AndyGrld/Digit_recognizer_model

Building a digit recognizer model

Getting started

pip install -r requirements.txt

digit_recognizer.py

Used KNeighbours Classifier, Support Vector Classifier and a Decision Tree Classifier to predict handwritten digits. Checked the accuracy of the three and saved the models as '.p' files using pickle library.

digit_recog_torch.py

Used a CNN in pytorch to build a model to predict handwritten digits.

All used data in datasets directory. All saved models stored in digit_recognizer_models directory.

Examples of digits