nn_fromScratch
所属分类:数值算法/人工智能
开发工具:Python
文件大小:0KB
下载次数:0
上传日期:2024-04-21 11:01:59
上 传 者:
sh-1993
说明: A multi-level perceptron neural network made from scratch, using NumPy and SkLearn. Implements backpropagation and utilized on the mnist dataset as a test case. Obtains acc > 97% test accuracy. *Some functions from BU ECE department and imported libraries.
文件列表:
mlp_nn.py
A multi-level perceptron neural network made from scratch, using NumPy and SkLearn. Implements backpropagation and utilized on the mnist dataset as a test case. Obtains acc > 97% test accuracy.
*Some functions from BU ECE department and imported libraries.
Test case on MNIST dataset:
![Untitled](https://github.com/corndog-overflow/nn_fromScratch/assets/104959725/30ea1b95-a75c-435f-8c78-2678e0c7b454)
Confusion matrix:
![image](https://github.com/corndog-overflow/nn_fromScratch/assets/104959725/c260b3cb-6902-4eab-906a-d36a5ae0f9a8)
近期下载者:
相关文件:
收藏者: