Content

Wisconsin Diagnostic Breast Cancer

Reference to code WDBC


The dataset is found on UCI.
Number of instances 569, features 30. The sample of only one record is shown below. 
842302,M,17.99,10.38,122.8,1001,0.1184,0.2776,
0.3001,0.1471,0.2419,0.07871,1.095,0.9053,8.589,153.4,0.006399,0.04904,
0.05373,0.01587,0.03003,0.006193,25.38,17.33,184.6,2019,0.1622,0.6656,0.7119,
0.2654,0.4601,0.1189
The training block in code is 62% of all records, the rest is validation, accuracy in percent $96.7 \mp 1.4$ which is the same as other people report. Execution takes near $0.002$ seconds. The performance is hard to compare, because I found only Python implementations. This result can be compared to this recent preprint. They report training time 0.06 for MLP and 0.09 for KAN.

I have to note that KAN implementation in preprint uses improved Broyden method and mine Kaczmarz, explained in KAN's Core.