Kaggle challenge.
We tackled the Kaggle challenge 'Digit Recognizer'. It consists of predicting with Machine Learning Regression models numbers that are shown in images.
The dataset was provided by Kaggle and was consisted in 42000 pixelated (28,28) images of hand drawn numbers.
- Dataset exploration
- Normalizing
- Modeling
- Final testing
- Submiting
- Either clone the repository or download the files
- Install requirements (requirements.txt.txt)
- Download the dataset from Kaggle
- Open the notebook: ML-Digit-Recognizer/main.ipynb
- Run the notebook
- Pandas: Dataset manipulation
- Numpy: Array manipulation
- Scipy, Scikitlearn: Train and test set
- Tensorflow, Keras: Modeling