Wave Dashboard for the OpenML AutoML Benchmark
Dashboard to compare AutoML frameworks by exploring the results from runs of the OpenML AutoML Benchmark.
App Goal: View & compare performance of different AutoML frameworks, as benchmarked on the OpenML AutoML Benchmark
Target Audience: Data scientists
Industry: Any
GitHub: https://github.com/h2oai/wave-amlb
Actively Being Maintained: Yes
Comes with Demo Mode (pre-loaded data, models, results, etc.): Yes
Allows for Uploading and Using New Data: Yes
Follow the instructions here to download and run the latest Wave Server, a requirement for apps.
Go to download the wave SDK for your platform
wave-0.12.0-darwin-amd64.tar.gz
Extract your download:
tar -xzf wave-0.12.0-darwin-amd64.tar.gz
Move it to a convenient location:
mv wave-0.12.0-darwin-amd64 <pathname>/wave
check your $home/wave directory, the output should look like the following below:
.
├── demo
├── examples
├── readme.txt
├── test
├── waved
└── www
cd wave
./waved
When runninng the command, you may get an issue about “cannot open file because identity of developer can’t be verified:
- go to system preferences
- security & privacy
- general
- you'll see a message about wave. select 'allow anyway'
- attempt to run ./waved again
If step 6 worked you should be able to go to the following link:
$ git clone https://github.com/h2oai/wave-amlb
$ cd wave-amlb
$ make setup
$ source venv/bin/activate
After you set-up the python environment you can start the app by running
Note! this is in a separate terminal tab then where you ran ‘./waved’ command to start the wave server:
wave run src.app
Note! If you did not activate your virtual environment this will be:
./venv/bin/wave run src.app
Point your web browser to localhost:10101