Skip to content
/ DL-Model-Imitation Public
forked from jhopkins39/DL-Model-Recreation

With this project we research and test out whether data generated by generative Deep Learning Models can be effectively utilized to train secondary models. Specifically, if images generated by DALL-E mini are suitable to train a classification model with a performance comparable to models trained on CIFAR-10.

1 star 1 fork Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

andrea-covre/DL-Model-Imitation

BranchesTags
 
 

Repository files navigation

Deep Learning Model Inference

Goal

Through this project, we investigate whether Generative Deep Learning Models (such as DALL-E Mini) can generate data that is suitable to effectively train a secondary DL Classification Model with an accuracy comparable to classification models trained on real data, such as the CIFAR-10 dataset.

Research Paper

Research Report
See full research report here

Research Poster

Research Poster

Contributors

Andrea Covre - [email protected]
Jake Hopkins - [email protected]
Connor Reitz - [email protected]

About

With this project we research and test out whether data generated by generative Deep Learning Models can be effectively utilized to train secondary models. Specifically, if images generated by DALL-E mini are suitable to train a classification model with a performance comparable to models trained on CIFAR-10.

Topics

deep-learning analysis mining generative-model research-project georgia-tech

Resources

Readme
Activity

Stars

1 star

Watchers

0 watching

Forks

0 forks
Report repository

Languages

  • Python 100.0%