{"id":13631340,"url":"https://github.com/GoogleCloudPlatform/tensorflow-without-a-phd","last_synced_at":"2025-04-17T22:30:53.939Z","repository":{"id":43331681,"uuid":"132798273","full_name":"GoogleCloudPlatform/tensorflow-without-a-phd","owner":"GoogleCloudPlatform","description":"A crash course in six episodes for software developers who want to become machine learning practitioners.","archived":false,"fork":false,"pushed_at":"2024-05-03T05:16:18.000Z","size":77412,"stargazers_count":2807,"open_issues_count":30,"forks_count":914,"subscribers_count":152,"default_branch":"master","last_synced_at":"2025-04-12T05:05:35.266Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/GoogleCloudPlatform.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","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,"publiccode":null,"codemeta":null}},"created_at":"2018-05-09T18:33:40.000Z","updated_at":"2025-04-10T23:03:22.000Z","dependencies_parsed_at":"2022-07-12T18:18:56.095Z","dependency_job_id":"a0d45b10-9509-4b72-b2b9-d30b4e3ba2df","html_url":"https://github.com/GoogleCloudPlatform/tensorflow-without-a-phd","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/GoogleCloudPlatform%2Ftensorflow-without-a-phd","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/GoogleCloudPlatform%2Ftensorflow-without-a-phd/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/GoogleCloudPlatform%2Ftensorflow-without-a-phd/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/GoogleCloudPlatform%2Ftensorflow-without-a-phd/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/GoogleCloudPlatform","download_url":"https://codeload.github.com/GoogleCloudPlatform/tensorflow-without-a-phd/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248732484,"owners_count":21152852,"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":"2024-08-01T22:02:21.507Z","updated_at":"2025-04-17T22:30:53.910Z","avatar_url":"https://github.com/GoogleCloudPlatform.png","language":"Jupyter Notebook","readme":"\u003ctable width=\"100%\"\u003e\n    \u003ctr\u003e\n        \u003ctd width=\"50%\"\u003e\n            \u003cH2\u003eFeatured code sample\u003c/H2\u003e\n            \u003cb\u003e\u003ca href=\"tensorflow-planespotting\"\u003etensorflow-planespotting\u003c/a\u003e\u003c/b\u003e\u003cbr/\u003e\n            Code from the Google Cloud NEXT 2018 session \"Tensorflow, deep\n            learning and modern convnets, without a PhD\". Other samples from the \"Tensorflow without a PhD\" series are in\n            this repository too.\n        \u003ctd width=\"50%\"\u003e\u003ca href=\"https://youtu.be/KC4201o83W0\"\u003e\u003cimg alt=\"Tensorflow, deep\n        learning and modern convnets, without a PhD\" src=\"tensorflow-planespotting/img/next2018thumb.jpg\"/\u003e\u003c/a\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n\u003c/table\u003e\n\u003cbr/\u003e\n\n## Tensorflow and deep learning without a PhD series by [@martin_gorner](https://twitter.com/martin_gorner).\n\nA crash course in six episodes for software developers who want to learn machine learning, with examples, theoretical concepts,\nand engineering tips, tricks and best practices to build and train the neural networks that solve your problems.\n\n\u003ctable width=\"100%\"\u003e\n    \u003ctr\u003e\n        \u003ctd width=\"50%\"\u003e\u003cimg alt=\"Tensorflow and deep learning without a PhD\" src=\"docs/images/flds1.png\"/\u003e\u003c/td\u003e\n        \u003ctd width=\"50%\"\u003e\n            \u003cdiv align=\"center\"\u003e\n                     \u003ca href=\"https://youtu.be/u4alGiomYP4\"\u003evideo\u003c/a\u003e |\n                     \u003ca href=\"https://docs.google.com/presentation/d/1TVixw6ItiZ8igjp6U17tcgoFrLSaHWQmMOwjlgQY9co/pub?slide=id.p\"\u003eslides\u003c/a\u003e |\n                     \u003ca href=\"https://codelabs.developers.google.com/codelabs/cloud-tensorflow-mnist/#0\"\u003ecodelab\u003c/a\u003e |\n                     \u003ca href=\"tensorflow-mnist-tutorial\"\u003ecode\u003c/a\u003e\u003cbr/\u003e\u003cbr/\u003e\u003c/div\u003e\n                     \u003cp\u003eThe basics of building neural networks for software engineers. Neural weights and biases, activation functions, supervised learning and gradient descent.