{"id":17384665,"url":"https://github.com/c17hawke/data-augmentation-dmls","last_synced_at":"2026-05-01T03:31:12.343Z","repository":{"id":210352580,"uuid":"726189532","full_name":"c17hawke/Data-augmentation-DMLS","owner":"c17hawke","description":"Summary of Data Augmentation topics from book - \"Huyen, C. (2022). Designing Machine Learning Systems\"","archived":false,"fork":false,"pushed_at":"2023-12-02T05:06:40.000Z","size":18602,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-13T23:41:56.598Z","etag":null,"topics":["data-augmentation","python"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/c17hawke.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,"governance":null}},"created_at":"2023-12-01T18:17:32.000Z","updated_at":"2024-12-13T17:35:30.000Z","dependencies_parsed_at":"2023-12-02T07:39:58.319Z","dependency_job_id":null,"html_url":"https://github.com/c17hawke/Data-augmentation-DMLS","commit_stats":null,"previous_names":["c17hawke/data-augmentation-dmls"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/c17hawke/Data-augmentation-DMLS","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/c17hawke%2FData-augmentation-DMLS","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/c17hawke%2FData-augmentation-DMLS/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/c17hawke%2FData-augmentation-DMLS/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/c17hawke%2FData-augmentation-DMLS/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/c17hawke","download_url":"https://codeload.github.com/c17hawke/Data-augmentation-DMLS/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/c17hawke%2FData-augmentation-DMLS/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32484352,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-30T13:12:12.517Z","status":"online","status_checked_at":"2026-05-01T02:00:05.856Z","response_time":64,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["data-augmentation","python"],"created_at":"2024-10-16T07:46:12.365Z","updated_at":"2026-05-01T03:31:12.325Z","avatar_url":"https://github.com/c17hawke.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Data-Augmentation\n\nIt is a family of techniques to increase the amount of training data\n\nThere are 3 kinds of data augmentation techniques as per the author -\n\n1. Simple label-preserving transformation\n2. Perturbation\n3. Data synthesis\n\n|  | Simple label-preserving transformation | Perturbation | Data synthesis |\n| --- | --- | --- | --- |\n| What? | Random modification of data while preserving the label. | Adding noise to the data while preserving the label. | Use GANs to generate synthetic data. Can use costly DALL-E-like services as well. |\n| Examples in CV | Random flipping, Random rotation, etc. | Adding noise patches, or changing a single pixel values | Using CycleGAN to synthesize or generate new samples. |\n| Examples in NLP | Replacing words in a sentence with its synonyms | Adding random symbols, or words in a sentence  | Using templating to generate new samples |\n| Why? | Increase training sample per label/class | To improve model performance as well as evaluate model performance (i.e. How good is our model to adversarial attacks) | Increase training data using GAN techniques. |\n\n---\n\n## Example notebooks -\n\n- Example notebooks for CV - [link](https://github.com/c17hawke/Data-augmentation-DMLS/tree/main/notebooks/CV)\n- Example notebooks for NLP - [link](https://github.com/c17hawke/Data-augmentation-DMLS/tree/main/notebooks/NLP)\n\n## References -\n\n- [1] Huyen, C. (2022). Designing Machine Learning Systems: An Iterative Process for Production-Ready Applications\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fc17hawke%2Fdata-augmentation-dmls","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fc17hawke%2Fdata-augmentation-dmls","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fc17hawke%2Fdata-augmentation-dmls/lists"}