mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-22 15:32:08 +00:00
c3bd8ef445
Co-authored-by: mrs303 <54661797+mrs303@users.noreply.github.com> Co-authored-by: Adam Stachowicz <105052242+astachowiczhabana@users.noreply.github.com> Co-authored-by: Grzegorz Morys <gmorys@habana.ai>
16 lines
569 B
Python
16 lines
569 B
Python
import os
|
|
import sys
|
|
|
|
assert "habana_frameworks" not in sys.modules
|
|
|
|
is_quantization_enabled = os.getenv("QUANT_CONFIG", "") != ""
|
|
|
|
if is_quantization_enabled:
|
|
os.environ.setdefault("ENABLE_EXPERIMENTAL_FLAGS", "true")
|
|
os.environ.setdefault("USE_DEFAULT_QUANT_PARAM", "true")
|
|
os.environ.setdefault("UPDATE_GRAPH_OUTPUT_MME", "false")
|
|
os.environ.setdefault("ENABLE_CALC_DYNAMIC_RANGE", "false")
|
|
os.environ.setdefault(
|
|
"UPDATE_MME_OUTPUT_PRECISION_FILTER", "v_proj,matmul_av")
|
|
os.environ.setdefault("EXPERIMENTAL_WEIGHT_SHARING", "FALSE")
|