Image classifier application to classify flowers to 102 categories, using TnensorFlow hub and Conv2D

A repository for machine learning problems and exploration of different ML libraries. The goal of this repository is to collect takeaways while developing ML models. This should improve my overall understanding of developing machine learning applications.

Handwriting digit recognition using keras.Conv2D and MNIST database.

Creating a classifier for the German Traffic Signs dataset that classifies images of traffic signs into 43 classes.

This project is to apply Convolutional Neural Networks (CNN) to recognize dog breeds.

Image classification based computer vision model CNN

This repository Investigates DCGAN using facedata. Serves as a personal cautionary tale when working with GANS.

Basic_CNN_Implementation

Convolutional Neural Network to Classify Dogs and Cat. I built a ImageClassifier which classifies and tells you whether its a Dog image or a Cat image. I built a convolutional network which consists of Three Convolution layer and Three MaxPooling layer. Each Convolutional layer has filters, kernel size. Maxpooling layer has stride and pooling si…

Deployed the super-resolution convolution neural network (SRCNN) using Keras. Recovers a high-resolution image from a low-resolution input.

layers

🐱 A deep learning model using CNN to classify between cat and dog images

This project aims to develop an advanced DL model using CNN to accurately detect and classify brain tumors from MRI scans.

Keras Convolutianl Networks

CNN

Reinforcement Learning with Actor-Critic to play Breakout-v4 (Atari) from OpenAI Gym

This model helps us classify 10 different real-life objects by undergoing training under tensorflow's CIFAR dataset which contains 60,000 32x32 color images with 6000 images of each class. I have made use of a stack of Conv2D and MaxPooling2D layers followed by a few densely connected layers.

In this project, used TensorFlow to build a neural network to classify road signs based on an image of those signs.