This is 100 days of Machine Learning challenge as instructed by Siraj Raval #learningbydoing
Machine Learning is the most transformative technology of our time. Whether its helping us discover new drugs for major diseases, fighting fraud, generating music, improving supply chain efficiency, the list of applications are truly endless. In order for us as a community to be able to make valuable contributions to the world, we need to master this technology. This is a call to action, a battle cry, a spark that will light a movement to radically improve the state of humanity. 100 Days of ML Code is a commitment to better your understanding of this powerful tool by dedicating at least 1 hour of your time everyday to studying and/or coding machine learning for 100 days.
It's the daily log to keep track on my progress.
- Today I got the overview of Machine Learning Algorithms with the Mind-maps and Cheat sheets.
- Set up the environment(anaconda.org) to complete this challenge and also completed choosing the courses I will follow.
Link to Progress - Overview
- Learnt the basics of Linear Regression and revisited the Machine Learning Course by Andrew Ng in Coursera. Alternative free course - Stanford YT
- Implemented Linear Regression without using popular Libraries and Frameworks on the 'bike_sharing_data.csv'.
Link to Progress - Linear Regression with NumPy and Python
- Learnt the basic intuition of Logistic Regression and enrolled to the guided project on Logistic Logistic Regression.
- Implemented Logistic Regression using Python.
Link to Progress - Logistic Regression with Numpy
- Continued with Regression and Coursera ML Course.
- Tried Implementing Linear Regression with Siraj Raval's tutorial
Link to Progress - How to Do Linear Regression using Gradient Descent
- Enrolled into a free course offered by Google - Machine Learning Crash Course and completed upto First steps with Tensorflow.
Link to Progress - First Steps with TF: Programming Exercises
- Continued the ML crash curse by google
- Implemented Linear Regression with Real Dataset in Google Colab
Link to Progress - Linear regression with tf.keras
- Today I analyzed Covid-19 Dataset using python on real dataset.
- I took the data and completed the project with help of Rhyme Project Network
Link to Progress - COVID-19 Data Analysis
- Completed the ML course upto Validation Set
- Enrolled into Project centric course - Predicting House Prices with Regression using TensorFlow
Link to Progress - Housing Price Prediction
- Implemented Housing Price Prediction with Boston_housing.csv data.
Link to Progress - Housing Price Prediction
- Predicted Profit of Food Truck with Regression with the previous data given by assignment page.
- I Implemented Linear regression with single variable from scratch.
Link to Progress - Food Truck Profit Prediction
- Enrolled into the CS50's AI course(audit) to freshen up with HarvardX: CS50AI.
- Explored with the Source code of the Maze from first lecture, the code is provided in the link below.
Link to Progress - Maze
- Completed the quiz and the project part after the first lecture.
Link to Progress - projects/2020/x/degrees
- Today, I Implemented Flight Price Prediction after watching the live stream by Krish Naik
- Learnt more about the data pre-processing from YouTube.
Link to Progress - Flight_price
Link to Progress - Baseline: Data, ML, AI
- Learned How to use Scikit-learn implementing this Project: Predict Employee Turnover with scikit-learn.
Link to Progress - Predicting Employee Turnover with scikit-learn.
- Finished the following quest - Perform Foundational Data, ML, and AI Tasks in Google Cloud.
- Learned a lot new things and how to implemented them in gcp, it's a great hand's on learning platform.
Link to Progress - Data, ML, and AI Tasks in Google Cloud.
- Get my hand dirty with the Titanic Survival Data.
- Submitted my first kaggle submission.
Link to Progress - Titanic: Machine Learning from Disaster.
- Exploring the dataset from kaggle.
- continuing the competition with Krish Naik.
Link to Progress - Advance House Price Prediction.
- Implemented Multiple Linear Regression with scikit-learn in Coursera.
- Coursera Network platform is a great place to learn by praticing real-time with tutorials & datasets.
Link to Progress - Predicting-sales-with-multiple-linear-regression.
- Completed upto data preprocessing with krishnaik06.
- Kaggle Competition - House Prices: Advanced Regression Techniques.
Link to Progress - Advance House Price Prediction.
- Submitted yesterday's progress along with deployment.
- Edited the model with hyperparameter tuning.
Link to Progress - kaggle submission: Housing Prices Advanced Regression.
- Predicting Car Prices from Vehicle dataset from cardekho with @krishnaik06.
- Completed upto model deployment part, will update the front-end by tommorow.
Link to Progress - Car Price Prediction
- Learnt more of Feature Selection in depth from Coursera Project Network.
- Enrolled into the most recommended open sourced course and completed up-to Random Forest.
- Random Forest is widely applicable ml model, You can find it here: lec. 1.
- I choose to complete at least four projects by the end of 100_Days_of_ML_Code by all alternative 25'th day.
- As I previously worked on some Housing Price Predictions, starting with a similar data - Delhi Real-Estate Prices by MagicBricks.com.
Link to Progress - Delhi Housing Price Prediction.
- Deployed the machine learning model.pkl with Heroku.
