Spaces:
Sleeping
Sleeping
File size: 99,195 Bytes
46b3a11 2797bac 46b3a11 1e97e9e 46b3a11 7972153 1e97e9e 46b3a11 2797bac 1e97e9e 7972153 46b3a11 7972153 46b3a11 1e97e9e 2797bac 1e97e9e 46b3a11 7972153 1e97e9e 2797bac 1e97e9e 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 7972153 46b3a11 1e97e9e 2797bac 76c801a 2797bac 76c801a 2797bac 76c801a 2797bac 76c801a 2797bac 76c801a 2797bac 76c801a 2797bac 1e97e9e 7972153 46b3a11 5d4f6a9 46b3a11 1e97e9e 2797bac 46b3a11 1e97e9e 46b3a11 34a90ad 1e97e9e 34a90ad 46b3a11 1e97e9e 46b3a11 34a90ad 1e97e9e 34a90ad 46b3a11 2797bac 1e97e9e 46b3a11 76c801a 46b3a11 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 |
import argparse
import copy
import io
import json
import os
import time
from pathlib import Path
from dataclasses import dataclass
import matplotlib.pyplot as plt
import gradio as gr
import numpy as np
import pandas as pd
import timm
import torch
import torch.nn as nn
import torch.nn.utils.prune as prune
import segmentation_models_pytorch as smp
from PIL import Image, ImageDraw, ImageFont
from torchvision import transforms
from torchvision.models.detection import (
FasterRCNN_ResNet50_FPN_Weights,
SSDLite320_MobileNet_V3_Large_Weights,
fasterrcnn_resnet50_fpn,
ssdlite320_mobilenet_v3_large,
)
try:
from ultralytics import YOLO as UltralyticsYOLO
except ModuleNotFoundError: # pragma: no cover - optional dependency
UltralyticsYOLO = None
try:
import albumentations as A
except ModuleNotFoundError: # pragma: no cover - optional dependency
A = None
# ---------------------------------------------
# Base Model Registry / Defaults
# ---------------------------------------------
MODEL_OPTIONS = [
"resnet50",
"mobilenetv3_large_100",
"efficientnet_b0",
"convnext_tiny",
"vit_base_patch16_224",
"regnety_016",
"efficientnet_lite0",
]
PRETRAINED_DEFAULT = os.getenv("MODEL_OPT_PRETRAINED", "1") == "1"
_PRETRAINED_DISABLED = False
_PRETRAINED_WARNED = False
PRESETS = {
"Edge CPU": {
"device": "cpu",
"channels_last": False,
"compile": False,
"quant": "dynamic",
"prune_amount": 0.3,
},
"Datacenter GPU": {
"device": "cuda",
"channels_last": True,
"compile": True,
"quant": "fp16",
"prune_amount": 0.2,
},
"Apple MPS": {
"device": "mps",
"channels_last": False,
"compile": False,
"quant": "fp16",
"prune_amount": 0.2,
},
}
_MODEL_CACHE: dict[str, torch.nn.Module] = {}
_TRANSFORM_CACHE: dict[str, transforms.Compose] = {}
@dataclass(frozen=True)
class SegmentationModelConfig:
name: str
checkpoint: str
classes: int = 150
dataset: str = "ADE20K"
SEGMENTATION_MODEL_CONFIGS: tuple[SegmentationModelConfig, ...] = (
SegmentationModelConfig("SegFormer B0 (ADE20K 512x512)", "smp-hub/segformer-b0-512x512-ade-160k"),
SegmentationModelConfig("SegFormer B4 (ADE20K 512x512)", "smp-hub/segformer-b4-512x512-ade-160k"),
SegmentationModelConfig("DPT Large (ADE20K)", "smp-hub/dpt-large-ade20k"),
SegmentationModelConfig("UPerNet ConvNeXt-Tiny (ADE20K)", "smp-hub/upernet-convnext-tiny"),
)
SEGMENTATION_MODEL_MAP = {cfg.name: cfg for cfg in SEGMENTATION_MODEL_CONFIGS}
_SEG_BASE_PALETTE = np.array(
[
[0, 0, 0],
[0, 114, 189],
[217, 83, 25],
[237, 177, 32],
[126, 47, 142],
[119, 172, 48],
[77, 190, 238],
[162, 20, 47],
[163, 200, 236],
[255, 127, 14],
[255, 188, 121],
[111, 118, 207],
[204, 121, 167],
[148, 103, 189],
[44, 160, 44],
[23, 190, 207],
[31, 119, 180],
[255, 152, 150],
[214, 39, 40],
[188, 189, 34],
],
dtype=np.uint8,
)
_SEG_MODEL_CACHE: dict[str, torch.nn.Module] = {}
_SEG_TRANSFORM_CACHE: dict[str, object] = {}
_SEG_PALETTE_CACHE: dict[int, np.ndarray] = {}
ADE20K_CLASS_NAMES = [
"wall", "building", "sky", "floor", "tree", "ceiling", "road", "bed", "windowpane", "grass",
"cabinet", "sidewalk", "person", "earth", "door", "table", "mountain", "plant", "curtain", "chair",
"car", "water", "painting", "sofa", "shelf", "house", "sea", "mirror", "rug", "field",
"armchair", "seat", "fence", "desk", "rock", "wardrobe", "lamp", "bathtub", "railing", "cushion",
"base", "box", "column", "signboard", "chest of drawers", "counter", "sand", "sink", "skyscraper", "fireplace",
"refrigerator", "grandstand", "path", "stairs", "runway", "case", "pool table", "pillow", "screen door", "stairway",
"river", "bridge", "bookcase", "blind", "coffee table", "toilet", "flower", "book", "hill", "bench",
"countertop", "stove", "palm", "kitchen island", "computer", "swivel chair", "boat", "bar", "arcade machine", "hovel",
"bus", "towel", "light", "truck", "tower", "chandelier", "awning", "streetlight", "booth", "television receiver",
"airplane", "dirt track", "apparel", "pole", "land", "bannister", "escalator", "ottoman", "bottle", "buffet",
"poster", "stage", "van", "ship", "fountain", "conveyer belt", "canopy", "washer", "plaything", "swimming pool",
"stool", "barrel", "basket", "waterfall", "tent", "bag", "minibike", "cradle", "oven", "ball",
"food", "step", "tank", "trade name", "microwave", "pot", "animal", "bicycle", "lake", "dishwasher",
"screen", "blanket", "sculpture", "hood", "sconce", "vase", "traffic light", "tray", "ashcan", "fan",
"pier", "crt screen", "plate", "monitor", "bulletin board", "shower", "radiator", "glass", "clock", "flag"
]
# ---------------------------------------------
# Object Detection Registry / Defaults
# ---------------------------------------------
DETECTION_MODEL_CONFIGS = {
"Faster R-CNN ResNet50 FPN (COCO)": {
"builder": fasterrcnn_resnet50_fpn,
"weights": FasterRCNN_ResNet50_FPN_Weights.DEFAULT,
"backend": "torchvision",
},
"SSDlite320 MobileNetV3 (COCO)": {
"builder": ssdlite320_mobilenet_v3_large,
"weights": SSDLite320_MobileNet_V3_Large_Weights.DEFAULT,
"backend": "torchvision",
},
}
COCO_CATEGORIES = list(FasterRCNN_ResNet50_FPN_Weights.DEFAULT.meta.get("categories", []))
# Optional YOLOv12 variants (Ultralytics) with size options: n/s/m/l/x
for _size in ("n", "s", "m", "l", "x"):
DETECTION_MODEL_CONFIGS[f"YOLO12-{_size} (COCO)"] = {
"backend": "ultralytics",
"weights": f"yolo12{_size}.pt",
"imgsz": 640,
"categories": COCO_CATEGORIES,
}
DETECTION_MODEL_OPTIONS = list(DETECTION_MODEL_CONFIGS.keys())
_DET_MODEL_CACHE: dict[str, nn.Module] = {}
_DET_TRANSFORM_CACHE: dict[str, object] = {}
_DET_LABELS_CACHE: dict[str, list[str]] = {}
def _require_ultralytics():
if UltralyticsYOLO is None:
raise RuntimeError(
"The 'ultralytics' package is required for YOLO12 models. "
"Install it with `pip install ultralytics` to enable these options."
