GANS assignment
Budget: $10 – $30 USD
I am seeking a deep learning engineer to implement and analyze several state-of-the-art generative deep learning models on image data as part of an advanced computer vision assignment. A PDF of the assignment and data will be provided later.
You will be responsible for:
1. Developing and training Variational Autoencoders (VAEs), a Deep Convolutional Generative Adversarial Network (DCGAN), two improved GAN models of your choice, and Wasserstein GAN
2. Analyzing results via visualizations of latent spaces and generated images
3. Preparing a summary report comparing findings across all implemented models with graphs and tables
4. Providing source code and trained models upon completion
Requirements
The ideal candidate will have strong skills in deep learning frameworks like TensorFlow or PyTorch and experience implementing variational autoencoders and generative adversarial networks. Familiarity with computer vision techniques is preferred.
Guidelines:
- It’s ok to get the models’ code from internet, just reference the source code link, and add some functions for small plagiarism check
- No use of opencv in model core implementation, for analysis is fine
- The comments on the report don’t have to be lengthy or new discovery, just have to be correct and short
The task’s deadline is before 30 Jan 10:00 AM GMT
Thank you in advance. Let me know if you have any other questions.
You will be responsible for:
1. Developing and training Variational Autoencoders (VAEs), a Deep Convolutional Generative Adversarial Network (DCGAN), two improved GAN models of your choice, and Wasserstein GAN
2. Analyzing results via visualizations of latent spaces and generated images
3. Preparing a summary report comparing findings across all implemented models with graphs and tables
4. Providing source code and trained models upon completion
Requirements
The ideal candidate will have strong skills in deep learning frameworks like TensorFlow or PyTorch and experience implementing variational autoencoders and generative adversarial networks. Familiarity with computer vision techniques is preferred.
Guidelines:
- It’s ok to get the models’ code from internet, just reference the source code link, and add some functions for small plagiarism check
- No use of opencv in model core implementation, for analysis is fine
- The comments on the report don’t have to be lengthy or new discovery, just have to be correct and short
The task’s deadline is before 30 Jan 10:00 AM GMT
Thank you in advance. Let me know if you have any other questions.