial Nets with Singular Value Clipping},\n    booktitle = {ICCV},\n    year = {2017},\n}\n```\n\n## License\n\nMIT License. Please see the LICENSE file for details.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpfnet-research%2Ftgan","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpfnet-research%2Ftgan","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpfnet-research%2Ftgan/lists"}