vikramnagashoka / deep-learning-specialization-projects

This repo has some of the projects from the Deep learning specialization that I completed in late 2020.

Deep-learning-specialization-projects

This repo has some of the projects from the Deep learning specialization that I completed in late 2020.

1 Neural machine translation using LSTM

Here I have built a Neural Machine Translation (NMT) model to translate human-readable dates ("25th of June, 2009") into machine-readable dates ("2009-06-25") using the attention sequence to sequence model.

2 Trigger word detection using RNN:

A unidirectional RNN model built using GRU layers that recognizes a trained word such as “Activate” from a given sound clip. A model that can be used in a voice assistant.

3 Neural Style Transfer

In this project I use the Gatys, et al algorithm to transfer styles from some famous paintings such as that of Picasso to real world images.

4 Face recognition and Face verification using deep learning.

Here I use ideas from two popular models FaceNet and Deepface to build two models, one for face verification trained on a public faces dataset. Another one for face recognition. Some of the tasks in the project include.

-> Implement the triplet loss function.

-> Use a pretrained model to map face images into 128-dimensional encodings.

-> Use these encodings to perform face verification and face recognition.

5 Object detection using YOLO network.

Here I build the original YOLO network to detect cars in a given image.

-> Use object detection on a car detection dataset

-> Deal with bounding boxes