Skip to content

Revert "Add quantize option to the coreml script (#5710)" #5906

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

Closed
wants to merge 1 commit into from
+3 −35
Closed

Revert "Add quantize option to the coreml script (#5710)" #5906

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
38 changes: 3 additions & 35 deletions examples/apple/coreml/scripts/export.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,20 +13,14 @@
import executorch.exir as exir

import torch
from coremltools.optimize.torch.quantization.quantization_config import (
LinearQuantizerConfig,
QuantizationScheme,
)

from executorch.backends.apple.coreml.compiler import CoreMLBackend

from executorch.backends.apple.coreml.partition import CoreMLPartitioner
from executorch.backends.apple.coreml.quantizer import CoreMLQuantizer
from executorch.devtools.etrecord import generate_etrecord
from executorch.exir import to_edge

from executorch.exir.backend.backend_api import to_backend
from torch.ao.quantization.quantize_pt2e import convert_pt2e, prepare_pt2e

from torch.export import export

Expand Down Expand Up @@ -81,13 +75,6 @@ def parse_args() -> argparse.ArgumentParser:
parser.add_argument("--generate_etrecord", action=argparse.BooleanOptionalAction)
parser.add_argument("--save_processed_bytes", action=argparse.BooleanOptionalAction)

parser.add_argument(
"--quantize",
action=argparse.BooleanOptionalAction,
required=False,
help="Quantize CoreML model",
)

args = parser.parse_args()
return args

Expand Down Expand Up @@ -123,10 +110,9 @@ def export_lowered_module_to_executorch_program(lowered_module, example_inputs):
return exec_prog


def save_executorch_program(exec_prog, model_name, compute_unit, quantize):
def save_executorch_program(exec_prog, model_name, compute_unit):
buffer = exec_prog.buffer
data_type = "quantize" if quantize else "fp"
filename = f"{model_name}_coreml_{compute_unit}_{data_type}.pte"
filename = f"{model_name}_coreml_{compute_unit}.pte"
print(f"Saving exported program to {filename}")
with open(filename, "wb") as file:
file.write(buffer)
Expand Down Expand Up @@ -182,22 +168,6 @@ def main():
if args.use_partitioner:
model.eval()
exir_program_aten = torch.export.export(model, example_inputs)
if args.quantize:
quantization_config = LinearQuantizerConfig.from_dict(
{
"global_config": {
"quantization_scheme": QuantizationScheme.affine,
"activation_dtype": torch.quint8,
"weight_dtype": torch.qint8,
"weight_per_channel": True,
}
}
)

quantizer = CoreMLQuantizer(quantization_config)
model = prepare_pt2e(model, quantizer) # pyre-fixme[6]
model(*example_inputs)
exir_program_aten = convert_pt2e(model)

edge_program_manager = exir.to_edge(exir_program_aten)
edge_copy = copy.deepcopy(edge_program_manager)
Expand All @@ -219,9 +189,7 @@ def main():
example_inputs,
)

save_executorch_program(
exec_program, args.model_name, args.compute_unit, args.quantize
)
save_executorch_program(exec_program, args.model_name, args.compute_unit)
generate_etrecord(f"{args.model_name}_coreml_etrecord.bin", edge_copy, exec_program)

if args.save_processed_bytes and lowered_module is not None:
Expand Down
Loading