Skip to content

[feat] add more op #35

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 86 commits into from
Mar 22, 2025
Merged

[feat] add more op #35

merged 86 commits into from
Mar 22, 2025

Conversation

chraac
Copy link
Owner

@chraac chraac commented Mar 22, 2025

Make sure to read the contributing guidelines before submitting a PR

Closing #19

chraac added 30 commits March 2, 2025 22:03
@chraac
move op key generate function to kOpCaps
b3ce765
@chraac
fix op desc print
c2076d8
@chraac
try fix rms_norm
33b2960
@chraac
Revert "try fix rms_norm"
fc300e3 
This reverts commit 33b2960.
@chraac
add quantization type support by converting them to float
9a0e397
@chraac
enable quantization tensor for mulmat in gpu/npu
72c587e
@chraac
Merge branch 'dev-refactoring' into dev-add-more-op
af16143
@chraac
fix asan error
cb3f8f7
@chraac
add log and assert
50b9a1d
@chraac
insert output convert operator after mulmat
ff240b0
@chraac
add log
8353922
@chraac
fix some error in running
fc1a683
@chraac
disable permute again
2550da7
@chraac
add log
7e71165
@chraac
add error function
f92ff47
@chraac
Revert "add error function"
b718b99 
This reverts commit f92ff47.
@chraac
add log
63e5f10
@chraac
more log
29a2147
@chraac
disable convert op in graph
e8387a7
@chraac
wip
1913372
@chraac
add f16 config for graph
a93a69e
@chraac
set f16 precision for f16 graph
b5f1ee9
@chraac
fix override data type
5d8cc5a
@chraac
add comment
a4bb747
@chraac
add config flag to enable quantize type
40b6cc1
@chraac
add log
9e545eb
@chraac
more quantized type for cpu and gpu backend
daf5862
@chraac
enable all quant types for cpu and gpu backend
3693ec2
@chraac
rename
e12b601
@chraac
wip
b5a8116
chraac added 25 commits March 14, 2025 18:01
@chraac
wip
5419435
@chraac
fix comment string
78a4423
@chraac
add file for profiler
68c58ea
@chraac
change qnn-cpu to GGML_BACKEND_DEVICE_TYPE_ACCEL, so that we can run …
1307033 
…tests on cpu
@chraac
wip
9d33c55
@chraac
profiler: refactoring
7063ff7
@chraac
wip
497a38d
@chraac
add implement for print_profile_events
4e3edf1
@chraac
set-up profiler for graph
a3a8660
@chraac
set profiler to graph execute
ff4351a
@chraac
pretty print events
7d4f780
@chraac
unified log print prefix
abb0437
@chraac
print event count
9761a22
@chraac
enable optrace
4d98a69
@chraac
print duration at event end
c4a7197
@chraac
wip
38f9e51
@chraac
add more detailed soc information
775e413
@chraac
wip
0a5e99e
@chraac
move device caps array into qnn-lib.cpp
e34228a
@chraac
remove lib_name in device_context
54dd18d
@chraac
move get_graph_key_from_cgraph to graph.cpp
186494c
@chraac
add override type for tensor key
ee1b6d1
@chraac
use override_type instead of original data type for graph key
86e07a0
@chraac
append op type to tensor name to fix error in qwen
2909892
@chraac
remove todo
b6829d1
@chraac chraac added the enhancement New feature or request label Mar 22, 2025
@chraac chraac self-assigned this Mar 22, 2025
@chraac chraac linked an issue Mar 22, 2025 that may be closed by this pull request
[feat] Supports more quantization data type #19
Closed
@chraac chraac merged commit a1ab674 into dev-refactoring Mar 22, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees

@chraac chraac

Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

[feat] Supports more quantization data type
1 participant
@chraac