Charmed Kubeflow Pros and Cons

I like the portability of it, which makes easier to work with any kubernete clusters whether it's on single computer or in cloud. Review collected by and hosted on G2.com.

It was difficult to setup initially we had to keep dedicated team members to setup it. Review collected by and hosted on G2.com.

1. It uses Kubernetes as a backend.

2. It adheres to follow best practices of Mlops & containerization.

3. Once a workflow is properly defined then it becomes very easy to automate it.

4. It does a great python sdk to design pipeline.

5. The Front end/UI to use Kubeflow pipeline is awesome.

6. It also displayed all the logs. Review collected by and hosted on G2.com.

1. Initial steep learning curve as it involves lot of variety of concepts under one roof.

2. So the user must have knowledge apart from usual ML stuffs about Docker/Container tech, kubernetes.

3. Even the initial setup process is not so initiative.

4. Based on what material is available on its docs, it seems setting it up is comparatively easy on GCP (in fact I have use it only on GCP) Review collected by and hosted on G2.com.

Automates flow of production machine learning. Kubeflow can be easily integrated with kubernetes on a lot of different cloud providers, such as Amazon web service (using Elastic Kubernetes Service), or with Google cloud (with Google Kubernetes Engine). It has API interface in different languages, espically easy to integrate with python and docker containers. Which helps users to build their own rerunnable and plugable machine learning pipelines. Review collected by and hosted on G2.com.

No easy integration with terraform and integration with domain name servers on Amazon web service. Which means that deploying kubeflow can be difficult dependent on what existing infrastructure looks like. If companies already have existing models to integrate with kubeflow that does not use containers, it could cost extra effort to implement them as Kubeflow is best used with docker containers and run on kubernetes. Review collected by and hosted on G2.com.

Scability, portability and distribute. The all-in-one feature of Kubeflow has made team easy to use and have saved lot amount of time .This is easy to use for new learner. Review collected by and hosted on G2.com.

There was a need of CI/CD feature to the team. On Kubeflow couldn't find the feature of CI/CD. Review collected by and hosted on G2.com.

1. The kubeflow is based on kubernetes, it makes the scaling of models and load balancer quite easy

2. The pipelines are very elegant and make the stages very clear Review collected by and hosted on G2.com.

1. The documents of kubeflow is incomplete and some examples of source codes ( especially for docker images ) are difficult to find

2. There are no simple examples of data passing in different stages in the pipelines

3. The learning curve of DSL is high for data scientists Review collected by and hosted on G2.com.

I especially like how it supports all the available ml frameworks starting from tfx,pytorch Caffe Review collected by and hosted on G2.com.

I would love to have a full-featured feature store with CRUD operation over REST endpoints, although that is in beat and will be released quickly for the stable release Review collected by and hosted on G2.com.

Pipeline and visualization and artifacts within the pipeline Review collected by and hosted on G2.com.

Writing code to create Pipeline. Kale is available but expect a Kubeflow ' s native soltuion to simplify the complete workflow. There is not enough documentation and a simple Google search doesn't provide a quick solution. Even stackoverflow community is not developed. A simple UI based approach to make the complete stack easy and accessible is required. Review collected by and hosted on G2.com.

It's usability, and easy launching of notebooks and creating models over the cloud!

Kubeflow can easily be setup over a cloud and many Data Engineers/Scientists can leverage this stuff. Review collected by and hosted on G2.com.

Nothing as of now.

UI can be improved a bit Review collected by and hosted on G2.com.

Organized way to work on data science projects. Experiment tracking. Review collected by and hosted on G2.com.

Complexity and learning curve for making a tailor made custom solutions Review collected by and hosted on G2.com.

so like kubeflow pipelines are the best way to build ML workflows. and it is an open-source community-driven project. Review collected by and hosted on G2.com.

in reality, installing kubernetes correctly not easy. kubeflow has many components that actually make kubeflow working more complexer. Review collected by and hosted on G2.com.