# MILDM-MNIST

**Repository Path**: inkiinki/mildm-mnist

## Basic Information

- **Project Name**: MILDM-MNIST
- **Description**: Multi-instance Learning with Discriminative Bag Mapping (2018MILDM)
Data set: Only using MNIST data set.
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2021-05-12
- **Last Updated**: 2021-05-12

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

Author：Inki
Contact：inki.yinji@qq.com
Parameters that can be modified:
1) po_label = 0 --> range: [0..9]
The number denotes that the main class is $c$-th class of MNIST.
2) mnist_path = "..."
The path of saved MNIST data.
3) loops = 5
The 5 times 10-cv, others you can set.
Data set description:
If you set po_label to $c \in [0..9]$, the $c$-th class will be set as the positive bags, others as the negative bags.
More parameters you could see original code.
Copyright: None.
Times: 20210512