Urias-T / Digit_Recognizer

Building neural network models that can accurately classify hand-written digits.

Computer Vision (Image Recognition) On MNIST Digit Dataset

Digit Recognizer

Data Source: Kaggle

Description: The Digit Recognizer competition is for participants to build a supervised learning model that can recognize grayscale images of handwritten numbers ranging from 0 - 9.

Results:

• Model1: Model1 attained an accuracy of ~99.16% on the test data.

Possible Area for Improvement:

To improve my model's performance, I could try a different model architecture.