GitHub - InkiInki/MILDM-MNIST: Multi-instance Learning with Discriminative Bag Mapping (2018MILDM) Data set: Only using MNIST data set.

InkiInki / MILDM-MNIST Public

Notifications You must be signed in to change notification settings
Fork 1
Star 1

Multi-instance Learning with Discriminative Bag Mapping (2018MILDM) Data set: Only using MNIST data set.

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
ClassifyTool.py		ClassifyTool.py
FunctionTool.py		FunctionTool.py
I2B.py		I2B.py
I2I.py		I2I.py
MILDM.py		MILDM.py
Main.py		Main.py
MnistLoadTool.py		MnistLoadTool.py
Prototype.py		Prototype.py
README.txt		README.txt

Repository files navigation

Author：Inki
Contact：inki.yinji@qq.com
Blog: https://blog.csdn.net/weixin_44575152/article/details/108139662
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