This repository holds the project of group JSON DERULO for the Applied Data Analysis EPFL cours.
Group members
- Adrián Fernández Lojo - adrian.fernandezlojo@epfl.ch
- Hanna Zwygart - hanna.zwygart@epfl.ch
- Jeremy Alexandre Serillon - jeremy.serillon@epfl.ch
- Mariem Maazoun - mariem.maazoun@epfl.ch
- Marina-Carmen Meredith Romero - marina.meredith@epfl.ch
.
├── README.md
├── data/
│ ├── IMPORTANT_README.md
│ ├── newyorker_caption_contest/
│ └── papers
└── src
├── Milestone_P2/
└── Milestone_P3/
The content of our milestones can be found as indicated above. The README for each milestone sits in their respective folders.
.
.
.
https://drive.google.com/file/d/1Es1XmglK5isHZuLiO2RDQfc1tYrDYzXz/view?usp=sharing
Zhang, J., Jain, L., Guo, Y.*, Chen, J., Zhou, K. L., Suresh, S., ... & Nowak, R. (2024). Humor in AI: Massive Scale Crowd-Sourced Preferences and Benchmarks for Cartoon Captioning. arXiv preprint arXiv:2406.10522.
- https://nextml.github.io/caption-contest-data/
- https://huggingface.co/datasets/jmhessel/newyorker_caption_contest
.
.
.
Download the dataset (see URL below), unzip/extract it, and place the resulting folder inside the data/ directory.
The folder name MUST remain exactly: newyorker_caption_contest
Do NOT rename this folder. If you change the name, the dataset may get added or saved in the GitHub repo (which is not ideal because the data set is a lot of storage) and code expecting data/newyorker_caption_contest will fail.
Your data/ folder should look like this:
data/
├── IMPORTANT_README.md
├── newyorker_caption_contest
│ ├── contests.json
│ ├── data/
│ └── images/
└── papers
├── Do Androids Laugh at Electric Sheep - Paper.pdf
└── Humor in AI - Paper.pdf
In VS Code the folder newyorker_caption_contest should be grayed out showing that git will not track it, like shown below:
