It's been two months since this profile received a new review

Fabric for Deep Learning (FfDL) Reviews & Product Details

Deep learning frameworks such as TensorFlow, PyTorch, Caffe, Torch, Theano, and MXNet have contributed to the popularity of deep learning by reducing the effort and skills needed to design, train, and use deep learning models. Fabric for Deep Learning (FfDL, pronounced ‚Äúfiddle‚Äù) provides a consistent way to run these deep-learning frameworks as a service on Kubernetes.

Seller

IBM

Discussions

Fabric for Deep Learning (FfDL) Community

Product Description

Deep learning frameworks such as TensorFlow, PyTorch, Caffe, Torch, Theano, and MXNet have contributed to the popularity of deep learning by reducing the effort and skills needed to design, train, and use deep learning models. Fabric for Deep Learning (FfDL, pronounced ‚Äúfiddle‚Äù) provides a consistent way to run these deep-learning frameworks as a service on Kubernetes.

Top-Rated Alternatives

NVIDIA Deep Learning GPU Training System (DIGITS)

4.5/5

(22)

View All Alternatives

G2 reviews are authentic and verified.

Here's how.

MS

Mohammad S.

Small-Business (50 or fewer emp.)

9/12/2023

"Navigating the Pros and Cons of Fabric for Deep Learning (FfDL)"

4.5/5

What do you like best about Fabric for Deep Learning (FfDL)?

I find Fabric for Deep Learning (FfDL) incredibly versatile and user-friendly. Its most helpful feature is its ability to seamlessly integrate with various deep learning frameworks, making it easy for users to work with their preferred tools and libraries. The upside of using FfDL lies in its robust scalability, allowing for efficient training of deep learning models on various infrastructures, whether it's on-premises or in the cloud. Additionally, the comprehensive documentation and active community support are invaluable resources for users seeking assistance and insghts. Review collected by and hosted on G2.com.

What do you dislike about Fabric for Deep Learning (FfDL)?

While FfDL offers many advantages, one downside is the learning curve for newcomers, especially those without prior experience in deploying deep learning models. The initial setup and configuration can be a bit challenging. Additionally, although the documentation is thorough, some users may still encounter issues that require more extensive troubleshooting. However, with time and community support, these challenges can be overcome, making FfDL a powerful tool for deep learning practitioners. Review collected by and hosted on G2.com.

What problems is Fabric for Deep Learning (FfDL) solving and how is that benefiting you?

Fabric for Deep Learning (FfDL) is a game-changer in the realm of deep learning by simplifying complex infrastructure management, offering framework agnosticism, and providing efficient scalability. It addresses the inherent challenges of deep learning, allowing users to focus on model development and reducing operational overhead. With its resource optimization and supportive community, FfDL streamlines deep learning projects, ultimately empowering users to tackle complex tasks and achieve success in this rapidly evolving field. Review collected by and hosted on G2.com.

PS

Pierre S.

Small-Business (50 or fewer emp.)

8/15/2023

"Smooth Sailing with Fabric for Deep Learning: An Honest Review As A Consultant"

5/5

What do you like best about Fabric for Deep Learning (FfDL)?

Well, the coolest thing about Fabric for Deep Learning (FfDL) is how it hooks up with Kubernetes. You can just throw in your machine learning or AI service, like TensorFlow, and boom, it's up and running on FfDL. You don't need to be a tech genius – just wrap your head around some basic kubectl, docker, and helm chart stuff, and you're good to go! Review collected by and hosted on G2.com.

What do you dislike about Fabric for Deep Learning (FfDL)?

Hmmm, the not-so-fun part about FfDL is that you've gotta be buddies with helm and kubectl. Before you start rolling with FfDL, you've gotta get these tools under your belt. Plus, it's not a solo mission – you gotta have your Kubernetes or EKS (Amazon's Kubernetes) cluster all set up and raring to go. Review collected by and hosted on G2.com.

What problems is Fabric for Deep Learning (FfDL) solving and how is that benefiting you?

FfDL's like a superhero for those of us diving deep into machine learning. It takes the hassle out of getting those heavyweight deep learning frameworks – think TensorFlow and PyTorch – up and running on Kubernetes or EKS. It’s like a smooth ride to setting up and managing your deep learning playground. Super beneficial, especially when you're dealing with hefty models that need some serious computing power. Review collected by and hosted on G2.com.

S

SHASHIDHAR KUDARI .

Small-Business (50 or fewer emp.)

8/17/2023

"Project is not active"

1.5/5

What do you like best about Fabric for Deep Learning (FfDL)?

They were trying to solve the problem of framework independent deep learning model training. Which is a very good usecase Review collected by and hosted on G2.com.

What do you dislike about Fabric for Deep Learning (FfDL)?

The project is no longer maintained and the last commit to the GitHub is around 5 years back Review collected by and hosted on G2.com.

UE

Verified User in E-Learning

Small-Business (50 or fewer emp.)

9/27/2022

"Fabric for deep learning is a platform to deploy services of ML / AI on kubernetes cluster"

4.5/5

What do you like best about Fabric for Deep Learning (FfDL)?

If you are working on kubernetes cluster and want to deploy any service of machine learning or AI, like TensorFlow, you can easily use FFDL. You just need to know basic commands of kubectl, docker, helm charts. Review collected by and hosted on G2.com.

What do you dislike about Fabric for Deep Learning (FfDL)?

You need to understand helm and kubectl commands before using Fabric for deep learning. You also need to have a working kubernetes/EKS cluster. Review collected by and hosted on G2.com.