TensorFlow 1.4 includes a Gradient Boosting implementation, aptly named TensorFlow Boosted Trees (TFBT). This repo contains the benchmarking code that I used to compare it XGBoost.
For more background, have a look at the article.
# Prepare the python environment
mkvirtualenv env
source env/bin/activate
pip install -r requirements.txt
# Download the dataset
wget http://stat-computing.org/dataexpo/2009/{2006,2007}.csv.bz2
bunzip2 {2006,2007}.csv.bz2
# Prepare the dataset
python preprocess_data.py
Train and run xgboost:
python do_xgboost.py
Train and run TensorFlow:
python do_tensorflow.py
Draw nice plots:
python analyze_results.py
./do_xgboost.py --num_trees=50 42.06s user 1.82s system 1727% cpu 2.540 total
./do_tensorflow.py --num_trees=50 --examples_per_layer=1000 124.12s user 27.50s system 374% cpu 40.456 total
./do_tensorflow.py --num_trees=50 --examples_per_layer=5000 659.74s user 188.80s system 356% cpu 3:58.30 total