A collection of Neural networks and NN models trained by me while I learn deep learning :)
This interactive notebook is a quick wakthrough of the ML workflow of loading up an image dataset and performing a multiclass classification. The dataset is the famous CIFAR_10 dataset having 10 classes. The model that I built is a simple custom model with couple of Convolution blocks and MaxPooling layers with a FC dense layer as the classifier head. THe dataset comes preprocessed from the TF Hub datasets but this is a good introduction notebook for beginners.
View the project:-here
Another multiclass classifier trained on the Fashion MNIST dataset with a deeper Neural network and more convolution blocks with an aim to understand if the depth of a model increases the accuracy of the classifier. This is a direct improvement over the Cifar model with evidently deeper architecture and more parameters.
Link to the project:-here
A Deep computer vision model with the ability to identify the age of an Indian and classify them into one among the three age groups- Young, Old, Middle. The dataset I used for this Project is The Indian Movie Face database (IMFDB) by Shankar setty et-al.[https://ieeexplore.ieee.org/document/6776225]
Link to the project folder:-here