|
|
cff4bb8aaa
|
doc: add section for quantization-related APIs and refactor
|
2025-10-16 13:38:24 -07:00 |
|
|
|
eed2287b10
|
doc: document quantization mode and clean up quantization page
|
2025-10-15 02:01:20 +02:00 |
|
|
|
8e660f58cb
|
doc: add link to f8 matmul example to doc
|
2025-10-03 11:44:28 -07:00 |
|
|
|
d559e96879
|
doc: minor renaming for nicer displayed list
|
2025-10-03 11:44:28 -07:00 |
|
|
|
c106dd4f49
|
doc: clarify saturation behavior
|
2025-09-04 13:46:23 +02:00 |
|
|
|
79fd09b6a3
|
doc: corrected GPU brand names
|
2025-08-26 17:03:01 -07:00 |
|
|
|
b1c9e6401d
|
doc: updated data type support tables for clarity
|
2025-08-26 17:03:01 -07:00 |
|
|
|
d2bf4fd4eb
|
api: introduce precomputed reductions primitive attribute
|
2025-08-26 10:50:04 -07:00 |
|
|
|
38cbb84874
|
doc: fixed broken links in quantization section
|
2025-08-22 10:18:48 -07:00 |
|
|
|
3634f41f51
|
doc: updates to atributes pages wrt host scale and zp
|
2025-08-20 13:09:13 -07:00 |
|
|
|
b2a546d283
|
doc: fixed serveral broken formulas in quantization section
|
2025-08-14 16:17:40 -07:00 |
|
|
|
fe299f1f58
|
doc: postops: update docs to describe binary select postop
|
2025-05-23 16:08:12 -07:00 |
|
|
|
1c6468b8ec
|
xe, doc: add alignment checks
|
2025-05-22 16:26:28 -07:00 |
|
|
|
17af1def52
|
doc: update fp4 support
|
2025-04-18 16:29:03 -07:00 |
|
|
|
1539f98865
|
doc: graph: extend Add and Subtract to support s32 dtype
|
2025-04-15 21:23:58 +08:00 |
|
|
|
e20f9b5a8a
|
doc: Add Xe2 architectures
|
2025-04-01 15:04:04 -07:00 |
|
|
|
4d9a89c531
|
doc: fixup fp8 support documentation
|
2025-04-01 15:04:04 -07:00 |
|
|
|
396fdfc9f7
|
all: clean graph compiler backend
|
2025-03-28 23:01:06 +08:00 |
|
|
|
ba5f1c7f13
|
doc: coloring the header and a function name for better visibility
|
2025-02-21 15:21:19 -08:00 |
|
|
|
69a8ee19de
|
doc: clarify conversions can be impacted by double-rounding
|
2025-01-14 14:25:34 +01:00 |
|
|
|
f6f966988b
|
doc: add f4_e3m0 in data-types section
|
2024-12-20 11:54:01 +01:00 |
|
|
|
d4ffaca7d9
|
doc: matmul: updated supported data types and minor edits (#2217)
Co-authored-by: Ranu Kundu <ranu.kundu@intel.com>
|
2024-11-20 09:26:17 -08:00 |
|
|
|
7fc333d554
|
doc: update datatype support tables
|
2024-09-13 10:40:20 +02:00 |
|
|
|
ee4ed242c1
|
doc: document rounding behavior and attribute
|
2024-08-28 14:31:54 +02:00 |
|
|
|
ed9f6d1b08
|
x64, api, benchdnn: revert dice post op and related changes
Reverts the following commits:
* 8bfcff59efbf5c9846d7f75a2a847e35676d582c verbose_converter: add prelu with scaleshift support
* d3ffb808d5e20068630c4052dbd0b37b7721b59c x64: add jitted prelu with scaleshift postop
* f2c62e01799eb62d68c2c0c5a348fe16d3dc6e6d x64: properly initialize postop injector constant tables
* 6ae2d9cc8c485fe4c8db720752324c871c7a302f benchdnn: add prelu postop with scaleshift coverage
* c11bab0c4970886c84eee1c2e7e98266d05dfa35 api: add dice post op
|
2024-08-06 13:46:23 -07:00 |
|
|
|
c11bab0c49
|
api: add dice post op
|
2024-08-05 09:49:29 +02:00 |
|
|
|
630e136cae
|
src: gpu: intel: ocl: matmul: add documentation for dropout attribute
|
2024-06-27 13:30:04 -07:00 |
|
|
|
ff2f84d23c
|
doc: add ukernel dev_guide
|
2024-05-14 00:15:48 -07:00 |
|
|
|
0a58dd98fb
|
doc: quantization: bias is not quantized
|
2024-04-19 12:50:08 -07:00 |
|
|
|
e1f30dc726
|
doc: correct gpu support for f16 training
|
2024-03-25 12:08:48 -07:00 |
|
|
|
f88eac7848
|
doc: data_types: add u4/s4
|
2024-02-26 12:18:21 -08:00 |
|
|
|
ea1a8207e3
|
doc: prim: add fp8 type documentation
|
2024-02-12 09:39:34 -08:00 |
|
|
|
56bf5efb5c
|
doc: minor additions for actual behavior and different ref fixes
|
2024-02-02 14:50:38 -08:00 |
|
|
|
79bc3f5ad7
|
doc: misc fixes
|
2024-01-26 09:52:26 -08:00 |
|
|
|
c48b62906b
|
doc, examples: fix typos
|
2024-01-23 10:50:22 +08:00 |
|
|
|
0c9d21d734
|
doc: programming_model: attr: extend fpmath mode
|
2024-01-22 14:42:02 -08:00 |
|
|
|
c8264ba066
|
api: add determinism attribute
|
2024-01-16 11:28:20 -08:00 |
|
|
|
94a79fc446
|
doc: add accumulation attribute
|
2023-11-13 15:33:14 +01:00 |
|
|
|
ac331c6344
|
doc: add information about bf8 and fp8 support in the library
|
2023-10-21 16:54:35 -07:00 |
|
|
|
b917b4d8b9
|
doc: convolution, reorder: move oscale -> scales
|
2023-06-29 16:10:50 -07:00 |
|
|
|
04e5e88e29
|
doc: updated URLs referring to software.intel.com
|
2023-06-13 13:10:49 -07:00 |
|
|
|
5aced11588
|
doc: fixed doxygen warnings
|
2023-05-05 09:13:19 -07:00 |
|
|
|
c452c13ccc
|
doc: programming model: sum post op: clarify quantization
|
2023-04-20 10:32:04 -07:00 |
|
|
|
0d65f31dcc
|
doc: programming_model: add boolean data type
|
2023-04-13 09:42:59 +08:00 |
|
|
|
b19cd37f7e
|
gpu: pool: enable f64 pooling in ref impl (#5343)
|
2023-03-01 20:22:54 +01:00 |
|
|
|
598fcb5017
|
doc: normalize tables
|
2023-02-24 09:44:46 -08:00 |
|
|
|
81555aad81
|
gpu: lnorm: added double datatype for ref impl (#5256)
* gpu: lnorm: added double datatype ref impl
|
2023-02-10 20:07:56 +01:00 |
|
|
|
83b167b80e
|
doc: update quantization documentation
|
2022-12-16 12:00:55 -08:00 |
|
|
|
f0a28862d1
|
doc: graph: add operation documents
|
2022-11-18 20:58:44 +08:00 |
|
|
|
80310a05e0
|
api, tests, doc: remove scale from eltwise post op
|
2022-11-04 15:43:22 -07:00 |
|
