Skip to content

Make INT8 activation PRESET_SCHEMES explicit #158

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

Merged
merged 1 commit into from
Sep 24, 2024
Merged

Make INT8 activation PRESET_SCHEMES explicit #158

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 5 additions & 4 deletions src/compressed_tensors/quantization/quant_scheme.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,7 +108,7 @@ def is_preset_scheme(name: str) -> bool:
UNQUANTIZED = dict()

# 8 bit integer weights and 8 bit activations quantization
W8A8 = dict(
INT8_W8A8 = dict(
weights=QuantizationArgs(
num_bits=8,
type=QuantizationType.INT,
Expand Down Expand Up @@ -149,7 +149,7 @@ def is_preset_scheme(name: str) -> bool:
)

# 4 bit integer weights and 8 bit activations quantization
W4A8 = dict(
INT8_W4A8 = dict(
weights=QuantizationArgs(
num_bits=4,
type=QuantizationType.INT,
Expand Down Expand Up @@ -210,8 +210,9 @@ def is_preset_scheme(name: str) -> bool:
"W8A16": W8A16,
"W4A16": W4A16,
# Integer weight and activation schemes
"W8A8": W8A8,
"W4A8": W4A8,
"W8A8": INT8_W8A8,
"INT8": INT8_W8A8, # alias for W8A8
"W4A8": INT8_W4A8,
# Float weight and activation schemes
"FP8": FP8,
"FP8_DYNAMIC": FP8_DYNAMIC,
Expand Down
Loading