{"id":21170209,"url":"https://github.com/dcarpintero/fastai-deeplearning","last_synced_at":"2025-03-14T17:24:37.077Z","repository":{"id":221401191,"uuid":"754229698","full_name":"dcarpintero/fastai-deeplearning","owner":"dcarpintero","description":"Learning notes and projects built for the Deep Learning course by fast.ai","archived":false,"fork":false,"pushed_at":"2024-03-10T23:56:28.000Z","size":6230,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-11-20T15:56:55.531Z","etag":null,"topics":["data-augmentation","deep-learning","fastai","resnet-50"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/dcarpintero.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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":"2024-02-07T16:45:10.000Z","updated_at":"2024-02-15T21:44:47.000Z","dependencies_parsed_at":"2024-03-10T23:38:58.571Z","dependency_job_id":null,"html_url":"https://github.com/dcarpintero/fastai-deeplearning","commit_stats":null,"previous_names":["dcarpintero/fastai-deeplearning"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dcarpintero%2Ffastai-deeplearning","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dcarpintero%2Ffastai-deeplearning/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dcarpintero%2Ffastai-deeplearning/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dcarpintero%2Ffastai-deeplearning/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/dcarpintero","download_url":"https://codeload.github.com/dcarpintero/fastai-deeplearning/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":234938508,"owners_count":18910259,"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":["data-augmentation","deep-learning","fastai","resnet-50"],"created_at":"2024-11-20T15:57:08.815Z","updated_at":"2025-01-21T10:42:18.069Z","avatar_url":"https://github.com/dcarpintero.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Practical Deep Learning @ fast.ai\n\nLearning notes and projects built for the Deep Learning course by [fast.ai](https://course.fast.ai/).\n\n## 01. Getting Started - Bird Classifier\n\n[Notebook](./course2024/lesson_01.ipynb) |\n[Summary](./course2024/lesson_01.summary.md) |\n[Quiz](./course2024/lesson_01.quiz.md)\n\n## 02. Deployment - Astronomy Classifier\n\n[Notebook](./course2024/lesson_02.astronomy.ipynb) |\n[Summary](./course2024/lesson_01.summary.md) |\n[Quiz](./course2024/lesson_02.quiz.md) |\n[Model](https://huggingface.co/dcarpintero/fastai-interstellar-class) |\n[Try in HuggingFace Spaces](https://huggingface.co/spaces/dcarpintero/interstellar) \n\n\n`[deep-learning]` `[data-augmentation]` `[ResNet-50]` `[transfer-learning]`\n\nVisual learner to classify images of astronomical objects using ResNet and transfer learning (1 + 3 epochs). We provide two versions: `class model`, and `object model`.\n\nIn the `class model`, the possible labels are: `galaxy`, `nebula`, `comet`, `asteroid`, `quasar`, and `star cluster`.\n\nIn the `object model` the labels are specific astronomy objects: `m31 andromeda galaxy`, `m33 triangulum galaxy`, `m81 bode galaxy`, `m82 cigar galaxy`, `ngc 1300 galaxy`, `m104 sombrero galaxy`, `m51 whirlpool galaxy`, `m42 orion nebula`, `m17 omega nebula`, and `m45 pleiades star cluster`.\n\nThe datasets have been created using [Bing Search API](https://www.microsoft.com/en-us/bing/apis/bing-web-search-api), 150 images per class with augmentation.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./course2024/static/hg.00.png\"\u003e\n\u003c/p\u003e\n\nThe model reaches 84% accuracy on class level:\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./course2024/static/hg.01.png\"\u003e\n\u003c/p\u003e\n\nand 91% accuracy at object level:\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./course2024/static/hg.02.png\"\u003e\n\u003c/p\u003e\n\n## 03. Neural Net Foundations - Digit Classifier w/ Multi-Layer Perceptron\n\n[Summary](./course2024/lesson_03.summary.md) |\n[Quiz](./course2024/lesson_03.quiz.md) |\n[Digit Classifier w/ Multi-Layer Perceptron](./course2024/lesson_03.full.mnist.mlp.ipynb) |\n[Try in HuggingFace Spaces](https://huggingface.co/spaces/dcarpintero/mlp-digit-classifier) \n\n`[deep-learning]` `[perceptron]` `[backpropagation]` `[gradient-descend]` `[linear-layer]` `[relu]` `[optimizer]` `[mnist]` `[multi-class]` `[universal-approximation-theorem]`\n\n[Annotated Multi-Layer Perceptron](./course2024/lesson_03.full.mnist.mlp.md) trained on the MNIST dataset to classify handwritten digits. It  implements from scratch the following modules: **linear layer**, **relu-activation-function**, **sequential-layer**, **flatten-layer**, **basic optimizer**, and **learner**.\n\nWe define a hidden layer of size `256, 64` wherein the input tensors (28 x 28) are flattened in the height, and width dimensions. The model achieves 95.6% accuracy with `15 training epochs` and `batch size = 64`.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./course2024/static/mnist.hg.png\"\u003e\n\u003c/p\u003e","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdcarpintero%2Ffastai-deeplearning","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdcarpintero%2Ffastai-deeplearning","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdcarpintero%2Ffastai-deeplearning/lists"}