Skip to content

ramakrishnan-21/MNIST-Dataset

Repository files navigation

Digit Recognizer: Learn computer vision fundamentals with the famous MNIST data

Data Source

Data can be either dowloaded from my GitHub profile or directly download from kaggle

Description

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.

In this article my solution goes through the following steps

  • Defining the problem statement
  • collecting the data
  • preprocessing
  • cnn
  • ensembling predictions

Prerequisites

You need to have installed following softwares and libraries in your machine before running this project.

  • Python 3
  • Anaconda: It will install ipython notebook and most of the libraries which are needed like sklearn, pandas, seaborn, matplotlib, numpy, scipy.

Installing

Performance

I recently achieved an accuracy score of 0.99728 on the public leaderboard. I am ranked among the top 6% of all teams that are currently competing.

energy

Important Note

This model will take a lot of time to train. Pre-trained model can be downloaded from here.You can refer the notebook if you have any difficulites in loading the pre-trained model

Authors

Ramakrishnan - Complete Work