Skip to content

Some fix to getting-started.md #94

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
May 25, 2018
Merged

Some fix to getting-started.md #94

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
698ed29
fix md
lluunn May 24, 2018
9da3e4c
fix
lluunn May 24, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 20 additions & 6 deletions docs/getting-start.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@
First, Copy CLI tool.

```bash
$ curl -Lo katib-cli https://github.com/kubeflow/katib/releases/download/v0.1.0-alpha/katib-cli-linux-amd64 && chmod +x katib-cli && sudo mv katib-cli /usr/local/bin/
$ curl -Lo katib-cli https://github.com/kubeflow/katib/releases/download/v0.1.1-alpha/katib-cli-linux-amd64 && chmod +x katib-cli && sudo mv katib-cli /usr/local/bin/
```

The cli tool will be put `/usr/local/bin/` directory.
Expand All @@ -26,6 +26,8 @@ $ ./scripts/deploy.sh

## Use CLI

vizier-core is a service of type NodePort, with port 30678.

Check which node the vizier-core was deployed.
Then access vizier API.

Expand All @@ -40,17 +42,29 @@ vizier-core-864dd6fdd4-r55qv 1/1 Running 0 11m
vizier-db-7b6f8c59bc-mjhh4 1/1 Running 0 11m 10.36.0.4 node1
vizier-suggestion-random-5895dc79b4-pbqkc 1/1 Running 0 11m 10.47.0.5 gpu-node3

$ katib-cli -s gpu-node2:30678 Getstudies
$ katib-cli -s gpu-node2:30678 get studies
2018/04/03 05:14:49 connecting gpu-node2:30678
StudyID Name Owner RunningTrial CompletedTrial
```

If your DNS cannot resolve the node name, connect it via IP. Get node's IP by

```
kubectl get -n katib node YOUR_NODE -o wide
```

If you are using GKE, create a firewall rule to allow traffic on port 30678.

```
gcloud compute firewall-rules create katibservice --allow tcp:30678
```

## Create Example Study

Try Createstudy. Study will be created and start hyperparameter search.

```bash
$ katib-cli -s gpu-node2:30678 -f ../examples/random.yml Createstudy
$ katib-cli -s gpu-node2:30678 -f ../examples/random.yml create study
2018/04/03 05:16:37 connecting gpu-node2:30678
2018/04/03 05:16:37 study conf{cifer10 root MAXIMIZE 0 configs:<name:"--lr" parameter_type:DOUBLE feasible:<max:"0.07" min:"0.03" > > configs:<name:"--lr-factor" parameter_type:DOUBLE feasible:<max:"0.2" min:"0.05" > > configs:<name:"--max-random-h" parameter_type:INT feasible:<max:"46" min:"26" > > configs:<name:"--max-random-l" parameter_type:INT feasible:<max:"75" min:"25" > > configs:<name:"--num-epochs" parameter_type:INT feasible:<max:"3" min:"3" > > [] random median [name:"SuggestionNum" value:"2" name:"MaxParallel" value:"2" ] [] Validation-accuracy [accuracy] mxnet/python:gpu [python /mxnet/example/image-classification/train_cifar10.py --batch-size=512 --gpus=0,1] 2 default-scheduler <nil> }
2018/04/03 05:16:37 req Createstudy
Expand All @@ -60,7 +74,7 @@ $ katib-cli -s gpu-node2:30678 -f ../examples/random.yml Createstudy
You can check the job is running with `kubectl` command.

```bash
$ katib-cli -s gpu-node2:30678 Getstudies
$ katib-cli -s gpu-node2:30678 get studies
2018/04/03 05:19:49 connecting gpu-node2:30678
StudyID Name Owner RunningTrial CompletedTrial
fef3711aa343fae6 cifer10 root 2 0
Expand All @@ -74,7 +88,7 @@ wbe8aabd6ad4f50e-worker-0 1 0 1m
Check the status of jobs with `katib-cli` command.

```bash
$ katib-cli -s gpu-node2:30678 Getstudies
$ katib-cli -s gpu-node2:30678 get studies
2018/04/03 05:26:20 connecting gpu-node2:30678
StudyID Name Owner RunningTrial CompletedTrial
fef3711aa343fae6 cifer10 root 1 1
Expand Down Expand Up @@ -215,7 +229,7 @@ parameterconfigs:
```

```bash
$ katib-cli -s gpu-node2:30678 -f ../examples/random-pv.yml Createstudy
$ katib-cli -s gpu-node2:30678 -f ../examples/random-pv.yml create study
2018/04/03 05:49:47 connecting gpu-node2:30678
2018/04/03 05:49:47 study conf{cifer10-pv-test root MAXIMIZE 0 configs:<name:"--lr" parameter_type:DOUBLE feasible:<max:"0.07" min:"0.03" > > configs:<name:"--lr-factor" parameter_type:DOUBLE feasible:<max:"0.2" min:"0.05" > > configs:<name:"--max-random-h" parameter_type:INT feasible:<max:"46" min:"26" > > configs:<name:"--max-random-l" parameter_type:INT feasible:<max:"75" min:"25" > > configs:<name:"--num-epochs" parameter_type:INT feasible:<max:"3" min:"3" > > [] random median [name:"SuggestionNum" value:"2" name:"MaxParallel" value:"2" ] [] Validation-accuracy [accuracy] mxnet/python:gpu [python /mxnet/example/image-classification/train_cifar10.py --batch-size=512 --gpus=0,1] 2 default-scheduler pvc:"nfs" path:"/nfs-mnt" }
2018/04/03 05:49:47 req Createstudy
Expand Down