Skip to content

Revert "Qualcomm AI Engine Direct - Quantizer refine for qat (#6513)" #6722

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
Nov 7, 2024
Merged

Revert "Qualcomm AI Engine Direct - Quantizer refine for qat (#6513)" #6722

Show file tree
Hide file tree
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
6 changes: 3 additions & 3 deletions backends/qualcomm/quantizer/custom_annotation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,12 +6,12 @@
from typing import Sequence

import torch
from executorch.backends.qualcomm.quantizer.annotators import QUANT_ANNOTATION_KEY
from executorch.backends.qualcomm.quantizer.quantizer import (
get_16a8w_qnn_ptq_config,
get_8a8w_qnn_ptq_config,
get_default_8bit_qnn_ptq_config,
QuantizationConfig,
)
from executorch.backends.qualcomm.quantizer.utils import QUANT_ANNOTATION_KEY
from executorch.exir.dialects._ops import ops as exir_ops
from torch.ao.quantization.quantizer import (
QuantizationAnnotation,
Expand Down Expand Up @@ -110,7 +110,7 @@ def annotate_matmul_input1(node: Node, quantization_config: QuantizationConfig):
# Annotate 16a8w for matmul op to get better performance
quantization_config_16a8w = get_16a8w_qnn_ptq_config()
# Annotate 8a8w for second input of matmul until past_kv_cache
quantization_config_8a8w = get_8a8w_qnn_ptq_config(act_symmetric=True)
quantization_config_8a8w = get_default_8bit_qnn_ptq_config(act_symmetric=True)
for node in gm.graph.nodes:
if node.op == "call_function" and node.target == torch.ops.aten.matmul.default:
if "nn_module_stack" in node.meta:
Expand Down
104 changes: 0 additions & 104 deletions backends/qualcomm/quantizer/observers/per_channel_param_observer.py
View file
Open in desktop

This file was deleted.

Loading
Loading