utkarshtrehan/i2dl-tum

Deep Learning WS2021 🧠 | TUM | Prof. Matthias Nießner

I2DL-TUM

Part of Introduction to Deep Learning course offered in WS2021 by TUM

Introduction to Deep Learning (I2DL) (IN2346)

Contents

1. Introduction to IPython notebook Introduction to ipython notebooks and Numpy library.
2. CIFAR-10: Image Dataset: Data Preparation Prepare The CIFAR-10 dataset which consists of 50000 32x32 colour images in 10 classes, which are plane, car, bird, cat, deer, dog, frog, horse, ship, truck, which will be then used for training Deep Learning models.
3. CIFAR-10: Image Dataset: DataLoader We load small subsets of the dataset at a time, instead of having to load each sample separately. (In machine learning, the small subsets are referred to as mini-batches.)
4. Introduction to Google Colab Google provides a free cloud service based on Jupyter Notebooks that supports free CPU and GPU. It allows you to develop deep learning applications using popular libraries such as PyTorch, TensorFlow, Keras, and OpenCV (without installation). All these libraries are pre-installed on Google Colab along wilt Python
5. House Prices Data Exploration Here we explore datasets in which the features are not images, but CSV files with different features, either numeric or categorical. such type of dataset is being used as an introduction to regression problems, which are problems in which we try to predict a continous variable (eg: the price of a house), rather than a discrete one (eg: the label of an image).