The recently concluded F8 developer conference in San Jose, California saw Facebook announcing Caffe2, a new lightweight, modular, and scalable deep learning framework to the world. Built on the original Caffe, Caffe2 is a trendy type of artificial intelligence (AI) that has been designed by keeping expression, speed, and modularity in mind.
For the uninitiated, deep learning is one of the latest advances in Artificial Intelligence (AI) and computer science in general, which basically involves training artificial neural networks on large amounts of data, like photos etc., and then getting them to make inferences based on the new data that has been fed to them.
According to experts, deep Learning possess the potential of bringing significant breakthroughs in the field of machine learning and artificial intelligence. Caffe2 aims to provide developers an easy and straightforward to experiment with deep learning first hand.
When a developer is getting started with deep learning, Caffe2 can help them in understanding the workflow of how they can create and deploy their deep learning application. There are basically two primary stages for working with a deep learning application built with Caffe2:
Create your model, which will learn from your inputs and information (classifiers) about the inputs and expected outputs.
Run the finished model elsewhere. e.g., on a smart phone, or as sub-component of a platform or a larger app.
The Caffe2 announcement kind of builds on the social networking giant's contributions to the Torch open source deep learning framework and the PyTorch framework that the Facebook Artificial Intelligence Research (FAIR) group had created.
But, according to Facebook AI Platform engineering lead Yangqing Jia's recent comment on Hacker News, Caffe2 is quite different from PyTorch. According to Jia, while PyTorch is good when it comes to research, experimentation and trying out exotic neural networks; Caffe2, on the other hand, is headed towards supporting more industrial-strength applications with a heavy focus on mobile.
While Caffe2 looks good on paper, it is surely going to face some tough competition as there are aplenty of other open source deep learning frameworks in the market for people to use for all kinds of purposes.
As soon as Facebook announced its new open source framework, Nvidia and Intel published blog posts showing some early performance numbers. Even Qualcomm revealed that it is working with Facebook to optimize Caffe2 and Qualcomm’s Snapdragon neural processing engine (NPE) framework.
The blogpost published by Nvidia read, “we’ve fine-tuned Caffe2 from the ground up to take full advantage of the NVIDIA GPU deep learning platform. Caffe2 uses the latest NVIDIA Deep Learning SDK libraries — cuDNN, cuBLAS and NCCL — to deliver high-performance, multi-GPU accelerated training and inference. As a result, users can focus on developing AI-powered applications, knowing that Caffe2 delivers the best performance on their NVIDIA GPU systems.”
Public cloud infrastructure provider Amazon Web Services (AWS) has also added experimental support for Caffe2 in its Deep Learning Amazon machine image (Read Here). Further, Microsoft Azure has also added Caffe2 support to its Data Science Virtual Machine (Read Here).
For the uninitiated, deep learning is one of the latest advances in Artificial Intelligence (AI) and computer science in general, which basically involves training artificial neural networks on large amounts of data, like photos etc., and then getting them to make inferences based on the new data that has been fed to them.
According to experts, deep Learning possess the potential of bringing significant breakthroughs in the field of machine learning and artificial intelligence. Caffe2 aims to provide developers an easy and straightforward to experiment with deep learning first hand.
When a developer is getting started with deep learning, Caffe2 can help them in understanding the workflow of how they can create and deploy their deep learning application. There are basically two primary stages for working with a deep learning application built with Caffe2:
Create your model, which will learn from your inputs and information (classifiers) about the inputs and expected outputs.
Run the finished model elsewhere. e.g., on a smart phone, or as sub-component of a platform or a larger app.
The Caffe2 announcement kind of builds on the social networking giant's contributions to the Torch open source deep learning framework and the PyTorch framework that the Facebook Artificial Intelligence Research (FAIR) group had created.
But, according to Facebook AI Platform engineering lead Yangqing Jia's recent comment on Hacker News, Caffe2 is quite different from PyTorch. According to Jia, while PyTorch is good when it comes to research, experimentation and trying out exotic neural networks; Caffe2, on the other hand, is headed towards supporting more industrial-strength applications with a heavy focus on mobile.
While Caffe2 looks good on paper, it is surely going to face some tough competition as there are aplenty of other open source deep learning frameworks in the market for people to use for all kinds of purposes.
As soon as Facebook announced its new open source framework, Nvidia and Intel published blog posts showing some early performance numbers. Even Qualcomm revealed that it is working with Facebook to optimize Caffe2 and Qualcomm’s Snapdragon neural processing engine (NPE) framework.
The blogpost published by Nvidia read, “we’ve fine-tuned Caffe2 from the ground up to take full advantage of the NVIDIA GPU deep learning platform. Caffe2 uses the latest NVIDIA Deep Learning SDK libraries — cuDNN, cuBLAS and NCCL — to deliver high-performance, multi-GPU accelerated training and inference. As a result, users can focus on developing AI-powered applications, knowing that Caffe2 delivers the best performance on their NVIDIA GPU systems.”
Public cloud infrastructure provider Amazon Web Services (AWS) has also added experimental support for Caffe2 in its Deep Learning Amazon machine image (Read Here). Further, Microsoft Azure has also added Caffe2 support to its Data Science Virtual Machine (Read Here).
Advertisements