Skip to content

Latest commit

 

History

History
41 lines (29 loc) · 1.26 KB

README.md

File metadata and controls

41 lines (29 loc) · 1.26 KB
Raw

CANN - C based Artificial Neural Network

This a simple library which integrates a lot of functionalities for basic ANN operations. For more details on the library, read the short paper summary - Paper

Comparison with python modules

1. Gate functionalities : Creating ANN for gates like XOR,AND,OR etc in a low level language like C is much faster and hiighly compatible too.

2. Regression : The Auto-Mpg dataset was used to solve a MULTIPLE LINEAR REGRESSION problem using the ANN.

Execution Time
ANN library Iterations RMSE Time(sec)
CANN 103 9.586 0.06558
Keras 103 5.516 13.9793

3.Classification : The IRIS dataset was used to solve a 3 class problem using the ANN.

Execution Time
ANN library Iterations Accuracy Time(sec)
CANN 103 96.7% 0.1834
Keras 103 96.6% 33.72608

Steps to compile and Execute

Compile

gcc -w cann.c <test_main_file.c> -lm

<test_main_file.c> can be :

  1. test_xor.c
  2. test_iris.c
  3. test_mpg.c

Corresponding python implementations are:

  1. iris_test.py
  2. mpg_test.py
Execute

./a.out