100 Days of ML Challenge

Day 1 : July 5, 2018

• Today's Progress: I picked up the Boston House Pricing dataset for implementing Linear Regression.
• Thoughts: I chose this dataset just as a warmup.
• Link to the code

Day 2 : July 6, 2018

• Today's Progress: After learning the basics of Keras, I implemented Sequential model on Breast Cancer Dataset available in scikit-learn.
• Thoughts: I am currently enrolled in a course, so this Keras lecture is a part of the course.
• Link to the code

Day 3 : July 7, 2018

• Today's Progress: I built a Convolutional Neural Network in Tensorflow to classify the images in MNIST dataset. This dataset has images of handwritten digits. The model runs at an accuracy of 98%.
• Link to the code

Day 4 : July 8, 2018

• Today's Progress: Improved the code done yesterday by building a Convolutional Neural Network in Keras to classify the images in MNIST dataset. The accuracy of the model is 99.4%.
• Link to the code

Day 5 : July 9, 2018

• Today's Progress: I used the Cifar10 dataset on the Convolutional Neural Network built using Keras. Cifar10 dataset has images which belong to one of the 10 classes like bird, truck, cat, frog, etc.
• Link to the code

Day 6 : July 10, 2018

• Today's Progress: I learnt RNN, which is a part of the course. So, I built a network containing SimpleRNN layer, in order to understand the whole working. The dataset used is International-Airline-Passengers.
• Link to the code

Day 7 : July 11, 2018

• Today's Progress: I implemented Naive Bayes classifier from scratch, and used it for predicting the classes in Iris Dataset available in Scikit-learn.
• Link to the code

Day 8 : July 12, 2018

• Today's Progress: I started learning KMeans clustering. On the Iris dataset, I implemented KMeans clustering from Scikit-learn, besided that I wrote another K_Means class.
• Link to the code

Day 9-10 : July 13, 14, 2018

• Progress in 2 days: I worked on Twitter US Airlines Sentiment Analysis project, which is a part of the coursework. The data preprocessing needed for this analyser was interesting. I used 3 different classifiers in order to see which one gives the best results.
• Link to the code

Day 11-12 : July 15-16, 2018

• Progress: I implemented Gradient Descent from scratch and used it for finding the result in Combined Cycle Power Plant Dataset. The dataset contains 9568 data points collected from a Combined Cycle Power Plant over 6 years (2006-2011), when the power plant was set to work with full load. Features consist of hourly average ambient variables Temperature (T), Ambient Pressure (AP), Relative Humidity (RH) and Exhaust Vacuum (V) to predict the net hourly electrical energy output (EP) of the plant.
• Link to the code

Day 13-14 : July 17-18, 2018

• Progress: I implemented Naive Bayes classifier on Iris Dataset. The code has been written from scratch.
• Link to the code