/30DaysofDataScience

My journey throughout the 100daysofdata challenge. I share daily learning, projects, resources for learning, and daily blogs.

Primary LanguageJupyter Notebook

Journey of 100DaysOfData Challenge

Books and Resources
1. DeepLearning Specialization
2. Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython
3. Hands On Machine Learning with Scikit Learn, Keras and TensorFlow
4. DeepLearning.AI TensorFlow Developer Professional Certificate

Day1 of 100DaysOfData!

  • On my journey of Machine learning and Deep Learning, Today I continued reading Chapter End-to-End Machine learning Project. After framing the problem, getting the data, exploring the data, sampling data into train and test set, and creating a whole transformation pipeline to clean up and prepare data for Machine Learning algorithms automatically, today I learned to select and train a Machine Learning Model using Scikit Learn. I am working on a District house price prediction model. It is a multiple(more than one feature/attribute), univariate(only one value to predict) regression(based on all the attributes we have to predict a single value) supervised(labeled data) Machine Learning System.
  • I learned to select and train a Machine Learning Model using Scikit Learn, learned about various performance metrics, and learned little about the Linear Regression model, Decision Tree Regressor, and Random Forest Regressor model. I also came across Cross-validation and using it easily from Scikit Learn.
  • Hands On Machine Learning with Scikit Learn, Keras and TensorFlow | (Book)
  • Day 1 of #100DaysOfData - Personal Blog | (Blog)

Day2 of 100DaysOfData!

  • On my journey of Machine Learning and Deep Learning, even Today I continued reading Chapter End-to-End Machine learning Project. After framing the problem, getting the data, exploring the data, sampling data into train and test set, creating a whole transformation pipeline to clean up and prepare data for Machine Learning algorithms automatically, and selecting and training a model, today I learned to fine-tune my selected model.
  • Here I have learned about Grid Search, Randomized Search, Ensemble methods, Analyze the best models, their error, and figuring out the importance of each attribute for making an accurate prediction, and finally to evaluate my system on the test set for production. I also took some time out to look into Artificial Neural Networks.

Day3 of 100DaysOfData!

  • On my journey of Machine Learning and Deep Learning, Today I started chapter 'Training Models' of Hands-on machine learning with scikit-learn and TensorFlow Book.I learned about the fundamentals of the Linear Regression model, cost function, Normal equation, the math behind scikit learn's Linear-regression model, Computational Complexity of Normal Equation.

Day4 of 100DaysOfData!

  • On my journey of Machine Learning and Deep Learning, Today I continued reading from chapter 'Training Models' of Hands-on machine learning with scikit-learn and TensorFlow Book. I learned and implemented Gradient Descent and its types: Batch Gradient Descent, Stochastic Gradient Descent, and Mini-batch Gradient Descent. I also learned about Polynomial Regression to fit non-linear data with a linear model and learning curves.
    • Hands On Machine Learning with Scikit Learn, Keras and TensorFlow | (Book)

Day5 of 100DaysOfData!

  • On my journey of Machine Learning and Deep Learning, Today I continued reading from chapter 'Training Models' of Hands-on machine learning with scikit-learn and TensorFlow Book. I started my day by reading about the Bias-Variance trade-off. Followed by that, I learned Regularization, Regularized linear models, different regularization methods using: Ridge Regression, Lasso Regression, Elastic Net, and Early Stopping. I also compared all those models and learned how they help to overcome overfitting.