- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
i understand 0.8.1 is the DLA version, 16x32 is k vectorization and channel depth and fp16 is the floating point precision. but what does arch13/09/06 mean ?
0-8-1_a10dk_fp11_16x32_arch16.aocx 0-8-1_a10dk_fp16_8x24_arch13.aocx 0-8-1_a10dk_fp16_8x48_arch06.aocx 0-8-1_a10dk_fp11_16x32_arch19.aocx 0-8-1_a10dk_fp16_8x32_arch09.aocx 0-8-1_a10dk_fp16_8x48_arch08.aocxLink Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I think they are different CNN architectures or variants in the same architecture. Eg: arch 17 is SqueezNet according to https://software.intel.com/en-us/articles/openvino-install-linux-fpga
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Since there is no description on these binaries, only way is to try out all of them and see which one performs better.
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page