Duplicate images in a dataset with a general adversarial network

$10-30 USD

Fechado

Publicado

há mais de 5 anos

$10-30 USD

Pago na entrega

I want to derive new images from the images I have collected. There are 150 pictures in each class of 20 class data. I'm supposed to give GAN a more of a painting that looks like her. I want to see at what stage each picture is produced in epoch. It needs to be made with Python programming language. When we give a picture we need to create a new picture in the project. You should do a work on the CPU because I'm going to run the CPU on the [login to view URL] need to make a fake picture of ourselves. You can ask if there is something that is not understood.

Machine Learning (ML)

Python

ID do Projeto: 18247251

Sobre o projeto

1 proposta

Projeto remoto

Ativo há 5 anos

Quer ganhar algum dinheiro?

Endereço de e-mail

Benefícios de ofertar no Freelancer

Defina seu orçamento e seu prazo

Seja pago pelo seu trabalho

Descreva sua proposta

É grátis para se inscrever e fazer ofertas em trabalhos

1 freelancers estão ofertando em média $470 USD for esse trabalho

@DoctorMatthew

I'm a Data Scientist with a PhD in Physics. I'm based in Northern Ireland, but work remotely for a US Fortune 100 company. In my day-to-day, I work primarily on Deep Learning tasks in the NLP and Computer Vision space. I've recently completed a project on Freelancer with an Australian artist - developing a GAN to generate artwork. Whilst I hadn't built such a network before, my skills in Python and Keras (Tensorflow) were highly transferrable. The artist has now applied for gallery funding to further develop the project into an exhibition. I like to provide code in a Jupyter notebook, hosted on a free platform for developing neural networks in Python. The major benefits are the powerful computation power (which makes training models fast!) and the ability for the client to easily run the code without having to install anything. It's also free to use the platform! Matthew

$470 USD em 4 dias