Citing from the Competition's Page: MNIST ("Modified National Institute of Standards and Technology") is the de facto “hello world” dataset of computer vision. Since its release in 1999, this classic dataset of handwritten images has served as the basis for benchmarking classification algorithms. As new machine learning techniques emerge, MNIST remains a reliable resource for researchers and learners alike.
In this competition, your goal is to correctly identify digits from a dataset of tens of thousands of handwritten images. We’ve curated a set of tutorial-style kernels which cover everything from regression to neural networks. We encourage you to experiment with different algorithms to learn first-hand what works well and how techniques compare.
This competition is the perfect for those who have some experience in Python and R but are new to Computer Vision since it introduces techniques like neural networks using a classic dataset including pre-extracted features or classification methods such as SVM and K-nearest neighbors.
Below is the description of each attached data files in this project. The files can be found under the Data folder in this repository.
- train.csv - the training set containing 42000 digit images of dimension 28x28
- test.csv - the test set containing 28000 digit images of dimension 28x28
We implemented a Convolutional Neural Network (CNN) which obtained an accuracy score of 0.988. Full implementation with all steps (EDA, Feature Engineering, Model Devlopment, and Evaluation) can be found under the Main Folder in this repository.