
Custom Object Detection using TensorFlow
A custom object detection model trained on a small image dataset of my back then College ID card as well as Spectacles images on TensorFlow, which could recognise objects with accuracy of about 80 percent.
Passionately courious about learning new technology. Data Science, Machine Learning, Internet of Things(IoT)
Passionately curious about learning new technologies
Background in Android application development and will be graduating in Big Data Solution Architecture from a Canadian institution in August 2020, with a GPA of 3.58 overall. Experience building complex mobile applications using Java, JSON, XML, and PHP API to fetch data from MySQL database as well as Android in-built database SQLite.
Very passionately curious about learning new technologies, having hands-on learning experience building a custom object detection model using TensorFlow 2.0.0 and implementation of various machine learning models based on machine learning algorithms to solve regression, classification, and clustering(segmentation) problems.
Thanking you for your precious time.
Thank you
Tapan Pandya.
As a passionate individual, I take a approach of breaking down bigger problem into small chunks of problems, which helps me understand the problem statement in-depth.
Below are some of my abilities and hobbies including soft skills as well as hard skills.
Throughout my Data Science, I would like to mention some of my successful endeavors, which includes Custom Object Detection using TensorFlow, the very first work I did during my first semester in COVID-19. Apart from that, my hands on learning experience on implementation of Machine Learning Algorithms and NLP.
A custom object detection model trained on a small image dataset of my back then College ID card as well as Spectacles images on TensorFlow, which could recognise objects with accuracy of about 80 percent.
A spam SMS classifier web application, trained on dataset from UCI Machine Learning repository. Python NLTK library is used to train the machine learning model on Naive-Bayes Multinomial algorithm and flask environment has been used to implement web application while Heroku Platform and Github collectively used for hosting the web application.
A breast cancer detection classifier web application, machine learning model trained on dataset taken from Kaggle. The web application devloped on flask environment and deployed on Heroku platform.
A car price prediction web application, regression machine learning model trained on Linear Regression algorithm while dataset is taken from Kaggle. The web application devloped on flask environment and deployed on Heroku platform.