)
def add_image_label(img: Image.Image, label: str) -> Image.Image:
"""Add a text label at the top of an image."""
img_array = np.array(img)
h, w = img_array.shape[:2]
# Create canvas with extra space at top for label
canvas = np.ones((h + 40, w, 3), dtype=np.uint8) * 255
canvas[40:, :] = img_array
# Convert back to PIL for text drawing
canvas_img = Image.fromarray(canvas)
draw = ImageDraw.Draw(canvas_img)
# Try to use a nice font, fall back to default if not available
try:
font = ImageFont.truetype("/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf", 20)
except:
try:
font = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", 20)
except:
font = ImageFont.load_default()
# Get text size and center it
bbox = draw.textbbox((0, 0), label, font=font)
text_width = bbox[2] - bbox[0]
text_x = (w - text_width) // 2
# Draw text
draw.text((text_x, 10), label, fill=(0, 0, 0), font=font)
return canvas_img
def select_device(device_str: str) -> torch.device:
"""Return a valid torch.device based on user selection."""
device_str = (device_str or "auto").lower()
if device_str == "cuda" and torch.cuda.is_available():
return torch.device("cuda")
if device_str == "mps" and getattr(torch.backends, "mps", None) and torch.backends.mps.is_available():
return torch.device("mps")
if device_str == "cpu":
return torch.device("cpu")
return torch.device("cuda" if torch.cuda.is_available() else "cpu")
def get_transform(model_name: str):
if model_name in _TRANSFORM_CACHE:
return _TRANSFORM_CACHE[model_name]
model = get_fp32_model(model_name)
if hasattr(timm.data, "resolve_model_data_config"):
data_cfg = timm.data.resolve_model_data_config(model)
else:
# Fallback for older timm versions
data_cfg = timm.data.resolve_data_config(model.pretrained_cfg)
_TRANSFORM_CACHE[model_name] = timm.data.create_transform(**data_cfg)
return _TRANSFORM_CACHE[model_name]
def get_fp32_model(model_name: str):
if model_name not in _MODEL_CACHE:
global _PRETRAINED_DISABLED, _PRETRAINED_WARNED
use_pretrained = PRETRAINED_DEFAULT and not _PRETRAINED_DISABLED
if not use_pretrained and not _PRETRAINED_WARNED:
print("INFO: MODEL_OPT_PRETRAINED disabled; using randomly initialized weights.")
_PRETRAINED_WARNED = True
try:
loaded = timm.create_model(model_name, pretrained=use_pretrained)
except Exception as exc:
print(f"Warning: pretrained weights unavailable ({exc}); using random init for {model_name}")
_PRETRAINED_DISABLED = True
loaded = timm.create_model(model_name, pretrained=False)
loaded.eval()
_MODEL_CACHE[model_name] = loaded
return _MODEL_CACHE[model_name]
def clone_model(model_name: str):
"""Create a fresh model loaded from the cached FP32 weights to avoid re-downloads."""
base = get_fp32_model(model_name)
fresh = timm.create_model(model_name, pretrained=False)
fresh.load_state_dict(base.state_dict())
fresh.eval()
return fresh
# ---------------------------------------------
# Segmentation Utilities
# ---------------------------------------------
def _require_albumentations():
if A is None:
raise RuntimeError(
"Albumentations is required for pretrained segmentation models. "
"Install it with `pip install albumentations` or add it to your environment."
)
def get_segmentation_model(config: SegmentationModelConfig) -> nn.Module:
key = config.checkpoint
if key not in _SEG_MODEL_CACHE:
model = smp.from_pretrained(config.checkpoint).eval()
_SEG_MODEL_CACHE[key] = model
return _SEG_MODEL_CACHE[key]
def clone_segmentation_model(config: SegmentationModelConfig) -> nn.Module:
base = get_segmentation_model(config)
fresh = smp.from_pretrained(config.checkpoint).eval()
fresh.load_state_dict(base.state_dict())
return fresh
def get_segmentation_transform(config: SegmentationModelConfig):
key = config.checkpoint
if key in _SEG_TRANSFORM_CACHE:
return _SEG_TRANSFORM_CACHE[key]
_require_albumentations()
try:
preprocessing = A.Compose.from_pretrained(config.checkpoint)
except Exception as exc: # pragma: no cover - depends on network availability
raise RuntimeError(f"Failed to load preprocessing pipeline for {config.checkpoint}: {exc}") from exc
def _transform(image):
if image is None:
raise ValueError("No image provided")
if not isinstance(image, Image.Image):
if isinstance(image, np.ndarray):
array = image
if array.dtype != np.uint8:
array = (np.clip(array, 0, 1) * 255).astype(np.uint8)
image_rgb = Image.fromarray(array)
else:
raise ValueError(f"Unsupported image type: {type(image)}")
else:
image_rgb = image
image_rgb = image_rgb.convert("RGB")
np_image = np.array(image_rgb)
processed = preprocessing(image=np_image)["image"]
if isinstance(processed, torch.Tensor):
processed_np = processed.detach().cpu().numpy()
else:
processed_np = np.asarray(processed, dtype=np.float32)
tensor = torch.from_numpy(processed_np.transpose(2, 0, 1)).float()
return tensor, image_rgb
_SEG_TRANSFORM_CACHE[key] = _transform
return _transform
def get_segmentation_palette(class_count: int) -> np.ndarray:
if class_count in _SEG_PALETTE_CACHE:
return _SEG_PALETTE_CACHE[class_count]
base_len = len(_SEG_BASE_PALETTE)
if class_count <= base_len:
palette = _SEG_BASE_PALETTE[:class_count]
else:
palette = np.zeros((class_count, 3), dtype=np.uint8)
palette[:base_len] = _SEG_BASE_PALETTE
rng = np.random.default_rng(1337)
palette[base_len:] = rng.integers(0, 256, size=(class_count - base_len, 3), endpoint=False, dtype=np.uint8)
palette[:, 0] |= 1 # ensure colors are not pure black except index 0
palette[0] = np.array([0, 0, 0], dtype=np.uint8)
_SEG_PALETTE_CACHE[class_count] = palette
return palette
def colorize_mask(mask: np.ndarray, class_count: int) -> Image.Image:
if mask.ndim != 2:
raise ValueError("Mask must be 2D for colorization")
palette = get_segmentation_palette(class_count)
indexed = np.mod(mask, class_count)
colored = palette[indexed]
return Image.fromarray(colored.astype(np.uint8))
def overlay_mask(image: Image.Image, mask_image: Image.Image, alpha: float = 0.5) -> Image.Image:
base = np.array(image.convert("RGB"), dtype=np.float32)
mask_resized = mask_image.resize(image.size, Image.NEAREST)
mask_arr = np.array(mask_resized, dtype=np.float32)
blended = (1.0 - alpha) * base + alpha * mask_arr
return Image.fromarray(np.clip(blended, 0, 255).astype(np.uint8))
def summarize_mask(mask: np.ndarray, class_count: int) -> list[dict[str, float]]:
flat = mask.reshape(-1)
counts = np.bincount(flat, minlength=class_count)
total = float(flat.size)
