Jobs are reproducible, run to completion instances that can be run in parallel.

You can use jobs to train your models with different parameters at the same time, compare results and iterate quickly.

| |
                pip install --no-binary :all: --user lightgbm && pip install --user lightgbm --install-option=--gpu && python non-blending-lgbm/ --debug=0 --learning_rate=.08 --min_child_samples=180 --max_bin=255 --nthread=7 --num_leaves=7 --max_depth=-1
(Spot) GPU: 1 (Tesla K80), CPU: 4, RAM: 61GB
Python 3, TensorFlow 1.6.0, Jupyter 5.3.0
April 29th 2018, 4:53:05 am
April 29th 2018, 5:04:14 am
April 29th 2018, 11:33:57 am