Datasets
Standard Dataset
Image Form of the SECOM Dataset
- Citation Author(s):
-
JianweiZhao
- Submitted by:
- John Zhao
- Last updated:
- Fri, 03/13/2020 - 11:39
- DOI:
- 10.21227/jkvr-n656
- License:
736 Views- Categories:
0 ratings - Please login to submit your rating.
Abstract
The original dataset SECOM is obtained from the the UC Irvine Machine Learning Repository (https://archive.ics.uci.edu/ml/datasets/secom). Then, each
sample is transformed to an image, with each pixel representing a feature. Therefore, image processing mechanisms such as convolutionary neural networks can be utilized for classification.
Instructions:
In the folder ``Classify_CNN'', files SECOM.xxxx.csv represent samples, in each of which, the first row can be ignored, and the remaining data is a matrix of size 24*20. The values indicate pixel values.
The labels is in file ``SECOM.label.csv''.
The folder ``FIGs'' contains the visualization of all samples.
