The purpose of this project is to gain more expernice and knowledge about the Data Science world and metholodigies.
In this project we used regular Machine Learning models such as Logistic Regression, Gaussian Naive Bayes, and KNN.
In addition we used an Ensemble Machine Learning models such as Random Forest, AdaBoost, XGBoost, Voting, Stacking, and Bagging.
We also used K-Means clustering and SVM classifier.
There are 4 notebooks to this project.
-
Rain Prediction - The first notebook's goal was to improve our classification notebook from last semester (Classification of whether it will be rainning tomorrow or not) by using new models and techniques we learned this semester.
-
Fashion MNIST - In the second notebook we created a dataset from clothes images and our goal was to find the best way to recognize for each image the right clothing. There are 10 types of clothing and our model should predict which number to give a new image of clothing (in range of 0-9).
-
Cats Vs Dogs - In the third notebook we created a dataset from cats and dogs images and we try to find the best way to recognize for each image the right answer. Our model will predict if the image is cat or a dog (cat = 1, dog = 2).
-
Kids Drawing - (Pandas) - In the fourth and last notebook we get drawing of kids and by understanding the data with the help of Pandas our goal was to answer the following questions:
a. How many different lines this drawing contains? b. What is the average pressure and the length(euclidean distance) of each line? c. How many times the kid who drew this doodle raised his hand during the doodling? d. Does this shape open or close?