This repository contains two sample implementations of the traditional Computer Vision task, Digit Recognizer from Kaggle. The URL of the dataset and relevant information can be found here: https://www.kaggle.com/c/digit-recognizer
The instructions to run the Notebooks are as follows:
- Download the train.csv and test.csv files from the Kaggle URL given above and put the two files in a folder named "data".
- Then run the "digitRecoDL_kaggle.ipynb" and "digitRecoML_kaggle.ipynb" to check the results produced by the Deep Learning-based CNN and ML (Random Forest) models.
- Running the programs will save the prediction results in the folder named "result".
Overall, the root directory should have a hierarchy as follows:
- data -- train.csv -- test.csv
- results
- digitRecoDL_kaggle.ipynb
- digitRecoML_kaggle.ipynb