When importing a dataset to mASKerAId the file structure of the training images must follow a certain format.
Below is an example directory structure that will allow mASKerAId to properly enroll all the training images to the MongoDB. To enroll these images into mASKerAId you simply just need to click on and select the example rootDirectory shown below.

rootDirectory   
│
└───person1
│   └───frontFace.png
│   └───leftSide.png 
|   └───rightSide.png
│   
└───person2
|   └───face.png
|
└───person3
|   └───normalFace.png
|   └───happyFace.png
|   └───sadFace.png
|   └───madFace.png
|
...

Note that the content of each person sub directory does not matter. For better training results having more images related to a single person will allow for better feature extraction, but multiple images per person is not required.