🚗 == 🍋? 1 : 0
Link to competition: https://www.kaggle.com/c/DontGetKicked/
$sudo pip install -r requirements.txt
$python main.py
- Read 'train.csv' and 'test.csv' files
- Preprocess the files
- Feature Engineering found in preprocess.py
- Perform Locality Sensitive Hashing (LSH) on the dataframe for test.csv
- Finds 2000 similar rows for a query vector
- Perform K Nearest Neighbours on the LSH result
- Finds k=1000 similar neighbours
- Pipeline of classification learning algorithms: Neural Network -> SVM