I am the wrong person to ask. We compete with tensorflow going after the larger scale stuff where hadoop is a requirement and the cloud is a swear word.
I work with .net and java shops where running on windows is a requirement and hippaa is considered "lite" and they dont know what a gpu is.
Despite the marketing noise that is still most of the world.
Look no further than another comment that showed java job postings vs machine learning.
My biased comment out of the way: if you can convince ops on why you need this deep learning thing then be my guest. Their first question to you is probably going to be: Does dell/hp sell this as a reference box?
It mostly comes down to roi and what your existing stack is.
My job is hard enough as it is. If you can convince IT to allocate an r&d budget be my guest. That is more or less our specialty. Email is in profile if you want specifics.
I work with .net and java shops where running on windows is a requirement and hippaa is considered "lite" and they dont know what a gpu is.
Despite the marketing noise that is still most of the world.
Look no further than another comment that showed java job postings vs machine learning.
My biased comment out of the way: if you can convince ops on why you need this deep learning thing then be my guest. Their first question to you is probably going to be: Does dell/hp sell this as a reference box?
It mostly comes down to roi and what your existing stack is.
My job is hard enough as it is. If you can convince IT to allocate an r&d budget be my guest. That is more or less our specialty. Email is in profile if you want specifics.