MICaptioning

Project Set-up:

first download all required packages
- create python virtualenv for the project first
- then pip install the requirement.txt
Once packages are setup, run data_download.py, the evaluation code in it is not important, just run download lines. It might be helpful to download that embedding file as well.
```
from bioCaption.data.downloads import DownloadData
downloads = DownloadData()
downloads.download_iu_xray()
```
Once data are donwloaded, run build_dataset_folder.py to automatically create the train/valid/test folders and populate with images and captions. This is required for dataloader to work properly.