nan
2ff7b04f56
- Introduced a new memory management policy for ONNX Runtime to release temporary GPU memory after inference completion. - Established a shared inference entry point for all ONNX models, differentiating between long-lived sessions and temporary memory arenas. - Updated CUDA sessions to utilize a shrinkable memory arena, reclaiming GPU memory post-inference. - Ensured consistent memory release semantics across BEN2, DA3, MoGe-2, and Upscale models while reusing existing sessions. - Added lifecycle tests to validate temporary memory release and session reuse. - Modified model adapters to integrate the new memory management features, ensuring backward compatibility with existing inference behavior. - Enhanced unit tests to cover new memory release scenarios and verify correct behavior across different execution providers.
83 lines
2.2 KiB
Python
83 lines
2.2 KiB
Python
import sys
|
|
import unittest
|
|
from pathlib import Path
|
|
from types import SimpleNamespace
|
|
from unittest.mock import patch
|
|
|
|
import numpy as np
|
|
from PIL import Image
|
|
|
|
|
|
SOURCE_DIR = (
|
|
Path(__file__).resolve().parents[1]
|
|
/ "src"
|
|
/ "anyimage"
|
|
/ "server"
|
|
/ "models"
|
|
)
|
|
sys.path.insert(0, str(SOURCE_DIR))
|
|
|
|
import onnx_ben2
|
|
|
|
|
|
class FakeSession:
|
|
def get_providers(self):
|
|
return ["CPUExecutionProvider"]
|
|
|
|
def get_inputs(self):
|
|
return [SimpleNamespace(name="input")]
|
|
|
|
def get_outputs(self):
|
|
return [SimpleNamespace(name="alpha")]
|
|
|
|
def run(self, output_names, feeds):
|
|
self.output_names = output_names
|
|
self.feeds = feeds
|
|
gradient = np.linspace(0.0, 1.0, 1024, dtype=np.float32)
|
|
return [np.tile(gradient, (1024, 1))[None, None]]
|
|
|
|
|
|
class Ben2OnnxTest(unittest.TestCase):
|
|
def test_infer_forwards_memory_release_boundary(self):
|
|
session = FakeSession()
|
|
|
|
with patch.object(
|
|
onnx_ben2,
|
|
"run_session",
|
|
side_effect=lambda model, names, feeds, **_options: model.run(
|
|
names, feeds
|
|
),
|
|
) as run:
|
|
result = onnx_ben2.infer(
|
|
session,
|
|
Image.new("RGB", (40, 20)),
|
|
release_memory=False,
|
|
)
|
|
|
|
self.assertEqual(result.size, (40, 20))
|
|
self.assertFalse(run.call_args.kwargs["release_memory"])
|
|
|
|
def test_preprocess_matches_model_layout(self):
|
|
image = Image.new("RGB", (40, 20), (64, 128, 192))
|
|
result = onnx_ben2.preprocess(image)
|
|
|
|
self.assertEqual(result.shape, (1, 3, 1024, 1024))
|
|
self.assertEqual(result.dtype, np.float32)
|
|
self.assertTrue(result.flags.c_contiguous)
|
|
|
|
def test_infer_restores_original_size_and_alpha(self):
|
|
session = FakeSession()
|
|
image = Image.new("RGB", (40, 20), (10, 20, 30))
|
|
|
|
result = onnx_ben2.infer(session, image)
|
|
|
|
self.assertEqual(result.mode, "RGBA")
|
|
self.assertEqual(result.size, image.size)
|
|
self.assertEqual(session.feeds["input"].shape, (1, 3, 1024, 1024))
|
|
alpha = np.asarray(result.getchannel("A"))
|
|
self.assertLess(int(alpha[:, 0].max()), int(alpha[:, -1].min()))
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|