{"id":22771816,"url":"https://github.com/stonet2000/modern-tensorflow-fast-style-transfer","last_synced_at":"2025-03-30T12:15:07.113Z","repository":{"id":55580527,"uuid":"323175599","full_name":"StoneT2000/modern-tensorflow-fast-style-transfer","owner":"StoneT2000","description":null,"archived":false,"fork":false,"pushed_at":"2020-12-20T23:43:08.000Z","size":3187,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-02-05T13:52:44.119Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/StoneT2000.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-12-20T22:17:46.000Z","updated_at":"2020-12-20T23:43:02.000Z","dependencies_parsed_at":"2022-08-15T03:31:23.532Z","dependency_job_id":null,"html_url":"https://github.com/StoneT2000/modern-tensorflow-fast-style-transfer","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StoneT2000%2Fmodern-tensorflow-fast-style-transfer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StoneT2000%2Fmodern-tensorflow-fast-style-transfer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StoneT2000%2Fmodern-tensorflow-fast-style-transfer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/StoneT2000%2Fmodern-tensorflow-fast-style-transfer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/StoneT2000","download_url":"https://codeload.github.com/StoneT2000/modern-tensorflow-fast-style-transfer/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246314149,"owners_count":20757463,"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":[],"created_at":"2024-12-11T16:17:26.004Z","updated_at":"2025-03-30T12:15:07.090Z","avatar_url":"https://github.com/StoneT2000.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Modern Tensorflow Fast Style Transfer\n\nThis code implements [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://arxiv.org/abs/1508.06576) by Johnson and [Instance Normalization](https://arxiv.org/abs/1607.08022) by Ulyanov\n\nCode is inspired by https://github.com/lengstrom/fast-style-transfer and improves it by using Tensorflow 2.0 modules and syntax, creating cleaner, more readable code, as well showing efficiency gains through more efficient training and data loading\n\nThere are still some bugs and it is a work in progress:\n\nResult so far after training for approximately 3 hours on google cloud colab with the free GPU environment and applying the following style image:\n\n\u003cimg src=\"images/style/wave.jpg\" alt=\"great-wave-style\" style=\"zoom:50%;\" /\u003e\n\n![chicago-great-wave](images/readme/chicago.png)\n\n![boat-great-wave](images/readme/boat.png)\n\nThere's still some error with high frequency artifacts left in the image that I'm debugging.\n\n## Styling:\n\nTo style an image, run the following:\n\n```\npython src/style.py --content=path/to/content/image \\\n\t\t\t\t--ckpt=path/to/model/checkpoint \\\n\t\t\t\t--save-path=path/to/save/file\n```\n\n## Training\n\nFirst run `setup.sh` to download the MS-COCO dataset to use for training. This can take a while as it is \u003e 10 gb in size.\n\n```\npython src/train.py --style=path/to/style/image \\\n\t\t\t\t--ckpt=path/to/model/checkpoint\n```\n\nModel checkpoint is optional and provided if you wish to continue training from beforer\n\nThis will also save model checkpoints during trraining every 100 steps to a auto-created `tf_ckpts` folder from which you can use checkpoints from\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fstonet2000%2Fmodern-tensorflow-fast-style-transfer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fstonet2000%2Fmodern-tensorflow-fast-style-transfer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fstonet2000%2Fmodern-tensorflow-fast-style-transfer/lists"}