Digit Recognizer is a Kaggle based competition. In this competition, goal is to correctly identify digits from a dataset of tens of thousands of handwritten images. Implemented different machine learning algorithms (Convolutional Neural Network, Random Forest, Support Vector Machine, Gradient Boosted Method) to improve the accuracy of model.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
DIGITaCC.PNG
LAYERVISULIZATION.PNG
LICENSE
MNIST-image-classification-using-TF.ipynb
MODELPREDICTION.PNG
README.md
png.png

README.md

Digit-Recognizer

Digit Recognizer is a Kaggle based competition. In this competition, goal is to correctly identify digits from a dataset of tens of thousands of handwritten images. Implemented different machine learning algorithms (Convolutional Neural Network, Random Forest, Support Vector Machine, Gradient Boosted Method) to improve the accuracy of model.

  • Accuracy Graph Accuracy
  • Prediction on Test data Prediction on Test dataset
  • Layer wise visualization Layer visulization
  • Model architecture Model Architecture