Vil du fortsette å lese, velg et av alternativene nedenfor
Is the cloud the key to democratizing AI?
At the peak of the Japanese harvest, Makoto Koike’s mother spends around eight hours a day sorting cucumbers from the family farm into different categories—a dull, time-consuming task that her son decided to automate. Although Makoto wasn’t a machine learning expert, he started playing around with TensorFlow, Google’s popular open-source machine learning framework, and developed a deep learning model that could sort cucumbers by size, shape and other attributes. The system isn’t perfect (it has an accuracy rate of around 75 percent). But it’s a sign of how AI could soon transform even the smallest family-run business.
Giants wins the expertise
AI AI Google’s improve learning Makoto: how Amazon, Facebook to of pool and have which beasts’ Most most system, companies big scientists, interest personal leaves smaller as same on of translation teams AI Fortune Google, this medium search in data dedicated other of course, are, Microsoft’s and transformative Amazon’s their many recommendation well Giants expertise. but power. But has also drained tools, the business, used underpins boat can and eager as Microsoft, and and widely services. applications enterprises short like assistant, the Deep Apple, well-aware 500 explore the place. in Cortana in
considerable believe experts analyze huge certain teach sums conscious overcome those issues—and or on them. spend Even computing neural enterprises a data power top way providers big of that and they prepare need to patterns are afford objects. to network hire to they’ve these sets, AI can cloud to their and help them people still However, to found recognize the
as learning a Machine Service
simplify to on business of image service, well deploying recognition, like building, of models a with offering is learning and say—as or process Azure, now platforms major applications Amazon as AI Web a by tools component the cloud providing Essentially Google Machine are IBM Cloud, in models—for customers to learning grunt deep the companies training their adding these pretrained customized as Services cloud Microsoft involved cloud. work that do the and Cloud. to (AWS), access AI,
Skymind, but who of enterprise. the to there can says if There an in against; to developers scientists deep through software you apps how the of tools give who tools of code; Nicholson, code there are a build how tools – CEO data them for are to API for Chris relate which know majority finally tools and world, that the people properly may provider tune covers are algorithms, basically vast for for know who GUIs, clickers, learning not
the more helping spectrum, or text simply built SageMaker, sit the Rekognition want serve are or videos Translation for that and ML Engine likes learning Cloud objects, models: Amazon pretrained while up you and models scale; platforms translated—and for Azure Google common Studio, the experts API scientist around feed want APIs their Amazon deploy are Microsoft to analyzed Google and data toward the end to ML the wait in at deep broadly your You of results. you data—images train, of similar tune,
to of can different The not different a the deep that cucumbers. identify business is may of words, solve learning you typically kittens to address. to is that breeds want recognize specific which In types used API be having latter good it with models other pretrained able if issue approach an no it’s problem,
Nicholson. make kind bunch predictions you says we of to it, on a ‘Hey, They’re about found images,’” it a we can of data, now use model saying, trained –
to train that’s want as But your still the – and just necessary that data solution. you model a customize it’s in a on own that if hard false solution sense to just as
Custom learning machine made
automatically models interface, models, first AutoML the Cloud Google recognition via custom neural pretrained deep basic the launched custom that model. uses is machine data the one-size-fits-all AutoML, bridge from a Cloud which more bid In to recently customer networks users new lets image to between a a a learning customized learning and service. for the build create Vision, system highly release gap drag-and-drop
AutoML companies Cloud for… testing been Several have