\n                     Tips and best practices for efficient training: learning rate decay, dropout regularisation and the intricacies of overfitting. Dense and convolutional neural networks. This session starts with low-level\n                     Tensorflow and also has a sample of high-level Tensorflow code using layers and Datasets. Code sample: MNIST handwritten digit recognition with 99% accuracy. Duration: 55 min\u003c/p\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd width=\"50%\"\u003e\u003cdiv align=\"center\"\u003e\n                                          \u003ca href=\"https://youtu.be/vq2nnJ4g6N0?t=76m\"\u003evideo\u003c/a\u003e |\n                                          \u003ca href=\"https://docs.google.com/presentation/d/18MiZndRCOxB7g-TcCl2EZOElS5udVaCuxnGznLnmOlE/pub?slide=id.g1245051c73_0_25\"\u003eslides\u003c/a\u003e |\n                                          \u003ca href=\"tensorflow-mnist-tutorial/README_BATCHNORM.md\"\u003ecode\u003c/a\u003e\u003cbr/\u003e\u003cbr/\u003e\u003c/div\u003e\n                                          \u003cp\u003eWhat is batch normalisation, how to use it appropriately and how to see if it is working or not.\n                                          Code sample: MNIST handwritten digit recognition with 99.5% accuracy. Duration: 25 min\u003c/p\u003e\u003c/td\u003e\n        \u003ctd width=\"50%\"\u003e\u003cimg alt=\"The superpower: batch normalization\" src=\"docs/images/flds2.png\"/\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd border=0 width=\"50%\"\u003e\u003cimg alt=\"Tensorflow, deep learning and recurrent neural networks, without a PhD\" src=\"docs/images/flds3.png\"/\u003e\u003c/td\u003e\n        \u003ctd border=0 width=\"50%\"\u003e\n            \u003cdiv align=\"center\"\u003e\n                 \u003ca href=\"https://youtu.be/fTUwdXUFfI8\"\u003evideo\u003c/a\u003e |\n                 \u003ca href=\"https://docs.google.com/presentation/d/18MiZndRCOxB7g-TcCl2EZOElS5udVaCuxnGznLnmOlE/pub?slide=id.p\"\u003eslides\u003c/a\u003e |\n                 \u003ca href=\"tensorflow-rnn-tutorial\"\u003ecodelab\u003c/a\u003e |\n                 \u003ca href=\"https://github.com/martin-gorner/tensorflow-rnn-shakespeare\"\u003ecode\u003c/a\u003e\u003cbr/\u003e\u003cbr/\u003e\u003c/div\u003e\n                 \u003cp\u003e RNN basics: the RNN cell as a state machine, training and unrolling (backpropagation through time).\n                 More complex RNN cells: LSTM and GRU cells. Application to language modeling and generation. Tensorflow APIs for RNNs.\n                 Code sample: RNN-generated Shakespeare play. Duration: 55 min\u003c/p\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd width=\"50%\"\u003e\u003cdiv align=\"center\"\u003e\n                  \u003ca href=\"https://youtu.be/KC4201o83W0\"\u003evideo\u003c/a\u003e |\n                  \u003ca href=\"https://docs.google.com/presentation/d/19u0Tm0JHL5tpzyarLILvy4qLSuDBFNNx2hwSvZsFPI0/pub\"\u003eslides\u003c/a\u003e |\n                  \u003ca href=\"tensorflow-planespotting\"\u003ecode\u003c/a\u003e\u003cbr/\u003e\u003cbr/\u003e\u003c/div\u003e\n                  \u003cp\u003eConvolutional neural network architectures for image processing. Convnet basics, convolution filters and how to stack them. \n                  Learnings from the Inception model: modules with parallel convolutions, 1x1 convolutions. A simple modern convnet architecture: Squeezenet.\n                  Convenets for detection: the YOLO (You Look Only Once) architecture. Full-scale model training and serving with Tensorflow's Estimator API on Google\n                  Cloud ML Engine and Cloud TPUs (Tensor Processing Units).