readme.md

Naruto Eyes Classifier using Neural Networks

---

Getting Started :

---

Classification on Naruto character eye images with various CNN for analysing the Neural Network accuracy.

Classes :

---

• Sharingan
• Byakugan
• Others (Sage)

How to Train model with custom data?

---

1. Prepare Images in directory structure

- train
    |_sharingan
    |_byakugan
    |_Others

2. Train Command:

python train.py -d data/train/

3. Test :

python test.py -m modules/LENET_naruto_eye.h5 -i data/test/sage_draw.jpeg

Neural Networks :

---

I have used the basic neural networks for testing purpose and will add more.

LeNet :

---

Structure:

Training result:

Thanks for the References :

---

• https://towardsdatascience.com/the-4-convolutional-neural-network-models-that-can-classify-your-fashion-images-9fe7f3e5399d
• https://medium.com/@sidereal/cnns-architectures-lenet-alexnet-vgg-googlenet-resnet-and-more-666091488df5
• https://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3
• https://machinelearningmastery.com/use-pre-trained-vgg-model-classify-objects-photographs/
• https://www.pyimagesearch.com/2017/12/11/image-classification-with-keras-and-deep-learning/
• Naruto fonts by Cufonfonts
• TensorFlow.js
• Release Notes of Tensorflow.js
• And All open source resources ..!