summary = []
for idx in range(class_count):
count = int(counts[idx])
percent = (count / total * 100.0) if total else 0.0
summary.append({"index": idx, "count": count, "percent": percent})
return summary
def get_class_labels(config: SegmentationModelConfig) -> list[str]:
# Try to get labels from model metadata first
model = get_segmentation_model(config)
meta = getattr(model, "meta", {}) or {}
dataset_meta = meta.get("dataset", {}) or {}
labels = dataset_meta.get("class_names") or dataset_meta.get("classes_names")
# If not in metadata, use dataset-specific labels
if not labels:
if config.dataset == "ADE20K" and config.classes == 150:
labels = ADE20K_CLASS_NAMES
else:
labels = [f"Class {idx}" for idx in range(config.classes)]
else:
labels = list(labels)
# Ensure we have the right number of labels
if len(labels) < config.classes:
labels.extend(f"Class {len(labels) + i}" for i in range(config.classes - len(labels)))
return labels[: config.classes]
# ---------------------------------------------
# Object Detection Utilities
# ---------------------------------------------
def get_detection_config(model_name: str) -> dict:
if model_name not in DETECTION_MODEL_CONFIGS:
raise ValueError(f"Unknown detection model: {model_name}")
return dict(DETECTION_MODEL_CONFIGS[model_name])
def get_detection_labels(model_name: str) -> list[str]:
if model_name in _DET_LABELS_CACHE:
return _DET_LABELS_CACHE[model_name]
cfg = get_detection_config(model_name)
categories = cfg.get("categories")
if categories:
labels = categories
else:
weights = cfg.get("weights")
labels = weights.meta.get("categories", []) if weights else []
_DET_LABELS_CACHE[model_name] = list(labels)
return _DET_LABELS_CACHE[model_name]
def get_detection_transform(model_name: str):
if model_name in _DET_TRANSFORM_CACHE:
return _DET_TRANSFORM_CACHE[model_name]
cfg = get_detection_config(model_name)
backend = cfg.get("backend", "torchvision")
if backend == "ultralytics":
transform = lambda img: img # Ultralytics handles preprocessing internally
else:
weights = cfg.get("weights")
transform = weights.transforms() if weights else transforms.Compose([transforms.ToTensor()])
_DET_TRANSFORM_CACHE[model_name] = transform
return transform
def get_detection_model(model_name: str) -> nn.Module:
if model_name not in _DET_MODEL_CACHE:
cfg = get_detection_config(model_name)
backend = cfg.get("backend", "torchvision")
if backend == "ultralytics":
_require_ultralytics()
weights = cfg.get("weights")
try:
model = UltralyticsYOLO(weights)
except Exception as exc:
raise RuntimeError(
f"Failed to load YOLO12 weights '{weights}'. Download or place the checkpoint locally first."
) from exc
if hasattr(model, "model"):
model.model.eval()
else:
weights = cfg.get("weights")
try:
model = cfg["builder"](weights=weights)
except Exception as exc:
print(f"Warning: detection weights unavailable ({exc}); using random init for {model_name}")
model = cfg["builder"](weights=None)
model.eval()
_DET_MODEL_CACHE[model_name] = model
return _DET_MODEL_CACHE[model_name]
def clone_detection_model(model_name: str) -> nn.Module:
base = get_detection_model(model_name)
cfg = get_detection_config(model_name)
backend = cfg.get("backend", "torchvision")
if backend == "ultralytics":
_require_ultralytics()
fresh = copy.deepcopy(base)
if hasattr(fresh, "model") and isinstance(fresh.model, nn.Module):
fresh.model.eval()
return fresh
fresh = cfg["builder"](weights=None)
fresh.load_state_dict(base.state_dict())
fresh.eval()
return fresh
def prepare_detection_input(image, transform_fn):
if image is None:
raise ValueError("No image provided")
if not isinstance(image, Image.Image):
if isinstance(image, np.ndarray) and image.dtype != np.uint8:
image = (np.clip(image, 0, 1) * 255).astype(np.uint8)
image = Image.fromarray(np.array(image).astype("uint8"))
image_rgb = image.convert("RGB")
tensor = transform_fn(image_rgb)
if tensor.ndim == 3:
tensor = tensor
else:
tensor = torch.as_tensor(tensor)
return tensor, image_rgb
def draw_detections(image: Image.Image, detections: list[dict], max_dets: int = 30) -> Image.Image:
canvas = image.copy()
draw = ImageDraw.Draw(canvas)
colors = _SEG_BASE_PALETTE # reuse palette for variety
for idx, det in enumerate(detections[:max_dets]):
box = det["box"]
color = tuple(int(c) for c in colors[idx % len(colors)])
draw.rectangle(box, outline=color, width=3)
label = f"{det['label']} {det['score']:.2f}"
draw.text((box[0] + 4, box[1] + 4), label, fill=color)
return canvas
def run_detection_inference(
model: nn.Module,
image,
device: torch.device,
transform_fn,
channels_last: bool,
warmup: bool,
use_amp: bool,
score_thresh: float = 0.25,
backend: str = "torchvision",
imgsz: int | None = None,
):
if backend == "ultralytics":
if image is None:
raise ValueError("No image provided")
if not isinstance(image, Image.Image):
if isinstance(image, np.ndarray) and image.dtype != np.uint8:
image = (np.clip(image, 0, 1) * 255).astype(np.uint8)
image = Image.fromarray(np.array(image).astype("uint8"))
image_rgb = image.convert("RGB")
device_arg = str(device) if isinstance(device, torch.device) else device
half = use_amp and isinstance(device, torch.device) and device.type == "cuda"
if hasattr(model, "model") and isinstance(model.model, nn.Module):
model.model.to(device)
if warmup:
with torch.no_grad():
model.predict(image_rgb, imgsz=imgsz, device=device_arg, verbose=False, half=half)
start = time.time()
with torch.no_grad():
results = model.predict(image_rgb, imgsz=imgsz, device=device_arg, verbose=False, half=half)
latency = (time.time() - start) * 1000
dets: list[dict] = []
if results:
res = results[0]
boxes = getattr(res, "boxes", None)
if boxes is not None:
xyxy = boxes.xyxy.detach().cpu().numpy()
confs = boxes.conf.detach().cpu().numpy()
labels = boxes.cls.detach().cpu().numpy()
for box, score, label_idx in zip(xyxy, confs, labels):
if score < score_thresh:
continue
dets.append(
{
"label": str(int(label_idx)),
"score": float(score),
"box": [float(x) for x in box],
}
)
return {"detections": dets, "latency": latency, "image": image_rgb}
tensor, image_rgb = prepare_detection_input(image, transform_fn)
model = model.to(device)
batch_tensor = tensor.to(device)
if channels_last and device.type == "cuda" and batch_tensor.dim() == 4:
batch_tensor = batch_tensor.to(memory_format=torch.channels_last)
elif channels_last and device.type == "cuda":
