AbelNN: Deep Learning Python module from scratch

I have implemented several neural networks from scratch using only Numpy. I have designed the module to be extremely easy to use even for people who have never used deep learning. However, if you have more experience using neural networks, you can also use my framework by fine-tuning and fully customizing the classes yourself from the constructor itself directly. In addition, if you want you can even call the internal procedures directly or even modify the source code.

You can consult all the information on how to use my library, whether you are an inexperienced user or not, in the documentation. You will also find detailed information on all classes, methods, procedures, and variables.

Implementation Class Code Examples Documentation
Multilayer perceptron MLP_Abel Click here Click here Click here
Convolutional neural network ConvNetAbel Click here Click here Click here

All algorithm classes are independent and you can directly import each file separately. However, you can also import the AbelNN.py file which imports everything.

GitHub

View Github