👕 Fashion-MNIST is a dataset of Zalando's article images—consisting of a training set of 60,000 examples and a test set of 10,000 examples. Each example is a 28x28 grayscale image, associated with a label from 10 classes. it was intend for Fashion-MNIST to serve as a direct drop-in replacement for the original MNIST dataset for benchmarking machine learning algorithms. It shares the same image size and structure of training and testing splits.
- you can find the notebook -> code
- The report of what we have done -> report and model evaluation
- Representing the images in 3 ways
- Pixel features representation (28×28) 784 pixel with a column for the label
- HOG features representation 70 features
- PCA 40 features
- Mix dataset between HOG and PCA
- Building Logistic Regression on each dataset Model hypertuning using GridSearch , Cross validation
- Kmeans (removing the label from the subset of dataset contain only 5 classses)
- Visualising K-means using PCA to create 3 components can be plotted as a 3D clusters
- Building a simpple GUI -> GUI
- you need to download the model from model
💡 House prices is regression data set about: Predict sales prices and practice feature engineering, RFs of training set is 81 column and test set is 79 column .
Cleaning and Preparing the data
- Feature Selection
- Dealing with missing data
- Dealing with outliers
- Feature Encoding
- Model Building and Enhancing
- Linear Regression
- KNN Regressor
- you can find the notebook -> code
- The report of what we have done -> report and model evaluation