# Channels-last requires NCHW (4D) input; detection tensors are 3D.
pass
if next(model.parameters()).dtype == torch.float16:
batch_tensor = batch_tensor.half()
inputs = [batch_tensor]
if warmup:
with torch.no_grad():
model(inputs)
amp_ctx = torch.cuda.amp.autocast(enabled=use_amp and device.type == "cuda")
start = time.time()
with torch.no_grad(), amp_ctx:
outputs = model(inputs)
latency = (time.time() - start) * 1000
out = outputs[0]
boxes = out["boxes"].detach().cpu().numpy()
scores = out["scores"].detach().cpu().numpy()
labels = out["labels"].detach().cpu().numpy()
dets = []
for box, score, label_idx in zip(boxes, scores, labels):
if score < score_thresh:
continue
dets.append(
{
"label": str(label_idx),
"score": float(score),
"box": [float(x) for x in box],
}
)
return {
"detections": dets,
"latency": latency,
"image": image_rgb,
}
def attach_detection_labels(detections: list[dict], label_names: list[str]) -> list[dict]:
labeled = []
for det in detections:
idx = int(det["label"])
name = label_names[idx] if idx < len(label_names) else f"Class {idx}"
labeled.append({**det, "label": name})
return labeled
def get_detection_state_module(model, backend: str):
if backend == "ultralytics" and hasattr(model, "model"):
return model.model
return model
def build_detection_metrics(
original_result: dict,
optimized_result: dict,
size_original: float,
size_optimized: float,
optimized_label: str,
score_thresh: float,
):
orig_dets = original_result["detections"]
opt_dets = optimized_result["detections"]
mean_score_orig = float(np.mean([d["score"] for d in orig_dets])) if orig_dets else 0.0
mean_score_opt = float(np.mean([d["score"] for d in opt_dets])) if opt_dets else 0.0
metrics_df = pd.DataFrame(
{
"Metric": [
"Latency (ms)",
f"Detections (score>={score_thresh})",
"Mean Score",
"Model Size (MB)",
],
"Original Model": [
f"{original_result['latency']:.2f}",
str(len(orig_dets)),
f"{mean_score_orig:.3f}",
f"{size_original:.2f}",
],
optimized_label: [
f"{optimized_result['latency']:.2f}",
str(len(opt_dets)),
f"{mean_score_opt:.3f}",
f"{size_optimized:.2f}",
],
}
)
return metrics_df
def build_detection_comparison_df(
orig_dets: list[dict],
opt_dets: list[dict],
optimized_label: str,
max_rows: int = 50,
) -> pd.DataFrame:
rows = []
for det in orig_dets:
rows.append(
{
"Model": "Original",
"Class": det["label"],
"Score": round(det["score"], 3),
"Box [x1,y1,x2,y2]": [round(x, 1) for x in det["box"]],
}
)
for det in opt_dets:
rows.append(
{
"Model": optimized_label,
"Class": det["label"],
"Score": round(det["score"], 3),
"Box [x1,y1,x2,y2]": [round(x, 1) for x in det["box"]],
}
)
if max_rows and len(rows) > max_rows:
rows = rows[:max_rows]
return pd.DataFrame(rows)
def run_segmentation_inference(
model: nn.Module,
image,
device: torch.device,
transform_fn,
channels_last: bool,
warmup: bool,
use_amp: bool,
class_count: int,
):
tensor, original_image = transform_fn(image)
model = model.to(device)
input_tensor = tensor.unsqueeze(0).to(device)
if channels_last and device.type == "cuda":
input_tensor = input_tensor.to(memory_format=torch.channels_last)
if next(model.parameters()).dtype == torch.float16:
input_tensor = input_tensor.half()
if warmup:
with torch.no_grad():
model(input_tensor)
amp_ctx = torch.cuda.amp.autocast(enabled=use_amp and device.type == "cuda")
start = time.time()
with torch.no_grad(), amp_ctx:
logits = model(input_tensor)
latency = (time.time() - start) * 1000
if isinstance(logits, (list, tuple)):
logits = logits[0]
logits = logits.detach().cpu()
probs = torch.softmax(logits, dim=1)
mask_tensor = torch.argmax(probs, dim=1)[0]
mask_processed = mask_tensor.cpu().numpy().astype(np.int64)
mean_conf = float(probs.max(dim=1)[0].mean().item())
mask_processed_image = colorize_mask(mask_processed, class_count)
mask_original_l = Image.fromarray(mask_processed.astype(np.uint8), mode="L").resize(original_image.size, Image.NEAREST)
mask_original_np = np.array(mask_original_l, dtype=np.int64)
mask_original_image = colorize_mask(mask_original_np, class_count)
overlay_original = overlay_mask(original_image, mask_original_image)
class_summary = summarize_mask(mask_original_np, class_count)
return {
"latency": latency,
"mask_processed": mask_processed,
"mask_original": mask_original_np,
"mask_image_processed": mask_processed_image,
"mask_image_original": mask_original_image,
"overlay_original": overlay_original,
"mean_confidence": mean_conf,
"class_summary": class_summary,
}
def build_segmentation_metrics(
original_result: dict,
optimized_result: dict,
size_original: float,
size_optimized: float,
optimized_label: str,
) -> pd.DataFrame:
mask_original = original_result["mask_original"]
mask_optimized = optimized_result["mask_original"]
agreement = float((mask_original == mask_optimized).mean() * 100.0)
metrics_df = pd.DataFrame(
{
"Metric": [
"Latency (ms)",
"Mean Confidence",
"Model Size (MB)",
"Mask Agreement (%)",
],
"Original Model": [
f"{original_result['latency']:.2f}",
f"{original_result['mean_confidence']:.4f}",
f"{size_original:.2f}",
"100.00",
],
optimized_label: [
f"{optimized_result['latency']:.2f}",
f"{optimized_result['mean_confidence']:.4f}",
f"{size_optimized:.2f}",
f"{agreement:.2f}",
],
}
)
return metrics_df
def build_class_distribution_df(
original_summary: list[dict[str, float]],
optimized_summary: list[dict[str, float]],
labels: list[str],
optimized_label: str,
max_rows: int = 25,
) -> pd.DataFrame:
rows = []
for idx, label in enumerate(labels):
orig_entry = original_summary[idx]
opt_entry = optimized_summary[idx]
if orig_entry["count"] == 0 and opt_entry["count"] == 0:
continue
rows.append(
{
"Class": label,
"Original %": round(orig_entry["percent"], 2),
f"{optimized_label} %": round(opt_entry["percent"], 2),
"Original Pixels": orig_entry["count"],
f"{optimized_label} Pixels": opt_entry["count"],
}
)
rows.sort(key=lambda item: max(item["Original %"], item[f"{optimized_label} %"]), reverse=True)
if max_rows and len(rows) > max_rows:
rows = rows[:max_rows]
return pd.DataFrame(rows)
# ---------------------------------------------
# Image Preprocess
# ---------------------------------------------
imagenet_info = timm.data.ImageNetInfo()
labels = [imagenet_info.index_to_description(i) for i in range(1000)]
# ---------------------------------------------
# PRUNING FUNCTION (dynamic)
# ---------------------------------------------
def apply_pruning(model, amount=0.5, method="unstructured"):
model = model.eval()
for module in model.modules():
if isinstance(module, nn.Conv2d):
if method == "unstructured":
prune.l1_unstructured(module, name="weight", amount=amount)
elif method == "structured":
prune.ln_structured(module, name="weight", amount=amount, n=2, dim=0)
# Make pruning permanent to reflect true sparsity/size
for module in model.modules():
if isinstance(module, nn.Conv2d) and hasattr(module, "weight_orig"):
prune.remove(module, "weight")
return model
# ---------------------------------------------
# QUANTIZATION FUNCTION (dynamic)
# ---------------------------------------------
def apply_quantization(model, q_type="dynamic"):
q_type = q_type or "dynamic"
if q_type in {"dynamic", "weight_only"}: # dynamic quantization is weight-only by default
return torch.ao.quantization.quantize_dynamic(model, {nn.Linear}, dtype=torch.qint8)
if q_type == "fp16":
return model.half().eval()
return model
def compute_sparsity(model: nn.Module) -> pd.DataFrame:
rows = []
for name, module in model.named_modules():
if hasattr(module, "weight"):
weight = module.weight.detach()
total = weight.numel()
zeros = (weight == 0).sum().item()
sparsity = 100.0 * zeros / max(total, 1)
rows.append({"Layer": name, "Params": total, "Sparsity %": round(sparsity, 2)})
return pd.DataFrame(rows)
def maybe_compile(model, use_compile: bool):
if not use_compile:
return model
if not hasattr(torch, "compile"):
return model
try:
return torch.compile(model)
except Exception:
return model
def get_state_dict_size_mb(model: nn.Module) -> float:
buffer = io.BytesIO()
torch.save(model.state_dict(), buffer)
return len(buffer.getbuffer()) / 1e6
def prepare_image(image, transform_fn):
if image is None:
raise ValueError("No image provided")
if not isinstance(image, Image.Image):
if isinstance(image, np.ndarray) and image.dtype != np.uint8:
image = (np.clip(image, 0, 1) * 255).astype(np.uint8)
image = Image.fromarray(image.astype("uint8"))
image = image.convert("RGB")
tensor = transform_fn(image).unsqueeze(0)
return tensor
# ---------------------------------------------
# Inference Function (shared)
# ---------------------------------------------
def run_inference(model, image, device, transform_fn, channels_last=False, warmup=False, use_amp=False):
print(f" run_inference called, image type: {type(image)}")
img = prepare_image(image, transform_fn)
model = model.to(device)
img = img.to(device)
if channels_last and device.type == "cuda":
img = img.to(memory_format=torch.channels_last)
if next(model.parameters()).dtype == torch.float16:
img = img.half()
if warmup:
with torch.no_grad():
model(img)
print(" Running model inference...")
