https://github.com/innixma/autogluon-benchmark

AutoGluon Benchmarking Repository
https://github.com/innixma/autogluon-benchmark

Last synced: about 1 month ago
JSON representation

AutoGluon Benchmarking Repository

• Host: GitHub
• URL: https://github.com/innixma/autogluon-benchmark
• Owner: Innixma
• License: apache-2.0
• Created: 2019-12-13T01:18:08.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2024-11-15T00:50:48.000Z (about 2 months ago)
• Last Synced: 2024-12-02T22:34:46.966Z (about 1 month ago)
• Language: Python
• Homepage:
• Size: 5.93 MB
• Stars: 2
• Watchers: 3
• Forks: 10
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
Code for benchmarking AutoGluon.

## Development Installation

To get started, run the following commands:

```

# Do this if you are locally developing AutoGluon to avoid installing it from pip:

git clone https://github.com/autogluon/autogluon

cd autogluon

./full_install.sh

cd ..

```

Install AutoGluon Bench if running evaluation code.

```

git clone https://github.com/autogluon/autogluon-bench.git

pip install -e autogluon-bench

```

Install AutoGluon Benchmark

```

# Install autogluon-benchmark

git clone https://github.com/Innixma/autogluon-benchmark.git

pip install -e autogluon-benchmark

```

## Full AutoMLBenchmark

To run AutoMLBenchmark, see instructions in:

`examples/automlbenchmark/README_automlbenchmark.md`

Please note that these instructions are quite technical.

## Local Testing

To benchmark individual OpenML datasets, you can check out the examples in `examples/train_***`

The scripts are fairly primitive. They mimic what AutoMLBenchmark does contained to a single dataset.

## Generating Task Metadata

To generate task metadata files, refer to autogluon_benchmark/data/metadata/README.md