English
Neu.ro Core
Neu.ro Toolbox
ADMINISTRATION
Hyperparameter Tuning with NNI
Introduction
This tutorial demonstrates how to use NNI (an open-source tool from Microsoft) for Hyperparameter Tuning on Neu.ro. You will create a new Neu.ro project, integrate it with NNI and run multiple tuning workers to speed up the search process.
Creating a Neu.ro project
To create a new Neu.ro project, run:
1
$ cookiecutter gh:neuro-inc/cookiecutter-neuro-project --checkout release
2
cd <project-id>
Copied!
Populating the Experiment Code and Integrating With Neu.ro
We're going to use this NNI example code with a MNIST dataset. Put the mnist.py file to the
modules folder and search_space.json to the config folder.Then, add the following lines to
requirements.txt:1
nni==2.0 # Required for Hyper-parameter search
2
neuro-sdk # Required by Neu.ro NNI integration script
3
Jinja2>=2.11.2 # Required by Neu.ro NNI integration script
Copied!
We are now ready to build our image:
1
$ neuro-flow build myimage
Copied!
While Docker builds our image, we can continue setting up the NNI integration.
Add the following lines at the end of
.neuro/live.yml:1
nni_worker:
2
image: $[[ images.myimage.ref ]]
3
life_span: 1d
4
multi: true
5
detach: true
6
volumes:
7
- $[[ volumes.code.ref_ro ]]
8
- $[[ volumes.config.ref_ro ]]
9
- $[[ volumes.project.ref ]]
10
env:
11
EXPOSE_SSH: "yes"
12
bash: |
13
sleep infinity
14
15
nni_master:
16
image: $[[ images.myimage.ref ]]
17
life_span: 1d
18
http_port: 8080
19
http_auth: false
20
pass_config: true
21
detach: true
22
browse: true
23
volumes:
24
- $[[ volumes.code.ref_ro ]]
25
- $[[ volumes.config.ref ]]
26
- $[[ volumes.project.ref ]]
27
bash: |
28
cd $[[ volumes.config.mount ]] && python3 prepare-nni-config.py
29
cd $[[ volumes.project.mount ]] && USER=root nnictl create --config $[[ volumes.config.mount ]]/nni-config.yml -f
Copied!
This will add some new jobs to work with NNI.
Finally, put
nni-config-template.yml and prepare-nni-config.py to the config folder and Makefile to the root folder of your project.Once
neuro-flow build myimage completes, your project is ready for running on Neu.ro.Running the Tuning Jobs
The only thing left is to run
1
make hypertrain
Copied!
This command will:
- Run 3 worker nodes. They can be configured via
N_JOBSin Makefile andpresetin.neuro/live.yamlparameters respectfully. - Run the master node with the
cpu-smallpreset. - Auto-generate a NNI configuration file for the master node pointing at the workers.
- Run the training process and automatically open the NNI web interface in your browser.
You can track experiment progress and intermediate results from this web UI. When the workers are done, you can get the final hyperparameter values and download the logs if needed.
NNI Hyperparameter Tuning GUI
Once you're done, you can shut down the workers and the master node by running
1
$ neuro-flow kill ALL
Copied!