Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
mirrors
ROCmSoftwarePlatform
iGEMMgen
Repository
Branches
Overview
Active
Stale
All
mfma_bwd_nhwc
efe82c3a
·
Fix in igemm_bwd_gtc_driver.h to clear the In tensor when it is needed
·
Apr 18, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
master
default
protected
4e03f4ca
·
add nhwc fp16 config
·
Apr 13, 2021
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
bwd_split_asm_files
a3229bb2
·
Update to compile and test all splitted kernel files
·
Apr 12, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
nhwc_fwd
ae7cef37
·
ignore print in wrw driver
·
Mar 23, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
nhwc_inference_test_gfx1030
eded7b0f
·
Merge branch 'nhwc_fwd' into nhwc_inference_test_gfx1030
·
Mar 22, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
nhwc_fwd_gemmk_split
c771a913
·
fix a bug in find max gks
·
Mar 22, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
fwd_vgpr_alloc_bug
merged
d26fc8e1
·
fix bug for vgpr alloc for fwd pass
·
Feb 10, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
mfma_fp16
0f964208
·
fix bug in param check for two script
·
Feb 02, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
hot_fix_bwd_zet_zero
11bc5bd9
·
hot fix bwd upsampling issue
·
Feb 01, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
split_4G_bwd_wrw
a1c575bd
·
tensor size >4G support for bwd/wrw
·
Feb 01, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
split_4G_pices_per_batch
f20f7fa3
·
remove confusing assert
·
Jan 23, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
wider_tensor_size
71c6256c
·
seperate block idx with thread idx into sgpr and vgpr, when compute n1b transform
·
Jan 21, 2021
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
wrw_b_padding_bug_fix
3fa80e50
·
fix bug in wrw when padding b
·
Dec 07, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
master_fix_fwd_step_repeat_both_2x2
62a13e98
·
fix bug when fwd's step and repeat are both 2x2
·
Nov 30, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
config-automatical-generation
b0442012
·
fix some format issue
·
Nov 06, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
mfma_fwd
bf3bf521
·
Merge branch 'master' into mfma_fwd
·
Oct 20, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
fp32_multi_k
04247997
·
add some configs and script for multi k instruction
·
Oct 15, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
gtc_wrw_2x2x2
491eca3d
·
remove debug code
·
Oct 14, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
wrw_xdlops_with_reduction
e30f2c50
·
add an xdlops version for 128x128 block
·
Aug 26, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
wrw_reduction
51e12271
·
right code for 32x32x4 kernel
·
Jul 20, 2020
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
Prev
1
2
3
4
5
Next