Skip to content
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

[GPU] Bugfix accumulate sum of FC dyn-quan #27657

Merged
merged 4 commits into from
Nov 26, 2024
Merged

[GPU] Bugfix accumulate sum of FC dyn-quan #27657

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
df02880
[GPU] Bugfix accumulate sum of FC dyn-quan
byungilm Nov 21, 2024
d449123
[GPU] Apply comments
byungilm Nov 22, 2024
99cb557
[GPU] Add _rte for conversion
byungilm Nov 22, 2024
f3cabe2
Merge branch 'master' into bugfix_dyn_qaun_int8_sd
byungilm Nov 25, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/plugins/intel_gpu/src/kernel_selector/cl_kernels/fully_connected_gpu_bf_tiled.cl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@ KERNEL(quantize_input)(
half4 buff = input_0[i] / (half4)quan_scale;
quantized_value[i] = CAT(CAT(convert_, MAKE_VECTOR_TYPE(DQ_TYPE, INPUT_LOAD_SIZE)), _rte)(buff);
#if COMPRESSED_WEIGHTS_INT8
quantized_sum += (buff[0] + buff[1] + buff[2] + buff[3]);
quantized_sum += ((half)quantized_value[i][0] + (half)quantized_value[i][1] + (half)quantized_value[i][2] + (half)quantized_value[i][3]);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

what about using int for accumulation instead of half? It will reduce quantization error

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Applied.

#endif
vstore4(quantized_value[i], 0, &quantized_input[input_offset + i * 4]);
}
Expand Down
Loading