start = time.time()
amp_ctx = torch.cuda.amp.autocast(enabled=use_amp and device.type == "cuda")
with torch.no_grad(), amp_ctx:
out = model(img)
latency = (time.time() - start) * 1000
out_cpu = out.detach().cpu()
prob = torch.softmax(out_cpu, dim=1)[0]
top5_prob, top5_idx = torch.topk(prob, 5)
results = [(labels[i], float(top5_prob[j])) for j, i in enumerate(top5_idx)]
print(f" Inference complete. Top prediction: {results[0][0]}")
return results, latency
def build_top5_plot(results_orig, results_other, other_label: str):
classes = []
for r in results_orig + results_other:
if r[0] not in classes:
classes.append(r[0])
orig_map = {r[0]: r[1] for r in results_orig}
other_map = {r[0]: r[1] for r in results_other}
orig_vals = [orig_map.get(c, 0.0) for c in classes]
other_vals = [other_map.get(c, 0.0) for c in classes]
x = np.arange(len(classes))
width = 0.35
fig, ax = plt.subplots(figsize=(8, 4))
ax.bar(x - width / 2, orig_vals, width, label="Original")
ax.bar(x + width / 2, other_vals, width, label=other_label)
ax.set_xticks(x)
ax.set_xticklabels(classes, rotation=20, ha="right")
ax.set_ylabel("Confidence")
ax.set_xlabel("Class")
ax.legend()
fig.tight_layout()
return fig
# ---------------------------------------------
# Gradio Functions With Options
# ---------------------------------------------
def run_pruned(
img,
model_name,
method,
amount,
device_choice="auto",
channels_last=False,
use_compile=False,
use_amp=False,
export_ts=False,
export_onnx=False,
export_report=False,
export_state=True,
preset=None,
):
print("\n=== RUN PRUNED CALLED ===")
print(f"Image type: {type(img)}, Model: {model_name}, Method: {method}, Amount: {amount}")
if img is None:
print("ERROR: Image is None")
return (
{"Metric": ["Error"], "Original Model": ["No image uploaded"], "Pruned Model": [""]},
{"Class": [], "Original": [], "Pruned": []},
pd.DataFrame(),
[]
)
if preset in PRESETS:
preset_cfg = PRESETS[preset]
device_choice = preset_cfg["device"]
channels_last = preset_cfg["channels_last"]
use_compile = preset_cfg["compile"]
use_amp = preset_cfg.get("amp", use_amp)
amount = preset_cfg.get("prune_amount", amount)
device = select_device(device_choice)
transform_fn = get_transform(model_name)
# Run original model
print("Running original model...")
fp32_model = get_fp32_model(model_name)
results_orig, latency_orig = run_inference(fp32_model, img, device, transform_fn, channels_last, warmup=True, use_amp=use_amp)
print(f"Original model done. Latency: {latency_orig:.2f}ms")
# Run pruned model
print("Creating fresh model...")
fresh_model = clone_model(model_name)
print("Applying pruning...")
pruned_model = apply_pruning(fresh_model, amount=float(amount), method=method)
pruned_model = maybe_compile(pruned_model, use_compile)
print("Running pruned model...")
results_pruned, latency_pruned = run_inference(pruned_model, img, device, transform_fn, channels_last, warmup=True, use_amp=use_amp)
print(f"Pruned model done. Latency: {latency_pruned:.2f}ms")
# Model sizes (in-memory)
size_orig = get_state_dict_size_mb(fp32_model)
size_pruned = get_state_dict_size_mb(pruned_model)
print(f"Model sizes - Original: {size_orig:.2f}MB, Pruned: {size_pruned:.2f}MB")
# Comparison metrics - as DataFrame for Gradio
metrics_df = pd.DataFrame({
"Metric": ["Top-1 Prediction", "Confidence", "Latency (ms)", "Model Size (MB)"],
"Original Model": [
results_orig[0][0],
f"{results_orig[0][1]:.4f}",
f"{latency_orig:.2f}",
f"{size_orig:.2f}"
],
"Pruned Model": [
results_pruned[0][0],
f"{results_pruned[0][1]:.4f}",
f"{latency_pruned:.2f}",
f"{size_pruned:.2f}"
]
})
chart_fig = build_top5_plot(results_orig, results_pruned, "Pruned")
sparsity_df = compute_sparsity(pruned_model.cpu())
downloads = []
export_dir = Path("exports")
export_dir.mkdir(exist_ok=True)
sample_cpu = prepare_image(img, transform_fn)
if export_report:
report_path = export_dir / "pruned_report.json"
report = {
"model": model_name,
"pruning": {"method": method, "amount": float(amount)},
"metrics": metrics_df.to_dict(),
"top5_pruned": results_pruned,
"top5_original": results_orig,
}
report_path.write_text(json.dumps(report, indent=2))
downloads.append(str(report_path))
# Always allow state_dict download for reproducibility
if export_state:
state_path = export_dir / "pruned_state_dict.pth"
torch.save(pruned_model.state_dict(), state_path)
downloads.append(str(state_path))
if export_ts:
ts_path = export_dir / "pruned_model.ts"
try:
scripted = torch.jit.trace(pruned_model.cpu(), sample_cpu)
scripted.save(ts_path)
downloads.append(str(ts_path))
except Exception as exc:
print(f"TorchScript export failed: {exc}")
if export_onnx:
onnx_path = export_dir / "pruned_model.onnx"
try:
torch.onnx.export(
pruned_model.cpu(),
sample_cpu,
onnx_path,
input_names=["input"],
output_names=["output"],
opset_version=13,
dynamic_axes={"input": {0: "batch"}, "output": {0: "batch"}},
)
downloads.append(str(onnx_path))
except Exception as exc:
print(f"ONNX export failed: {exc}")
print("=== RUN PRUNED COMPLETE ===")
return metrics_df, chart_fig, sparsity_df, downloads
def run_quantized(
img,
model_name,
q_type,
device_choice="auto",
channels_last=False,
use_compile=False,
use_amp=False,
export_ts=False,
export_onnx=False,
export_report=False,
export_state=True,
preset=None,
):
print("\n=== RUN QUANTIZED CALLED ===")
print(f"Image type: {type(img)}, Model: {model_name}, Q-type: {q_type}")
if img is None:
print("ERROR: Image is None")
return (
{"Metric": ["Error"], "Original Model": ["No image uploaded"], "Quantized Model": [""]},
{"Class": [], "Original": [], "Quantized": []},
[]
)
if preset in PRESETS:
preset_cfg = PRESETS[preset]
device_choice = preset_cfg["device"]
channels_last = preset_cfg["channels_last"]
use_compile = preset_cfg["compile"]
use_amp = preset_cfg.get("amp", use_amp)
q_type = preset_cfg.get("quant", q_type)
device = select_device(device_choice)
if q_type in {"dynamic", "weight_only"} and device.type != "cpu":
print("Dynamic/weight-only quantization uses CPU kernels; switching device to CPU.")
