-
Notifications
You must be signed in to change notification settings - Fork 346
Pull requests: google/XNNPACK
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Implement PRELU as a binary operator instead of a one-off operator type
#7034
opened Sep 5, 2024 by
copybara-service
bot
Loading…
Add
xnn_float16_t
for use instead of uint16_t
#7023
opened Sep 4, 2024 by
copybara-service
bot
Loading…
enable AVX_VNNI_INT8 instruction for qd8-f32-qc8w-gemm/igemm
#7019
opened Sep 4, 2024 by
xujuntwt95329
Loading…
Use the new
pthreadpool_parallelize_1d_dynamic
strategy in the unary and binary ops, where appropriate.
#7008
opened Sep 3, 2024 by
copybara-service
bot
Loading…
Fix filepath formatting for Windows in update-microkernels.py
#6963
opened Aug 29, 2024 by
Ch3nYuY
Loading…
add f32-gemm-5x16-minmax-fma3-broadcast-prfm microkernel
#6915
opened Aug 22, 2024 by
Ch3nYuY
Loading…
Introduce flags for qb4 scale format in xnn_define_blockwise_quantized_tensor_value
#6899
opened Aug 20, 2024 by
GregoryComer
Loading…
Enable wasmusdot qd8_f32_qc8w gemm/igemm in gemm-config
#6863
opened Aug 14, 2024 by
fanchenkong1
Loading…
AVX2 vpmadd GEMM move constants to functions to params
#6802
opened Aug 5, 2024 by
copybara-service
bot
Loading…
vpmaddubsw version of avx512 mimic to vnni
#6733
opened Jul 17, 2024 by
copybara-service
bot
Loading…
XNN_ARCH_WASMSIMDRELAXED
implies XNN_ARCH_WASMSIMD
.
#6686
opened Jul 5, 2024 by
copybara-service
bot
Loading…
Fix duplicate test and benchmark generation
#6670
opened Jul 2, 2024 by
copybara-service
bot
Loading…
ProTip!
Filter pull requests by the default branch with base:master.