AIMunchkin

8 ms·

1. 31.▲
   by AIMunchkin 10y ago

   It does... Badly... And you used to have to pay for it...
2. 32.▲
   by AIMunchkin 10y ago

   It would have been fantastic if Intel had stopped beating the linpack horse a lot sooner and built a viable competitor to GPUs by now. Not this timeline though alas... Maybe 2020?
3. 33.▲
   by AIMunchkin 10y ago

   It would send data from CPU space at PCIE speeds (~10.5 GB/s). What's the bandwidth from that SSD? <1 GB/s? Seems like plenty of margin... That said, onboard SSD is cool if it can read without pre-empting the PCIE bus.
4. 34.▲
   by AIMunchkin 10y ago

   OpenCL IMO is an ugly API born of the equivalently ugly CUDA driver API because Steve Jobs got butthurt at Jensen Huang for announcing a deal with Apple prematurely. Downvote all you like, but as John Oliver would say "That's jus
5. 35.▲
   by AIMunchkin 10y ago

   Except that Pascal class GPUs support true unified virtual memory so some minor code changes might provide mostly the same performance using memory-mapped files. Caveat: I don't own a Pascal class GPU yet.
6. 36.▲
   by AIMunchkin 10y ago

   Google has done a far better job of abstracting the underlying math of machine learning into vertical-oriented services like the CloudVision API and the Cloud Natural Language API. In contrast, the AWS Machine Learning service provides a 10
7. 37.▲
   by AIMunchkin 10y ago

   I think the TensorFlow the API has indeed won. TensorFlow the Machine Learning Engine not so much. It's not derisively referred to as TensorSlow(tm) without cause.