device = torch.device("cpu")
channels_last = False
use_amp = False
transform_fn = get_transform(model_name)
# Run original model
print("Running original model...")
fp32_model = get_fp32_model(model_name)
results_orig, latency_orig = run_inference(fp32_model, img, device, transform_fn, channels_last, warmup=True, use_amp=use_amp)
print(f"Original model done. Latency: {latency_orig:.2f}ms")
# Run quantized model
fresh_model = clone_model(model_name)
quant_model = apply_quantization(fresh_model, q_type)
quant_model = maybe_compile(quant_model, use_compile)
results_quant, latency_quant = run_inference(quant_model, img, device, transform_fn, channels_last, warmup=True, use_amp=use_amp)
size_orig = get_state_dict_size_mb(fp32_model)
size_quant = get_state_dict_size_mb(quant_model)
metrics_df = pd.DataFrame({
"Metric": ["Top-1 Prediction", "Confidence", "Latency (ms)", "Model Size (MB)"],
"Original Model": [
results_orig[0][0],
f"{results_orig[0][1]:.4f}",
f"{latency_orig:.2f}",
f"{size_orig:.2f}"
],
"Quantized Model": [
results_quant[0][0],
f"{results_quant[0][1]:.4f}",
f"{latency_quant:.2f}",
f"{size_quant:.2f}"
]
})
chart_fig = build_top5_plot(results_orig, results_quant, "Quantized")
downloads = []
export_dir = Path("exports")
export_dir.mkdir(exist_ok=True)
sample_cpu = prepare_image(img, transform_fn)
if export_report:
report_path = export_dir / "quant_report.json"
report = {
"model": model_name,
"quantization": q_type,
"metrics": metrics_df.to_dict(),
"top5_quantized": results_quant,
"top5_original": results_orig,
}
report_path.write_text(json.dumps(report, indent=2))
downloads.append(str(report_path))
if export_state:
state_path = export_dir / "quantized_state_dict.pth"
torch.save(quant_model.state_dict(), state_path)
downloads.append(str(state_path))
if export_ts:
ts_path = export_dir / "quantized_model.ts"
try:
scripted = torch.jit.trace(quant_model.cpu(), sample_cpu)
scripted.save(ts_path)
downloads.append(str(ts_path))
except Exception as exc:
print(f"TorchScript export failed: {exc}")
if export_onnx:
onnx_path = export_dir / "quantized_model.onnx"
try:
torch.onnx.export(
quant_model.cpu(),
sample_cpu,
onnx_path,
input_names=["input"],
output_names=["output"],
opset_version=13,
dynamic_axes={"input": {0: "batch"}, "output": {0: "batch"}},
)
downloads.append(str(onnx_path))
except Exception as exc:
print(f"ONNX export failed: {exc}")
print("=== RUN QUANTIZED COMPLETE ===")
return metrics_df, chart_fig, downloads
def run_pruned_detection(
img,
model_choice,
method,
amount,
device_choice="auto",
channels_last=False,
use_compile=False,
use_amp=False,
export_ts=False,
export_onnx=False,
export_report=False,
export_state=True,
preset=None,
score_thresh=0.25,
):
print("\n=== RUN DETECTION PRUNED CALLED ===")
if img is None:
print("ERROR: Image is None")
empty_metrics = pd.DataFrame({"Metric": ["Error"], "Original Model": ["No image"], "Pruned Model": [""]})
return empty_metrics, None, pd.DataFrame(), []
if preset in PRESETS:
preset_cfg = PRESETS[preset]
device_choice = preset_cfg["device"]
channels_last = preset_cfg["channels_last"]
use_compile = preset_cfg["compile"]
use_amp = preset_cfg.get("amp", use_amp)
amount = preset_cfg.get("prune_amount", amount)
device = select_device(device_choice)
cfg = get_detection_config(model_choice)
backend = cfg.get("backend", "torchvision")
imgsz = cfg.get("imgsz")
labels = get_detection_labels(model_choice)
transform_fn = get_detection_transform(model_choice)
base_model = get_detection_model(model_choice)
original_result = run_detection_inference(
base_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
score_thresh=score_thresh,
backend=backend,
imgsz=imgsz,
)
original_result["detections"] = attach_detection_labels(original_result["detections"], labels)
fresh_model = clone_detection_model(model_choice)
pruned_module = apply_pruning(get_detection_state_module(fresh_model, backend), amount=float(amount), method=method)
pruned_module = maybe_compile(pruned_module, use_compile)
if backend == "ultralytics" and hasattr(fresh_model, "model"):
fresh_model.model = pruned_module
pruned_model = fresh_model
else:
pruned_model = pruned_module
pruned_result = run_detection_inference(
pruned_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
score_thresh=score_thresh,
backend=backend,
imgsz=imgsz,
)
pruned_result["detections"] = attach_detection_labels(pruned_result["detections"], labels)
size_orig = get_state_dict_size_mb(get_detection_state_module(base_model, backend))
size_pruned = get_state_dict_size_mb(get_detection_state_module(pruned_model, backend))
metrics_df = build_detection_metrics(
original_result, pruned_result, size_orig, size_pruned, "Pruned Model", score_thresh
)
det_df = build_detection_comparison_df(original_result["detections"], pruned_result["detections"], "Pruned")
overlay_orig = add_image_label(
draw_detections(original_result["image"], original_result["detections"]),
"Original Model",
)
overlay_pruned = add_image_label(
draw_detections(pruned_result["image"], pruned_result["detections"]),
"Pruned Model",
)
overlay_slider_value = (
overlay_orig,
overlay_pruned,
)
downloads: list[str] = []
export_dir = Path("exports")
export_dir.mkdir(exist_ok=True)
trace_inputs = None
if backend != "ultralytics":
sample_tensor, _ = prepare_detection_input(img, transform_fn)
sample_batch = [sample_tensor]
trace_inputs = (sample_batch,)
else:
if export_ts or export_onnx:
print("TorchScript/ONNX export is not enabled for YOLO12 models in this app.")
export_ts = False
export_onnx = False
if export_report:
report_path = export_dir / "pruned_det_report.json"
report = {
"model": model_choice,
"pruning": {"method": method, "amount": float(amount)},
"score_threshold": score_thresh,
"metrics": metrics_df.to_dict(),
"detections": {
"original": original_result["detections"],
"pruned": pruned_result["detections"],
},
}
report_path.write_text(json.dumps(report, indent=2))
downloads.append(str(report_path))
if export_state:
state_path = export_dir / "pruned_det_state_dict.pth"
torch.save(get_detection_state_module(pruned_model, backend).state_dict(), state_path)
downloads.append(str(state_path))
if export_ts and trace_inputs is not None:
ts_path = export_dir / "pruned_det_model.ts"
try:
scripted = torch.jit.trace(pruned_model.cpu(), trace_inputs)
scripted.save(ts_path)
downloads.append(str(ts_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"TorchScript export failed: {exc}")
if export_onnx and trace_inputs is not None:
onnx_path = export_dir / "pruned_det_model.onnx"
try:
torch.onnx.export(
pruned_model.cpu(),
trace_inputs,
onnx_path,
input_names=["images"],
output_names=["detections"],
opset_version=13,
dynamic_axes={"images": {0: "batch", 2: "height", 3: "width"}},
)
downloads.append(str(onnx_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"ONNX export failed: {exc}")
print("=== RUN DETECTION PRUNED COMPLETE ===")
return metrics_df, overlay_slider_value, det_df, downloads
def run_quantized_detection(
img,
model_choice,
q_type,
device_choice="auto",
channels_last=False,
use_compile=False,
use_amp=False,
export_ts=False,
export_onnx=False,
export_report=False,
export_state=True,
preset=None,
score_thresh=0.25,
):
print("\n=== RUN DETECTION QUANTIZED CALLED ===")
if img is None:
print("ERROR: Image is None")
empty_metrics = pd.DataFrame({"Metric": ["Error"], "Original Model": ["No image"], "Quantized Model": [""]})
return empty_metrics, None, pd.DataFrame(), []
if preset in PRESETS:
preset_cfg = PRESETS[preset]
device_choice = preset_cfg["device"]
channels_last = preset_cfg["channels_last"]
use_compile = preset_cfg["compile"]
use_amp = preset_cfg.get("amp", use_amp)
q_type = preset_cfg.get("quant", q_type)
device = select_device(device_choice)
if q_type in {"dynamic", "weight_only"} and device.type != "cpu":
print("Dynamic/weight-only quantization uses CPU kernels; switching device to CPU.")
device = torch.device("cpu")
channels_last = False
use_amp = False
cfg = get_detection_config(model_choice)
backend = cfg.get("backend", "torchvision")
imgsz = cfg.get("imgsz")
labels = get_detection_labels(model_choice)
transform_fn = get_detection_transform(model_choice)
base_model = get_detection_model(model_choice)
original_result = run_detection_inference(
base_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
score_thresh=score_thresh,
backend=backend,
imgsz=imgsz,
)
original_result["detections"] = attach_detection_labels(original_result["detections"], labels)
fresh_model = clone_detection_model(model_choice)
quant_module = apply_quantization(get_detection_state_module(fresh_model, backend), q_type)
quant_module = maybe_compile(quant_module, use_compile)
if backend == "ultralytics" and hasattr(fresh_model, "model"):
fresh_model.model = quant_module
quant_model = fresh_model
else:
quant_model = quant_module
quant_result = run_detection_inference(
quant_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
score_thresh=score_thresh,
backend=backend,
imgsz=imgsz,
)
quant_result["detections"] = attach_detection_labels(quant_result["detections"], labels)
size_orig = get_state_dict_size_mb(get_detection_state_module(base_model, backend))
size_quant = get_state_dict_size_mb(get_detection_state_module(quant_model, backend))
metrics_df = build_detection_metrics(
original_result, quant_result, size_orig, size_quant, "Quantized Model", score_thresh
)
det_df = build_detection_comparison_df(original_result["detections"], quant_result["detections"], "Quantized")
overlay_orig = add_image_label(
draw_detections(original_result["image"], original_result["detections"]),
"Original Model",
)
overlay_quant = add_image_label(
draw_detections(quant_result["image"], quant_result["detections"]),
"Quantized Model",
)
overlay_slider_value = (
overlay_orig,
overlay_quant,
)
downloads: list[str] = []
export_dir = Path("exports")
export_dir.mkdir(exist_ok=True)
trace_inputs = None
if backend != "ultralytics":
sample_tensor, _ = prepare_detection_input(img, transform_fn)
sample_batch = [sample_tensor]
trace_inputs = (sample_batch,)
else:
if export_ts or export_onnx:
print("TorchScript/ONNX export is not enabled for YOLO12 models in this app.")
export_ts = False
export_onnx = False
if export_report:
report_path = export_dir / "quant_det_report.json"
report = {
"model": model_choice,
"quantization": q_type,
"score_threshold": score_thresh,
"metrics": metrics_df.to_dict(),
"detections": {
"original": original_result["detections"],
"quantized": quant_result["detections"],
},
}
report_path.write_text(json.dumps(report, indent=2))
downloads.append(str(report_path))
if export_state:
state_path = export_dir / "quant_det_state_dict.pth"
torch.save(get_detection_state_module(quant_model, backend).state_dict(), state_path)
downloads.append(str(state_path))
if export_ts and trace_inputs is not None:
ts_path = export_dir / "quant_det_model.ts"
try:
scripted = torch.jit.trace(quant_model.cpu(), trace_inputs)
scripted.save(ts_path)
downloads.append(str(ts_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"TorchScript export failed: {exc}")
if export_onnx and trace_inputs is not None:
onnx_path = export_dir / "quant_det_model.onnx"
try:
torch.onnx.export(
quant_model.cpu(),
trace_inputs,
onnx_path,
input_names=["images"],
output_names=["detections"],
opset_version=13,
dynamic_axes={"images": {0: "batch", 2: "height", 3: "width"}},
)
downloads.append(str(onnx_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"ONNX export failed: {exc}")
print("=== RUN DETECTION QUANTIZED COMPLETE ===")
return metrics_df, overlay_slider_value, det_df, downloads
def run_pruned_segmentation(
img,
model_choice,
method,
amount,
device_choice="auto",
channels_last=False,
use_compile=False,
use_amp=False,
export_ts=False,
export_onnx=False,
export_report=False,
export_state=True,
preset=None,
):
print("\n=== RUN SEGMENTATION PRUNED CALLED ===")
if img is None:
print("ERROR: Image is None")
empty_metrics = pd.DataFrame({"Metric": ["Error"], "Original Model": ["No image"], "Pruned Model": [""]})
empty_dist = pd.DataFrame({"Class": [], "Original %": [], "Pruned %": []})
return empty_metrics, empty_dist, None, None, pd.DataFrame(), []
config = SEGMENTATION_MODEL_MAP.get(model_choice, SEGMENTATION_MODEL_CONFIGS[0])
if preset in PRESETS:
preset_cfg = PRESETS[preset]
device_choice = preset_cfg["device"]
channels_last = preset_cfg["channels_last"]
use_compile = preset_cfg["compile"]
use_amp = preset_cfg.get("amp", use_amp)
amount = preset_cfg.get("prune_amount", amount)
device = select_device(device_choice)
base_model = get_segmentation_model(config)
transform_fn = get_segmentation_transform(config)
class_labels = get_class_labels(config)
class_count = config.classes
