I accept the challenge of Siraj Raval of 100 days of ML. I will learn and code machine learning for at least 1 hour everyday.
Day 1 : Stock Price Predictor
Day 2 : Twitter Sentiment Analysis and Udacity Intro to Machine Learning - Lesson 11 Text Learning
Day 3 : Created labelled CSV dataset of twitter tweets and its sentiments and trained a neural network model to classify images of clothing, like sneakers and shirts.
Day 4 : Learned Bayes Theorem and Naive Bayes Classification Algorithm. Wrote a small program using sklearn to demonstrate Gaussian Naive Bayes.
Day 5 : Built a Naive Bayes Classifier from Scratch using python and numpy package. Reference: https://machinelearningmastery.com/naive-bayes-classifier-scratch-python/
Day 6 : Learned Decision Tree Classifier and implemented it using sklearn on iris dataset. Started learning Pandas library.
Day 7 : Visualizing Data from Data Science from Scratch : First Principles with Python.
Day 8 : Learnt Pandas from Kaggle -
1.Creating, reading and writing workbook,
2.Indexing, Selecting & Assigning,
3. Summary functions and maps workbook
Day 9 : 4. Grouping and Sorting
5. Datatypes and missing data workbook
6. Renaming and combining workbook
7. Method chaining workbook
Day 10 : Movie Recommendation using LightFM dependency and movielens dataset
Day 11 : Done some coding in Python for All Women Hackthon - Hacckerrank
Day 12 : Studied Feature Selection
Day 13 : Built a spam filter using Naive Bayes classifier and enron dataset
Day 14 : Working on building Spam classifier from scratch using Pandas and Numpy library
Day 15 : Predicted house prices using Linear Regression.
Day 16 : Visualization with Seaborn
Day 17 : Built a single layer neural network using Python Numpy.
Day 18 : Built decision tree from scratch using Python.
Day 19 : Built KNN classifier from scratch
Day 20 : Chapter 1 of Data Science from Scratch : Motivating Hypothetical - DataSciencester
Day 21 : Chapter 4 : Linear Algebra and Chapter 5: Statistics (Revised some mathemathical concepts).
Day 22 : Completed Quiz project task of Feature Selection and started new concept - PCA - Principal Component Analysis in Udacity Into to ML
Day 23 : Completed PCA and Validation lessons
Day 24 : Completed Evaluation Metric lesson
Day 25, 26, 27: Completed project on Enron dataset
Day 28 : Implemented simple linear regression from scratch.
Day 29 : Studied gradient descent.
Day 30 : Watched Andrew Ng videos on multiple linear regression and implemented it.
Day 31 : Studied Logistic Regression
Day 32, 33 : Built Binary Classification Logistic Regression from scratch using python
Day 34, 35 : Learned Tensorflow basics
Day 36, 37, 38: Working on Kaggle's Titanic: Machine Learning from Disaster
Day 39, 40: Watched youtube videos of Siraj and read medium posts.
Day 41, 42: Started Statistic course of Udacity.
Day 43 : Learned from Coursera CNN course - Edge Detection, Padding, Strided cnn, pooling.
Day 44 : Implemnted linear regression using Tensorflow
Day 45 : Working on MNIST digit recognizer.
Day 46 : Started learning Angular and Node.js side by side.
Day 47 : Predicting Boston Housing Prices using Tensorflow.
Day 48 : Completed 7 lessons of Udacity Intro to Statistics
Day 49 : Completed Kaggle's Digit Recognizer.
Day 50 : Learned basics of CNN and wrote edge detection code.
Day 51 : Learned classic neural network architectures - LeNet, AlexNet, VGG-16, ResNet.
Day 52 : Used Google Collab, learned to load Kaggle's dataset in it. Also learnt its basic functionality
Day 53 : Working on Cats vs Dogs with ConvNet using Keras
Day 54 : Created conv model similar to VGG-16 in Keras to classify between a dog and cat, also learned data augmentation
Day 55 : Published my first story on Medium regarding Google Colab
Day 56 : Completed the Cat vs Dog classification. Although achieved just 70% accuracy, studying related to overfitting problems, and use of optimizers and callbacks. Will try to achieve better accuracy.
Day 57 : Wrote my 2nd @Medium story - on how to import data to google colab.
Day 58 : Read and watched videos on Transfer learning
Day 59 : Wrote ML code using Logistic Regression to predict whether a patient has diabetes or not.
Day 60 : Started learning Flask.
Day 61 : Created my first webapp using Flask to predict diabetes. Still work in progress, will try to make it look little classy.
Day 62 & 63 : Did nothing related to ML, but taught JAVA programming to one of my relative🙂
Day 64 : Watched @sirajraval 's video of How to Generate Art and How to Do Style Transfer with Tensorflow. Now reading a research paper - Neural Algorithm of Artistic Style
Day 65 : Wrote a program to generate art.
Day 66 : Worked on YOLO object detection using darkflow.
Day 67 : Built Sentiment Analyser using Tflearn - A step towards NLP.
Day 68 : Studied statistics
Day 69 : Wrote a code to detect faces with OpenCV and Deep learning. Reference @PyImageSearch tutorials
Day 70 : Revised basics of OpenCV - array slicing, resizing, rotating, smoothing, drawing on image, etc
Day 71 : Learned the fundamentals of image processing like edge detection, thresholding, erosion, dilation, masking and many more using OpenCV
Day 72 : Learned to track colored object in OpenCV. Referred @PyImageSearch tutorials.😀
Day 73 : Read a blog on - How OpenCV’s blobFromImage works.
Day 74, 75 : Using previously trained model of Digit Recognizer in keras implemented code to predict digits from the image with OpenCV.
Day 76, 77 : Real-time digit recognition using OpenCV.
Day 78 : Updated projects to github with detailed readme
Day 79 : Today I practiced SQL queries. Didn't do much related to ML, watched siraj's video on AI in China
Day 80 : Worked on data visualization and feature selection on Breast Cancer Dataset
Day 81 : Learned voilon plot, swarm plot and box plot plotting techniques.<br/ >
Day 82 : Read blogs on facial landmark detection using dlib library
Day 83, 84 : Started working on facial expression dataset
Day 85, 86 : Read various blogs on Data Augmentation and wrote a @Medium story 👉 Is deep learning useless if there is no large dataset available? NO!!!
Day 87 : Trained a model to predict sign language digits in Google Colab.
Day 88 : Having an issue - Error 104 when downloading the model from Google Colab🙄 Trying to solve it.
Day 89 : Loaded Keras saved model and trying to build a real time sign language detector
Day 90, 91 : Practiced discrete mathematics sets, relation and function concepts.
Day 92 : Started learning Web Scraping