This is an open solution to the Santander Value Prediction Challenge :smiley:
Check collection of public projects :gift:, where you can find multiple Kaggle competitions with code, experiments and outputs.
We are building entirely open solution to this competition. Specifically:
LightGBM train and validation performance on folds :bar_chart: | LightGBM experiment logged values :bar_chart: |
---|---|
In this open source solution you will find references to the neptune.ml. It is free platform for community Users, which we use daily to keep track of our experiments. Please note that using neptune.ml is not necessary to proceed with this solution. You may run it as plain Python script :wink:.
link to code | name | CV | LB | link to the description |
---|---|---|---|---|
solution 1 | honey bee :honeybee: | 1.39 | 1.43 | LightGBM and 5fold CV |
solution 2 | beetle :beetle: | 1.60 | 1.77 | LightGBM on binarized dataset |
solution 3 | dromedary camel :dromedary_camel: | 1.35 | 1.41 | LightGBM with row aggregations |
solution 4 | whale :whale: | 1.3416 | 1.41 | LightGBM on dimension reduced dataset |
solution 5 | water buffalo :water_buffalo: | 1.336 | 1.39 | Exploring various dimension reduction techniques |
solution 6 | blowfish :blowfish: | 1.333 | 1.38 | bucketing row aggregations |
You can jump start your participation in the competition by using our starter pack. Installation instruction below will guide you through the setup.
:trident:
neptune run --config neptune_random_search.yaml main.py train_evaluate_predict --pipeline_name SOME_NAME
:snake:
python main.py -- train_evaluate_predict --pipeline_name SOME_NAME
git clone https://github.com/minerva-ml/open-solution-value-prediction.git
pip3 install -r requirements.txt
:trident:
neptune login
neptune run --config neptune_random_search.yaml main.py train_evaluate_predict --pipeline_name SOME_NAME
:snake:
python main.py -- train_evaluate_predict --pipeline_name SOME_NAME
experiment_directory
specified in the neptune.yamlYou are welcome to contribute your code and ideas to this open solution. To get started:
There are several ways to seek help: