{"id":21567689,"url":"https://github.com/sohaib90/ml-coding-practice","last_synced_at":"2025-03-18T05:42:07.019Z","repository":{"id":112154996,"uuid":"154486410","full_name":"Sohaib90/ML-Coding-Practice","owner":"Sohaib90","description":"ML Coding Practices","archived":false,"fork":false,"pushed_at":"2019-12-15T13:24:39.000Z","size":62257,"stargazers_count":1,"open_issues_count":1,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-01-24T12:25:16.583Z","etag":null,"topics":["100daysofmlcode","datascience","deeplearnign-ai","machinelearning-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/Sohaib90.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,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2018-10-24T11:02:58.000Z","updated_at":"2022-07-06T18:12:53.000Z","dependencies_parsed_at":null,"dependency_job_id":"d2f9c552-ae3c-4e9f-b5b6-6a49c58c85d7","html_url":"https://github.com/Sohaib90/ML-Coding-Practice","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/Sohaib90%2FML-Coding-Practice","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sohaib90%2FML-Coding-Practice/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sohaib90%2FML-Coding-Practice/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sohaib90%2FML-Coding-Practice/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Sohaib90","download_url":"https://codeload.github.com/Sohaib90/ML-Coding-Practice/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244166637,"owners_count":20409177,"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":["100daysofmlcode","datascience","deeplearnign-ai","machinelearning-python"],"created_at":"2024-11-24T10:32:20.192Z","updated_at":"2025-03-18T05:42:06.998Z","avatar_url":"https://github.com/Sohaib90.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Machine Learning Practice\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day1\"\u003e Day 1 : \u003c/a\u003e \u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day1/Data%20Pre-processing%20.ipynb\"\u003e Data Preprocessing for Machine Leanring Models\u003c/a\u003e \u003c/h3\u003e\n\u003cp\u003e Things learned : \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eData Selection: Consider what data is available, what data is missing and what data can be removed.\u003c/li\u003e\n\u003cli\u003eData Preprocessing: Organize selected data by formatting, cleaning and sampling from it. \u003c/li\u003e\n\u003cli\u003eData Transformation: Transform preprocessed data ready for machine learning by engineering features using scaling, attribute decomposition and attribute aggregation.\u003c/li\u003e\n  \u003cli\u003e Used numpy, matplotlib, numpy and sklearn libraries to handle, visualize and manipulate data \u003c/li\u003e\n\u003c/ul\u003e\n\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%202\"\u003e Day 2 : \u003c/a\u003e \u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%202/Support%20vector%20machine.ipynb\"\u003e Support Vector Machines: Implementation in Python \u003c/a\u003e \u003c/h3\u003e\n\u003cp\u003e Things learned : \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhat is SVM? Concepts and theory\u003c/li\u003e\n\u003cli\u003eImplementation in python: SVM and Kernel SVM\u003c/li\u003e\n\u003cli\u003eDifferent Kernel SVMs and comparison\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%203\"\u003e Day 3 : \u003c/a\u003e \u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%203/Decision-Trees-From-Scratch%20.ipynb\"\u003e Decison Trees: Implementation in Python (also sklearn implementation) \u003c/a\u003e \u003c/h3\u003e\n\u003cp\u003e Things learned : (tutorial helps from machinelearningmastery) \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eHow Decision Trees work\u003c/li\u003e\n\u003cli\u003eSplits made on the basis of entropy or Gini Index\u003c/li\u003e\n\u003cli\u003eImplementation in sklearn and from scratch\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%204\"\u003e Day 4 : \u003c/a\u003e \u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%204\"\u003e Logistic Regression and Feed Forward network to recognize handwritten digits (Tensorflow) \u003c/a\u003e \u003c/h3\u003e\n\u003cp\u003e Things learned : (Fundamentals of Deep Learning, Chapter 3) \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eLogistic Regression on MNIST Data\u003c/li\u003e\n\u003cli\u003eFeed Forward Network on MNIST data and comparison\u003c/li\u003e\n\u003cli\u003eTensorflow implementation : using variable scope and name scope for network\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e Day 5 : \u003c/h2\u003e\n\u003ch3\u003e Beyond Gradient Descent (Chapter 4 of Fundamentals of Deep Learning by Nikhil Budma)\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eChallenges with Gradient Descent: Local Minima and their effect in deep learning error surfaces\u003c/li\u003e\n\u003cli\u003eMomentum based Optimization: keeping memory of grdients for smoother error surfaces\u003c/li\u003e\n\u003cli\u003eLearning Rate Adaptation: (1) Adagrad  (2) RMSProp  (3) Adam \u003c/li\u003e\n  \u003cli\u003e Adagrad accumaltes and adapts the global learning rate using istorical gradients \u003c/li\u003e\n  \u003cli\u003e RMSProp is exponentially weighted moving average of gradients: it enables us to \"toss out\" measurements we made a long time ago \u003c/li\u003e\n  \u003cli\u003e Adam is the variant of both RMSProp and AutoGrad \u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e Day 6 : \u003c/h2\u003e\n\u003ch3\u003eMADL-Videos Day (Machine and Deep Learning- Videos Day)\u003c/h3\u003e\n\u003cp\u003e Watched Documentaries and videos relating ML and DL\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"https://www.netflix.com/watch/80190844?trackId=13752289\u0026tctx=0%2C0%2C4a044771-5e15-4157-8f03-2e9481d9a476-103080659%2C%2C\"\u003eWatched AlphaGo Documentary to learn how the game was built to be smart enough to beat the world champion\u003c/a\u003e\u003c/li\u003e\n  \u003cli\u003e\u003ca href=\"https://www.ted.com/talks/jeremy_howard_the_wonderful_and_terrifying_implications_of_computers_that_can_learn?language=en#t-1173351\"\u003eWatched Jeremy Howard's account on machine learning and computer vision integrated with machine learning \u003c/a\u003e\u003c/li\u003e\n  \u003cli\u003e\u003ca href=\"https://www.ted.com/talks/fei_fei_li_how_we_re_teaching_computers_to_understand_pictures#t-1052631\"\u003e Fei-Fei Li: How we're teaching computers to understand pictures \u003c/a\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%207\"\u003e Day 7 : \u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%207/naive_bayes_classifier.ipynb\"\u003eNaive Bayes Classification on diabetes dataset\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eImplementation of Naive Bayes Classifier in python\u003c/li\u003e\n  \u003cli\u003eClass probability and attribute probability based classifier\u003c/li\u003e\n  \u003cli\u003eFunctions for class probabilities and attribute probabilities\u003c/li\u003e\n  \u003cli\u003eMore work required on the concepts \u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e Day 8 : \u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://bloomberg.github.io/foml/#lectures\"\u003eLecture 2/3 of Bloomberg Foundation of Machine Learning: (2) Churn Prediction (3) Statistical Machine Learning\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eHow to think about a machine learning problem\u003c/li\u003e\n  \u003cli\u003eHowto think about the output and to analyse what we want to predict from the model\u003c/li\u003e\n  \u003cli\u003eWhile thinking about features and input values, think about the availability of all the data at deployment time\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%209\"\u003e Day 9 : \u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%209/guess%20word.py\"\u003eSpeech Recognition with Python, a simple Guess word game\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eHow to use speech recognition library in python for speech_recognition from microphone\u003c/li\u003e\n  \u003cli\u003eDevelop a small guessing game based on the recognized speech from microphone\u003c/li\u003e\n  \u003cli\u003eTheory of how it all works\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2010\"\u003e Day 10 : \u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%2010/convolutional_neural_networks.ipynb\"\u003eConvolutional Neural Networks: Introduction and Implementation\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhat are convolutional neural networks and what was the need? \u003c/li\u003e\n  \u003cli\u003eWhat are filters and feature maps and how convolution helps extracting features\u003c/li\u003e\n  \u003cli\u003eImplementation on MNIST data using Tensorflow\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2011\"\u003e Day 11 : \u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%2011/CIFAR10_convo_neural_network.ipynb\"\u003eConvolutional Neural Networks: Day 10 continued\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eBatch normalization and how it is helpful for training \u003c/li\u003e\n  \u003cli\u003eHow CIFAR 10 dataset is handled by using batch normalization\u003c/li\u003e\n  \u003cli\u003eImplementation of the network\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003eDay 12\u003c/h2\u003e\n\u003ch3\u003eChapter 6: Fundamentals of Deep Learning Book (Embedding and Representation Learning)\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eEmbedding and Representation Learning: A way to escape the curse of dimensionality \u003c/li\u003e\n  \u003cli\u003ePrincipal Component Analysis: concepts and mathematical formulation study\u003c/li\u003e\n  \u003cli\u003eAutoEncoders: Introduction and basic concepts\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e \u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2013\"\u003eDay 13\u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003eChapter 6 (continued): Fundamentals of Deep Learning Book (Embedding and Representation Learning)\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eDenoising