4c660dff3c
Tiles are fixed-shape, so fused compilation applies the same as full-frame. Forcing compile=False in the tiled branch left performance on the table: measured on the same 12-frame 2048px green-screen set (alpha output bit-identical, IoU 0.9316): M3 Ultra: 3.64 -> 2.48 s/frame (1.47x) M1 Ultra: 4.65 -> 4.19 s/frame (1.11x) Peak memory unchanged (~2.3 GB). |
||
|---|---|---|
| .. | ||
| convert | ||
| inference | ||
| io | ||
| model | ||
| utils | ||
| __init__.py | ||
| __main__.py | ||
| engine.py | ||
| weights_cli.py | ||
| weights.py | ||