From 0251e9df74e2aa0995e701b5dd7168acd0cbaf86 Mon Sep 17 00:00:00 2001 From: Kento Nozawa Date: Fri, 4 Oct 2024 14:13:53 +0900 Subject: [PATCH] Fix inline code in kubernetes --- kubernetes/README.md | 4 ++-- kubernetes/mlflow/README.md | 4 ++-- kubernetes/simple/README.md | 6 +++--- 3 files changed, 7 insertions(+), 7 deletions(-) diff --git a/kubernetes/README.md b/kubernetes/README.md index 75cdd4de..71ad1e26 100644 --- a/kubernetes/README.md +++ b/kubernetes/README.md @@ -1,6 +1,6 @@ # Distributed Optimization on Kubernetes -This folder contains two kinds of examples with Kubernetes: one is based on [sklearn_simple.py](../sklearn/sklearn_simple.py) and the other is based on [pytorch_lightning_simple.py](../pytorch/pytorch_lightning_simple.py) with MLflow. +This folder contains two kinds of examples with Kubernetes: one is based on [`sklearn_simple.py`](../sklearn/sklearn_simple.py) and the other is based on [`pytorch_lightning_simple.py`](../pytorch/pytorch_lightning_simple.py) with MLflow. -Currently, both [simple/sklearn_distributed.py](./simple/sklearn_distributed.py) and [mlflow/pytorch_lightning_distributed.py](./mlflow/pytorch_lightning_distributed.py) use POSTGRESQL for their backend of `optuna.Study.optimize` to be parallelized. +Currently, both [`simple/sklearn_distributed.py`](./simple/sklearn_distributed.py) and [`mlflow/pytorch_lightning_distributed.py`](./mlflow/pytorch_lightning_distributed.py) use POSTGRESQL for their backend of `optuna.Study.optimize` to be parallelized. Though we do not use it for MLflow records. Of course, you can use POSTGRESQL as backend store of MLflow (https://mlflow.org/docs/latest/tracking.html#where-runs-are-recorded), current example uses HTTP server. diff --git a/kubernetes/mlflow/README.md b/kubernetes/mlflow/README.md index f9d5b79b..789c3a9a 100644 --- a/kubernetes/mlflow/README.md +++ b/kubernetes/mlflow/README.md @@ -2,7 +2,7 @@ This example is only verified on minikube. -This example's code is based on ../../pytorch/pytorch_lightning_simple.py example with the following changes: +This example's code is based on [`pytorch_lightning_simple.py`](../../pytorch/pytorch_lightning_simple.py) example with the following changes: 1. It gives a name to the study and sets `load_if_exists` to `True` in order to avoid errors when the code is run from multiple workers. 2. It sets the storage address to the postgres pod deployed with the workers. @@ -18,7 +18,7 @@ First run `run.sh` which takes two arguments `$IsMinikube` and `$IMAGE_NAME` $ bash run.sh True optuna-kubernetes-mlflow:example ``` -- If you want to run in cloud, please change the `IMAGE_NAME` accordingly in k8s-manifest.yaml and run as follows. Also please make sure that your kubernetes context is set correctly. +- If you want to run in cloud, please change the `IMAGE_NAME` accordingly in `k8s-manifest.yaml` and run as follows. Also please make sure that your kubernetes context is set correctly. ```bash $ bash run.sh False $IMAGE_NAME diff --git a/kubernetes/simple/README.md b/kubernetes/simple/README.md index 0ad4f634..8e53394e 100644 --- a/kubernetes/simple/README.md +++ b/kubernetes/simple/README.md @@ -1,9 +1,9 @@ # Distributed Optimization on Kubernetes -This example's code is mostly the same as the sklearn_simple.py example, +This example's code is mostly the same as the [`sklearn_simple.py`](../../sklearn/sklearn_simple.py) example, except for two things: -1 - It gives a name to the study and sets load_if_exists to True +1 - It gives a name to the study and sets `load_if_exists` to `True` in order to avoid errors when the code is run from multiple workers. 2 - It sets the storage address to the postgres pod deployed with the workers. @@ -18,7 +18,7 @@ Run `run.sh` which takes two arguments `$IsMinikube` and `$IMAGE_NAME` $ bash run.sh True optuna-kubernetes:example ``` -- If you want to run in cloud, please change the IMAGE_NAME accordingly in k8s-manifest.yaml and run as follows. Also please make sure that you kubernetes context is set correctly. +- If you want to run in cloud, please change the `IMAGE_NAME` accordingly in `k8s-manifest.yaml` and run as follows. Also please make sure that you kubernetes context is set correctly. ```bash $ bash run.sh False $IMAGE_NAME