In this project, we will implement Domain-Adversarial Training of Neural Networks on SVHN → MNIST task. The DANN is a popular unsupervised Domain Adaptation paper which uses the principle of adversarial learning to align the source and target datasets. A classifier trained on the source data can now classify the target data effectively.
In this project we will treat the SVHN dataset of digits as the source domain and the MNIST dataset of digits as the target domain. Note that both the domains need to have the same categories.
1 freelance fait une offre moyenne de $140 pour ce travail
Hi, I just read your job posting and it sounds like you need a python expert. Over the last 5 years, I developed many python projects related to web scraping, machine learning, AI, data processing etc. Especially, I ha Plus