Skip to content
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

Updating minikube get started for newest version of the wrappers #41

Merged
merged 1 commit into from
Jan 17, 2018
Merged

Updating minikube get started for newest version of the wrappers #41

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
43 changes: 22 additions & 21 deletions docs/getting_started/minikube.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
# Getting started on Minikube


In this guide, we will show how to create, deploy and serve a iris classification model using seldon-core running on a Minikube cluster. Seldon-core uses [helm](https://github.com/kubernetes/helm) charts to start and runs on [kubernetes](https://kubernetes.io/) clusters. Minikube is a tool that makes it easy to run Kubernetes locally, and runs a single-node Kubernetes cluster inside a VM on your laptop.
In this guide, we will show how to create, deploy and serve an Iris classification model using Seldon Core running on a Minikube cluster. Seldon Core uses [helm](https://github.com/kubernetes/helm) charts to start and runs on [kubernetes](https://kubernetes.io/) clusters. Minikube is a tool that makes it easy to run Kubernetes locally, and runs a single-node Kubernetes cluster inside a virtual machine on your laptop.


### Prerequisites
Expand All @@ -13,7 +13,7 @@ The following packages need to be installed on your machine.
* [helm](https://github.com/kubernetes/helm/blob/master/docs/install.md) >= 2.7.0

* [sklearn](http://scikit-learn.org/stable/)
- Sklearn is needed only to train the iris classifier example below. Seldon-core doesn't not require sklearn installed on your machine to run.
- Sklearn is needed only to train the iris classifier example below. Seldon Core doesn't not require sklearn installed on your machine to run.


### Before starting: run a Minikube cluster locally
Expand All @@ -30,38 +30,38 @@ Once the cluster is created, Minikube should automatically point your kubectl cl

### Start seldon-core

You can now start seldon-core in your minikube cluster.
You can now start Seldon Core in your minikube cluster.


1. Seldon-core uses helm charts to start. To use seldon-core, you need [helm installed](https://github.com/kubernetes/helm/blob/master/docs/install.md) in your machine. To Initialize helm, type on command line:
1. Seldon Core uses helm charts to start. To use Seldon Core, you need [helm installed](https://github.com/kubernetes/helm/blob/master/docs/install.md) in your machine. To Initialize helm, type on command line:

```bash
helm init
```

1. Seldon-core uses helm charts to start which are stored in google storage. To start seldon-core using helm:
1. Seldon Core uses helm charts to start which are stored in google storage. To start Seldon Core using helm:

```bash
helm install seldon-core --name seldon-core \
--repo https://storage.googleapis.com/seldon-charts
```

Seldon-core should now be running on your cluster. You can verify if all the pods are up and running typing on command line ```helm status seldon-core``` or ```kubectl get pods```
Seldon Core should now be running on your cluster. You can verify if all the pods are up and running typing on command line ```helm status seldon-core``` or ```kubectl get pods```

### Wrap your model

In this session, we show how to wrap the sklearn iris classifier in the [seldon-core-example](https://github.com/SeldonIO/seldon-core-examples) repository using seldon-core python wrappers. The example consist of a logistic regression model trained on the [iris dataset](http://scikit-learn.org/stable/auto_examples/datasets/plot_iris_dataset.html).
In this session, we show how to wrap the sklearn iris classifier in the [seldon-core](https://github.com/SeldonIO/seldon-core) repository using Seldon Core python wrappers. The example consists of a logistic regression model trained on the [Iris dataset](http://scikit-learn.org/stable/auto_examples/datasets/plot_iris_dataset.html).

1. Clone seldon-core-examples repository:
1. Clone the seldon-core repository:

```bash
git clone https://github.com/SeldonIO/seldon-core-examples
git clone https://github.com/SeldonIO/seldon-core
```

2. Train and save the sklearn iris classifier example model using the provided script ```train_iris.py```:

```bash
cd seldon-core-examples/models/sklearn_iris/
cd seldon-core/examples/models/sklearn_iris/
```
```bash
python train_iris.py
Expand All @@ -70,20 +70,22 @@ In this session, we show how to wrap the sklearn iris classifier in the [seldon-
This will train a simple logistic regression model on the iris dataset and save the model in the same folder.


3. Wrap your saved model using the bash script wrap-model-in-minikube in ```seldon-core-examples```::
3. Wrap your saved model using the core-python-wrapper docker image:
```bash
cd ../../
docker run -v $(pwd):/model seldonio/core-python-wrapper:0.4 /model IrisClassifier 0.1 seldonio --force
```

4. Build the docker image locally
```bash
./wrap-model-in-minikube models/sklearn_iris IrisClassifier 0.1 seldonio --force
cd build
./build_image.sh
```

This will create the docker image ```seldonio/irisclassifier:0.1``` inside the minikube cluster which is ready for deployment with seldon-core.
This will create the docker image ```seldonio/irisclassifier:0.1``` inside the minikube cluster which is ready for deployment with Seldon Core.


### Deploy your model

The docker image version of your model is deployed through a json configuration file. A general template for the configuration can be found [here](https://github.com/SeldonIO/seldon-core-examples/blob/master/models/sklearn_iris/sklearn_iris_deployment.json). For the sklearn iris example, we have already created a deployment file ```sklearn_iris_deployment.json``` in the ```sklearn_iris``` folder:
The docker image version of your model is deployed through a json configuration file. A general template for the configuration can be found [here](https://github.com/SeldonIO/seldon-core/blob/master/examples/models/sklearn_iris/sklearn_iris_deployment.json). For the sklearn iris example, we have already created a deployment file ```sklearn_iris_deployment.json``` in the ```sklearn_iris``` folder:

```json
{
Expand Down Expand Up @@ -128,7 +130,6 @@ The docker image version of your model is deployed through a json configuration
"endpoint": {
"type" : "REST"
},
"subtype": "MICROSERVICE",
"type": "MODEL"
},
"name": "sklearn-iris-predictor",
Expand All @@ -142,10 +143,10 @@ The docker image version of your model is deployed through a json configuration
}
```

1. To deploy the model in seldon core, type on command line:
1. To deploy the model in seldon core, type on command line (from the that contains the sklearn_iris_deployment.json file):

```bash
kubectl apply -f models/sklearn_iris/sklearn_iris_deployment.json
kubectl apply -f sklearn_iris_deployment.json
```

2. The deployment will take a few seconds to be ready. To check if your deployment is ready:
Expand All @@ -158,7 +159,7 @@ The docker image version of your model is deployed through a json configuration

### Serve your model:

In order to send a prediction request to your model, you need to query the seldon-core api server and obtain an authentication token from your model key and secret (in this example key and secret are set to "oauth-key" and "oauth-secret" for simplicity). The api server is running on minikube with default port 30032. To query your model you need to
In order to send a prediction request to your model, you need to query the Seldon Core api server and obtain an authentication token from your model key and secret (in this example key and secret are set to "oauth-key" and "oauth-secret" for simplicity). The api server is running on minikube with default port 30032. To query your model you need to

1. Set the api server IP and port:

Expand Down Expand Up @@ -205,4 +206,4 @@ The response contains:
* "routing": This field is relevant when the deployment contain a more complex graph. In this case is empty since we are deploying a single model.
* "data" dictionary: This dictionary contains the predictions for your model classes
* "names": The names of your classes.
* "ndarray": The predicted probabilities for each class.
* "ndarray": The predicted probabilities for each class.