- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Folks,
I am Guru, Software Developer from India. I have the following systems, Ubuntu 16.04, Intel® Core™ i3-5005U CPU @ 2.00GHz × 4 processor and Intel® HD Graphics 5500 (Broadwell GT2) graphics. I know its a fifth generation processor, still can i install OpenVINO, as I work on core Computer Vision using OpenCV and Caffe or OpenCV, TensorFlow, SSD Mobilenet. as per the notes, I know its good on sixth generation processors.
Can you please let me know, would it be effective to install OpenVINO to increase the performance of my systems.
Thanks
Guru
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Guru,
CPU:
- 6th to 8th Generation Intel Core and Intel Xeon processors
- Intel® Pentium® processor N4200/5, N3350/5, N3450/5 with Intel® HD Graphics
Graphics:
- 6th to 8th generation Intel Core processor with Iris® Pro graphics and Intel HD Graphics
- 6th to 8th generation Intel Xeon processor with Iris Pro graphics and Intel HD Graphics (excluding the e5 product family, which does not have graphics)
These are our target platform from " https://software.intel.com/en-us/openvino-toolkit "
Regards,
Peter.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Guru,
Looks like, you are talking about 2 different things.
* OpenVINO provides multiple inference engine plugins for each compute device.
. CPU: mklDNN (Intel math kernel library)
. GPU: clDNN (Intel OpenCL)
. MYRIAD: Movidius VPU plugin
. FPGA: FPGA plugin
You can get performance advantage through each Intel plugins.
* Python is one of "framework"s, which it does not related to deep learning performance itself.
OpenVINO provides a python wrapper but still need to be improved.
Please check inference engine samples also.
Regards,
Peter.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks again for the detailed response Peter.
One last question, its reg.Intel FPGA, while googling i figured out that OpenCV can be used in FPGAs, Can you please let me know, If I would have the liberty to code and deploy Deep Learning models like OpenCV DNN using Cafee, Tensorflow(using python) in FPGAs, just like an Intel powered PC.
Regards
Guru
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Guru,
Have you read this?
https://www.intel.com/content/www/us/en/programmable/solutions/acceleration-hub/overview.html
Regards,
Peter.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Have you actually tried running OpenVINO? 6-8 generation is officially tested configuration, but in reality there should be nothing limiting you from running on Broadwell CPU. Given that there are SSE optimizations, you may benefit in terms of performance.
Yury
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page