- Fixed some issues and improved the accuracy a little bit.
Link to Progress - Delhi Real-estate Price Prediction.
- Learnt basic html and css as I've no prior experience with Front-end.
- CS50's web Programming course is a great resource to learn.
Link to Progress - CS50's Web Development.
- Enrolled into Google Machine Learning Specialization Course and completed upto modeule 3.
- Exploring Rest API's from Qwicklab.
Link to Progress - How Google does Machine Learning.
- Finished yesterday's course - How Google Does Machine Learning.
- Joined the live class by Krish Naik on hyperparameter-tuning with diabetes data.
Link to Progress - Hyper Parameter Tuning.
- Completed upto Decision Trees and re-visited archived courses to note-making for future self XD.
Link to Progress - Check Resources Column
- Taking notes of CS229: Machine Learning, this is great alternative of Coursera's Andrew Ng ml course.
- Continuing with the epic Fast.ai course and finished Launching into Machine Learning course.
Link to Progress - CS229: Machine Learning.
- Prediction of Diabetes Dataset on kaggle done with RandomForestRegressor.
- Follow Krish Naik's Live project videos to learn more - Diabetes Prediction using Machine Learning
- Submitted the first Problem Set of CS50AI.
- Started Intro to TensorFlow from Qwicklab and completed first two lab with basic operations.
- Brushing up python with NumPy and Panda.
- Continuing the GCP course on TensorFlow, the course is well structured but beginner friendly.
- Predicted the cab price data with Linear Regression and DNN, got rmse < 10.
- This one is the part of the course assignment, the notebook is available in the following link.
Link to Progress - training-data-analyst.
- Scaling up cab price model.py file using Cloud AI Platform on GCP.
- This is also part course assignment, check this repo.
Link to Progress - Scaling up ML using Cloud AI Platform.
- A decision tree is a flowchart-like structure in which each internal node represents a "test" on an attribute, each branch represents the outcome of the test, and each leaf node represents a class label.
Link to Progress - Classification Trees in Python, From Start To Finish.
- Started learning Deep Learning with the MIT Deep Learning Playlist available on YouTube.
- Completed the first task of assignment - 1, Deep learning basics.
Link to Progress - Boston Housing Price Prediction with FFNN.
Day 40 : August 3, 2020 | Planar data classification with a hidden layer
- Continuing the epic Deep Learning course thought by Andrew Ng.
- Finished the 3'rd week's assignment on Planar data classification with a hidden layer.
Link to Progress - Planar data classification with a hidden layer.
-
- In week 4 of deeplearning.ai course, built a Neural Network from scratch.
Link to Progress - Building Deep Neural Network.
- Completed the first assignment of MIT6.S191, thaught by Alexander Amini.
- The part - 1 describes basic operations in TensorFlow and the aumated differentiation.
Link to Progress - Intro to TensorFlow.
- Completed the second part of assignment-1 of Lex Fridman DL Lectures.
- It build the overview of how hand written digits can be recognised using CNN.
Link to Progress - Classification of MNIST Dreams with Convolutional Neural Networks.
Link to Progress -
Link to Progress -
Link to Progress -
- Reducing image noises with auto-encoders in TensorFlow.
- This rhyme interface developed the understanding of auto-encoders.
Link to Progress -
- Converted images to grayscale, performed normalization, then applied CNN to it.
- This rhyme course is one of the most useful as it's a great project of computer vision.
Link to Progress -
- Today I learnt k-means clustering and applied it to compress images.
- This Coursera Project is most useful so far.
Link to Progress -
- Detecting Smiles in Images with What-if Tool in qwiklabs.
Link to Progress - What-If Tool with Image Recognition Models.
- Build a Classification Tree, which uses continuous and categorical data from the UCI Machine Learning Repository to predict whether or not a patient has heart disease.
Link to Progress Classification Trees in Python, From Start To Finish.
- So, this is the first assignment of CNN course by deeplearning.ai, the lectures are very useful, also available on their YouTube channel.
- Implemented single Convolution layer from scratch.
Link to Progress - Convolutional Model: step by step.
- Completed the second assignment of CNN (deeplearning.ai).
Link to Progress - Convolutional model: application.
- Started the epic Fast.ai course-v4 - Deeplearning For Coders.
- I used google colab to follow the course, check the documentation if faced any error in setup.
Link to Progress -
- Got 84% accuracy on predicting Tomato leaf diseases.
- The tutorial is avaible on here if you wish to follow along.
Link to Progress -
- Predicted signs with Resnet50 in Keras, it's the optional assignment of deeplearning.ai cnn course.
- check out the repo to follow along with the Keras tutorial.
Link to Progress -
- Continuing with previous day, completed week 3 assignment, detecting cars with YOLO algorithm.
Link to Progress -
- Implemented face verification and recognition with Inception model.
- Though it's the second part of the week 4, completed it as the lecture part of the Neural style transfer isn't completed.
Link to Progress -
- Finished CNN course of deeplearning.ai with the final assignment. The assignments are great for the future projects as well.
Link to Progress -
Link to Progress -