Autoencoders: More Robust Autoencoders\u003c/li\u003e\n  \u003cli\u003eIntroducing Sparsity in Autoencoders\u003c/li\u003e\n  \u003cli\u003eWhen context is important in representations: English language as an example and their representation learning\u003c/li\u003e\n  \u003cli\u003e Coded an mnist autoencoder (using dense layers) in keras \u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e \u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2014\"\u003eDay 14\u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%2014/autoencoder_using_convo.ipynb\"\u003eAutoencoder: Continued (Using convolutional neural networks)\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eImplementing Autoencoders using convolutional neural networks\u003c/li\u003e\n  \u003cli\u003eUsing convo layers works better than fully connected layer in terms of reconstruction\u003c/li\u003e\n  \u003cliComparison on the basis of number of parameters and reconstruction, also error\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e \u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2015\"\u003eDay 15\u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%2015/pytorch_tutorials.ipynb\"\u003eDeep Learning and tutorials of Pytorch (A numpy based deep learning framework)\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e\u003ca href=\"https://pytorch.org/tutorials/beginner/deep_learning_60min_blitz.html\"\u003e Pytorch tutorial: \u003c/a\u003e\u003c/p\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003epytorch methods for building neural networks\u003c/li\u003e\n  \u003cli\u003eData loading and manipulations to tensors in pytorch\u003c/li\u003e\n  \u003cli\u003eAutograd and backprop concepts in pytorch\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e \u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2016\"\u003eDay 16\u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%2016/data_loading_preprocessing.ipynb\"\u003eData Loading and Processing with Pytorch (pytorch tutorials continued)\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e\u003ca href=\"https://pytorch.org/tutorials/beginner/data_loading_tutorial.html\"\u003e Pytorch tutorial: \u003c/a\u003e\u003c/p\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eHow to create your own custom dataloader\u003c/li\u003e\n  \u003cli\u003eHow to transform your data to make it same\u003c/li\u003e\n  \u003cli\u003eHow to use DataLoader pytorch class to enable batching, shuffling and parallel loading of data\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e Day 17\u003c/h2\u003e\n\u003ch3\u003eLeraning Pytorch with Examples\u003c/h3\u003e\n\u003cp\u003e\u003ca href=\"https://pytorch.org/tutorials/beginner/pytorch_with_examples.html\"\u003e Pytorch tutorial: \u003c/a\u003e\u003c/p\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eMain Features of Pytorch: n-dimensional tensors, kinda like numpy arrays but can run on GPU. Second important is Automatic differentiation for building and training networks\u003c/li\u003e\n  \u003cli\u003eDeeper understanding of Autograd and how pytorch builds computational graphs to computer gradients and weight updates\u003c/li\u003e\n  \u003cli\u003eDifference between Tensorflow and Pytorch: Dynamic and static graph\u003c/li\u003e\n  \u003cli\u003e Hw to build custom nn modules and optimizers \u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/tree/master/code/Day%2017\"\u003e Day 18 \u003c/a\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003ca href=\"https://github.com/Sohaib90/100-Days-of-ML-Code/blob/master/code/Day%2017/hymenoptera_data.zip\"\u003eTransfer Learning with Pytorch\u003c/a\u003e\u003c/h3\u003e\n\u003cp\u003e\u003ca href=\"https://pytorch.org/tutorials/beginner/transfer_learning_tutorial.html\"\u003e Pytorch tutorial: \u003c/a\u003e\u003c/p\u003e\n\u003cp\u003e Things learned \u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhat is transfer learning and the importance of transfer learning: fine-tuning, as a fixed feature extractor\u003c/li\u003e\n  \u003cli\u003eConstraints of Transfer learning\u003c/li\u003e\n  \u003cli\u003eImplementation of transfer learning using pytorch (help from pytorch tutorial)\u003c/li\u003e\n\u003c/ul\u003e\n\n\u003ch2\u003eDay 19 \u003c/h2\u003e\n\u003ch3\u003eRecurrent Neural Networks and Sequence-to-Sequence (Tutorials and concepts)\u003c/h3\u003e\n\u003cp\u003e\u003ca href=\"https://www.youtube.com/watch?time_continue=1777\u0026v=G5RY_SUJih4\"\u003e Lecture \u003c/a\u003e\u003c/p\u003e\n\n\u003ch2\u003eDay 20 \u003c/h2\u003e\n\u003ch3\u003eHow Google does Machine Learning (Learning google cloud platform and ML APIs)\u003c/h3\u003e\n\u003cp\u003eThe course took 5 days with extensive lab introduction and practice\u003c/p\u003e\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsohaib90%2Fml-coding-practice","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsohaib90%2Fml-coding-practice","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsohaib90%2Fml-coding-practice/lists"}