\n                  Application: airplane detection in aerial imagery. Duration: 55 min\u003c/p\u003e\u003c/td\u003e\n        \u003ctd width=\"50%\"\u003e\u003cimg alt=\"Tensorflow, deep learning and modern convnets, without a PhD\" src=\"docs/images/flds4.png\"/\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n            \u003ctd border=0 width=\"50%\"\u003e\u003cimg alt=\"Tensorflow, deep learning and modern RNN architectures, without a PhD\" src=\"docs/images/flds5.png\"/\u003e\u003c/td\u003e\n            \u003ctd border=0 width=\"50%\"\u003e\n                \u003cdiv align=\"center\"\u003e\n                     \u003ca href=\"https://youtu.be/pzOzmxCR37I\"\u003evideo\u003c/a\u003e |\n                     \u003ca href=\"https://docs.google.com/presentation/d/17gLPozfb-l3WCR8FnejNJD9tEI_igTq1YqIXzCtOR14/pub\"\u003eslides\u003c/a\u003e |\n                     \u003ca href=\"https://github.com/conversationai/conversationai-models/tree/master/attention-tutorial\"\u003ecode\u003c/a\u003e\u003cbr/\u003e\u003cbr/\u003e\u003c/div\u003e\n                     \u003cp\u003eAdvanced RNN architectures for natural language processing. Word embeddings, text classification,\n                     bidirectional models, sequence to sequence models for translation. Attention mechanisms. This session also explores\n                     Tensorflow's powerful seq2seq API. Applications: toxic comment detection and langauge translation.\n                     Co-author: Nithum Thain. Duration: 55 min\u003c/p\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd width=\"50%\"\u003e\u003cdiv align=\"center\"\u003e\n            \u003ca href=\"https://youtu.be/t1A3NTttvBA\"\u003evideo\u003c/a\u003e |\n            \u003ca href=\"https://docs.google.com/presentation/d/1qLVvgKxZlM6_oOZ4-ZoOAB0wTh2IdhbFvuBhsMvmK9I/pub\"\u003eslides\u003c/a\u003e |\n            \u003ca href=\"tensorflow-rl-pong\"\u003ecode\u003c/a\u003e\u003cbr/\u003e\u003cbr/\u003e\u003c/div\u003e\n            \u003cp\u003e\n            A neural network trained to play the game of Pong from just the pixels of the game.\n            Uses reinforcement learning and policy gradients. The approach can be generalized to\n            other problems involving a non-differentiable step that cannot be trained using traditional supervised learning techniques.\n            A practical application: neural architecture search - neural networks designing neural networks. Co-author: Yu-Han Liu. Duration: 40 min\u003c/p\u003e\u003c/td\u003e\n        \u003ctd width=\"50%\"\u003e\u003cimg alt=\"Tensorflow and deep reinforcement learning, without a PhD\" src=\"docs/images/flds6.png\"/\u003e\u003c/td\u003e\n        \u003c/tr\u003e\n\u003c/table\u003e\n\u003cbr/\u003e\n\u003cbr/\u003e\n\u003cbr/\u003e\n\u003ctable width=\"75%\"\u003e\n    \u003ctr\u003e\u003ctd colspan=\"4\"\u003eQuick access to all code samples:\u003c/td\u003e\u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd width=\"33%\"\u003e\n                    \u003cb\u003e\u003ca href=\"tensorflow-mnist-tutorial\"\u003etensorflow-mnist-tutorial\u003c/a\u003e\u003c/b\u003e\u003cbr/\u003e\n                    dense and convolutional neural network tutorial\n                \u003c/td\u003e\n        \u003ctd width=\"33%\"\u003e\n            \u003cb\u003e\u003ca href=\"tensorflow-rnn-tutorial\"\u003etensorflow-rnn-tutorial\u003c/a\u003e\u003c/b\u003e\u003cbr/\u003e\n            recurrent neural network tutorial using temperature series\n        \u003c/td\u003e\n        \u003ctd width=\"33%\"\u003e\n            \u003cb\u003e\u003ca href=\"tensorflow-rl-pong\"\u003etensorflow-rl-pong\u003c/a\u003e\u003c/b\u003e\u003cbr/\u003e\n            \"pong\" with reinforcement learning\n        \u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n    \u003ctd width=\"33%\"\u003e\n        \u003cb\u003e\u003ca href=\"tensorflow-planespotting\"\u003etensorflow-planespotting\u003c/a\u003e\u003c/b\u003e\u003cbr/\u003e\n        airplane detection model\n    \u003c/td\u003e\n    \u003ctd width=\"33%\"\u003e\n            \u003cb\u003e\u003ca href=\"https://github.com/conversationai/conversationai-models/tree/master/attention-tutorial\"\u003econversationai: attention-tutorial\u003c/a\u003e\u003c/b\u003e\u003cbr/\u003e\n            Toxic comment detection with RNNs and attention\n        \u003c/td\u003e\n    \u003c/tr\u003e\n\u003c/table\u003e\n\u003cbr/\u003e\n\u003cbr/\u003e\n\u003cbr/\u003e\n*Disclaimer: This is not an official Google product but sample code provided for an educational purpose*","funding_links":[],"categories":["Jupyter Notebook","Deep Learning"],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FGoogleCloudPlatform%2Ftensorflow-without-a-phd","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FGoogleCloudPlatform%2Ftensorflow-without-a-phd","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FGoogleCloudPlatform%2Ftensorflow-without-a-phd/lists"}