Portfolio

Create a classifier that can help diagnose breast cancer.

Train a model to predict whether or not a mushroom is poisonous.

Train multiple models and evaluate how effectively they predict credit card transactions fraud.

Predict whether a given blight ticket will be paid on time.

Train multiple classification models to predict if a message is spam or not.

Analyze real-world datasets from the city of Chicago using SQLite and Python.

Analyze teaching ratings of professors with different characteristics and see if there are external influences on the teaching evaluation score.

Train a Deep Neural Network and a Convolutional Neural Network for cat recognition.

Train a Convolutional Neural Network and a Residual Neural Network to classify a collection of six finger signs representing numbers from 0 to 5.

Implement transfer learning using a pretrained Convolutional Neural Network to build an alpaca classifier.

Implement object detection for autonomous driving using the "You Only Look Once" algorithm.

Build a U-Net to predict a label for every pixel in an image from an autonomous driving dataset.

Build a facial recognition system using a pre-trained FaceNet model.

Implement a Neural Style Transfer model to generate novel artistic images.

Implement a Shakespeare poem generator using a Long Short-Term Memory network.

Train a Long Short-Term Memory network to generate music.

Build a Long Short-Term Memory model that takes word embeddings as input to predict the most appropriate emoji.

Build a Neural Machine Translation model to translate human-readable dates into machine-readable dates.

Implement an algorithm for trigger word detection.

Implement Simple and Multiple Linear Regression to predict CO2 emissions from light-duty vehicles.

Use k-Nearest Neighbors to classify customers into group memberships for a telecommunications provider.

Build a Decision Tree classifier to find out which drug might be appropriate for a future patient with the same illness.

Build a model that predicts whether a credit card transaction is fraudulent or not on an imbalanced dataset.

Build a Decision Tree regressor that can predict the median price of houses for various areas of Boston.

Train a regression model to predict the amount of a taxi tip.

Train a Logistic Regression classifier to predict customer churn for a telecom company.

Build and train a model to classify human cells samples as benign or malignant.

Train a model using the classic iris dataset for multi-class classification.

Use k-Means Clustering for customer segmentation.

Train multiple classification models to predict the weather forecast in Australia.

Build a regression model using Keras to predict the compressive strength of concrete.

Build a Deep Neural Network and a Convolutional Neural Network to classify handwritten digits.

Build a Deep Neural Network and a Convolutional Neural Network using Pytorch to classify fashion products.

Implement pedestrian detection using a pretrained Faster R-CNN model.

Create a Recurrent Neural Network focused on Language Modelling and reach low levels of perplexity on the Penn Treebank dataset.

Reconstruct an image using a Restricted Boltzmann Machine.

Create a simple autoencoder for feature extraction, dimensionality reduction and reconstruction of images.

Leverage pre-trained models to build image classifiers to detect concrete with and without cracks.