{"id":21415686,"url":"https://github.com/npvisual/fast.ai-course","last_synced_at":"2026-07-16T00:32:22.940Z","repository":{"id":70645300,"uuid":"106113195","full_name":"npvisual/fast.ai-course","owner":"npvisual","description":"Notes, homeworks and other topics from the fast.ai course","archived":false,"fork":false,"pushed_at":"2017-10-07T17:45:14.000Z","size":826,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-03-16T18:56:40.894Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"wtfpl","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/npvisual.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"COPYING.WTFPL","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,"publiccode":null,"codemeta":null}},"created_at":"2017-10-07T16:35:21.000Z","updated_at":"2017-12-27T19:38:03.000Z","dependencies_parsed_at":"2023-06-10T17:00:37.868Z","dependency_job_id":null,"html_url":"https://github.com/npvisual/fast.ai-course","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/npvisual/fast.ai-course","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/npvisual%2Ffast.ai-course","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/npvisual%2Ffast.ai-course/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/npvisual%2Ffast.ai-course/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/npvisual%2Ffast.ai-course/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/npvisual","download_url":"https://codeload.github.com/npvisual/fast.ai-course/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/npvisual%2Ffast.ai-course/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35526112,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-15T02:00:06.706Z","response_time":131,"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":[],"created_at":"2024-11-22T18:51:47.122Z","updated_at":"2026-07-16T00:32:22.922Z","avatar_url":"https://github.com/npvisual.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Practical Deep Learning for Coders : My Journey\n\nCollection of notebooks and other documents generated as a result of following the [fast.ai class](http://course.fast.ai) : Practical Deep Learning for Coders.\n\n##### Table of contents\n\n* [Lesson 2 - Reusing VGG](#lesson2-reusingvgg) : notebook showing how to avoid importing all images to run predictions.\n* [Lesson 2 - Linear model]()\n\n\u003ca name=\"lesson2-reusingvgg\"/\u003e\n\n## Lesson 2 Homework : Reusing VGG\n\n### Introduction\n\n[Lesson 2 Homework - Training VGG](Lesson2/Lesson%202%20-%20Training%20VGG.ipynb) is a [Jupyter Notebook](https://jupyter.org) that I created as a hands-on approach to the [fast.ai class](http://course.fast.ai) Lesson 2.  Specifically, the [last part of that lesson](https://youtu.be/e3aM6XTekJc?t=1h29m41s) was somewhat problematic for me.\n\n### Reason for `get_data()` ?\n\nPart of the problem was my lack of understanding for using `get_data()` defined in `utils.py`. The more I read and discovered about it, the more I thought it was completely redundant to the use of batches that's done in that same notebook.\n\nIn addition there was a huge amount of memory being taken while running that notebook, regardless of whether or not prediction and training were performed.\n\n### Using batches\n\nSo after looking at some of the comments in the [Lesson 2 forum](http://forums.fast.ai/t/lesson-2-discussion/161) I decided to *re-write* the section of the notebook that uses predictions from the VGG model to feed to a single layer linear model. I used `predict_generator()` to run predictions on the batches we already defined and completely dropped the in-memory import of all the images done via `get_data()`.\n\nThis worked out pretty well as memory was kept in check and the original intent of the lesson was, in my mind, still preserved.\n\nHave fun going over it ! \n\n## References and Licenses\n\nAs far as my comments and not-so-original code, feel free to use [at will](COPYING.WTFPL). If you want to add attributions, it's always appreciated.\n\nFor the original content, please see below for attributions :\n\n* [Lesson 2 Notes](http://wiki.fast.ai/index.php/Lesson_2_Notes) : © fast.ai 2017. All rights reserved.\n* [Lesson 2 Video](https://youtu.be/e3aM6XTekJc) : [Creative Commons](https://creativecommons.org/)\n\nNote that I have had a hard time finding license information on the course notes, so I used the copyright show by the fast.ai web site. Please let me know if you know otherwise.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnpvisual%2Ffast.ai-course","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnpvisual%2Ffast.ai-course","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnpvisual%2Ffast.ai-course/lists"}