original_result = run_segmentation_inference(
base_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
class_count=class_count,
)
fresh_model = clone_segmentation_model(config)
pruned_model = apply_pruning(fresh_model, amount=float(amount), method=method)
pruned_model = maybe_compile(pruned_model, use_compile)
pruned_result = run_segmentation_inference(
pruned_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
class_count=class_count,
)
size_orig = get_state_dict_size_mb(base_model)
size_pruned = get_state_dict_size_mb(pruned_model)
metrics_df = build_segmentation_metrics(original_result, pruned_result, size_orig, size_pruned, "Pruned Model")
class_df = build_class_distribution_df(
original_result["class_summary"],
pruned_result["class_summary"],
class_labels,
"Pruned",
)
# Add labels to images for slider comparison
overlay_orig_labeled = add_image_label(original_result["overlay_original"], "Original Model")
overlay_pruned_labeled = add_image_label(pruned_result["overlay_original"], "Pruned Model")
mask_orig_labeled = add_image_label(original_result["mask_image_original"], "Original Mask")
mask_pruned_labeled = add_image_label(pruned_result["mask_image_original"], "Pruned Mask")
overlay_slider_value = (
overlay_orig_labeled,
overlay_pruned_labeled,
)
mask_slider_value = (
mask_orig_labeled,
mask_pruned_labeled,
)
sparsity_df = compute_sparsity(pruned_model.cpu())
downloads: list[str] = []
export_dir = Path("exports")
export_dir.mkdir(exist_ok=True)
if export_report:
report_path = export_dir / "pruned_seg_report.json"
report = {
"model": config.name,
"checkpoint": config.checkpoint,
"dataset": config.dataset,
"pruning": {"method": method, "amount": float(amount)},
"metrics": metrics_df.to_dict(),
"class_distribution": class_df.to_dict(),
}
report_path.write_text(json.dumps(report, indent=2))
downloads.append(str(report_path))
if export_state:
state_path = export_dir / "pruned_seg_state_dict.pth"
torch.save(pruned_model.state_dict(), state_path)
downloads.append(str(state_path))
sample_tensor, _ = transform_fn(img)
sample_batch = sample_tensor.unsqueeze(0)
if export_ts:
ts_path = export_dir / "pruned_seg_model.ts"
try:
scripted = torch.jit.trace(pruned_model.cpu(), sample_batch)
scripted.save(ts_path)
downloads.append(str(ts_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"TorchScript export failed: {exc}")
if export_onnx:
onnx_path = export_dir / "pruned_seg_model.onnx"
try:
torch.onnx.export(
pruned_model.cpu(),
sample_batch,
onnx_path,
input_names=["input"],
output_names=["mask"],
opset_version=13,
dynamic_axes={"input": {0: "batch"}, "mask": {0: "batch"}},
)
downloads.append(str(onnx_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"ONNX export failed: {exc}")
return (
metrics_df,
class_df,
overlay_slider_value,
mask_slider_value,
sparsity_df,
downloads,
)
def run_quantized_segmentation(
img,
model_choice,
q_type,
device_choice="auto",
channels_last=False,
use_compile=False,
use_amp=False,
export_ts=False,
export_onnx=False,
export_report=False,
export_state=True,
preset=None,
):
print("\n=== RUN SEGMENTATION QUANTIZED CALLED ===")
if img is None:
print("ERROR: Image is None")
empty_metrics = pd.DataFrame({"Metric": ["Error"], "Original Model": ["No image"], "Quantized Model": [""]})
empty_dist = pd.DataFrame({"Class": [], "Original %": [], "Quantized %": []})
return empty_metrics, empty_dist, None, None, []
config = SEGMENTATION_MODEL_MAP.get(model_choice, SEGMENTATION_MODEL_CONFIGS[0])
if preset in PRESETS:
preset_cfg = PRESETS[preset]
device_choice = preset_cfg["device"]
channels_last = preset_cfg["channels_last"]
use_compile = preset_cfg["compile"]
use_amp = preset_cfg.get("amp", use_amp)
q_type = preset_cfg.get("quant", q_type)
device = select_device(device_choice)
if q_type in {"dynamic", "weight_only"} and device.type != "cpu":
print("Dynamic quantization runs on CPU; switching device to CPU.")
device = torch.device("cpu")
channels_last = False
use_amp = False
base_model = get_segmentation_model(config)
transform_fn = get_segmentation_transform(config)
class_labels = get_class_labels(config)
class_count = config.classes
original_result = run_segmentation_inference(
base_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
class_count=class_count,
)
fresh_model = clone_segmentation_model(config)
quant_model = apply_quantization(fresh_model, q_type)
quant_model = maybe_compile(quant_model, use_compile)
quant_result = run_segmentation_inference(
quant_model,
img,
device,
transform_fn,
channels_last=channels_last,
warmup=True,
use_amp=use_amp,
class_count=class_count,
)
size_orig = get_state_dict_size_mb(base_model)
size_quant = get_state_dict_size_mb(quant_model)
metrics_df = build_segmentation_metrics(original_result, quant_result, size_orig, size_quant, "Quantized Model")
class_df = build_class_distribution_df(
original_result["class_summary"],
quant_result["class_summary"],
class_labels,
"Quantized",
)
# Add labels to images for slider comparison
overlay_orig_labeled = add_image_label(original_result["overlay_original"], "Original Model")
overlay_quant_labeled = add_image_label(quant_result["overlay_original"], "Quantized Model")
mask_orig_labeled = add_image_label(original_result["mask_image_original"], "Original Mask")
mask_quant_labeled = add_image_label(quant_result["mask_image_original"], "Quantized Mask")
overlay_slider_value = (
overlay_orig_labeled,
overlay_quant_labeled,
)
mask_slider_value = (
mask_orig_labeled,
mask_quant_labeled,
)
downloads: list[str] = []
export_dir = Path("exports")
export_dir.mkdir(exist_ok=True)
if export_report:
report_path = export_dir / "quant_seg_report.json"
report = {
"model": config.name,
"checkpoint": config.checkpoint,
"dataset": config.dataset,
"quantization": q_type,
"metrics": metrics_df.to_dict(),
"class_distribution": class_df.to_dict(),
}
report_path.write_text(json.dumps(report, indent=2))
downloads.append(str(report_path))
if export_state:
state_path = export_dir / "quant_seg_state_dict.pth"
torch.save(quant_model.state_dict(), state_path)
downloads.append(str(state_path))
sample_tensor, _ = transform_fn(img)
sample_batch = sample_tensor.unsqueeze(0)
if export_ts:
ts_path = export_dir / "quant_seg_model.ts"
try:
scripted = torch.jit.trace(quant_model.cpu(), sample_batch)
scripted.save(ts_path)
downloads.append(str(ts_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"TorchScript export failed: {exc}")
if export_onnx:
onnx_path = export_dir / "quant_seg_model.onnx"
try:
torch.onnx.export(
quant_model.cpu(),
sample_batch,
onnx_path,
input_names=["input"],
output_names=["mask"],
opset_version=13,
dynamic_axes={"input": {0: "batch"}, "mask": {0: "batch"}},
)
downloads.append(str(onnx_path))
except Exception as exc: # pragma: no cover - export best effort
print(f"ONNX export failed: {exc}")
return (
metrics_df,
class_df,
overlay_slider_value,
mask_slider_value,
downloads,
)
# ---------------------------------------------
# GRADIO UI
# ---------------------------------------------
examples = [["examples/cat.jpg"], ["examples/dog.jpg"], ["examples/bird.jpg"], ["examples/car.jpg"], ["examples/elephant.jpg"]]
ade_examples = [["examples/ADE_val_00000001.jpg"], ["examples/ADE_val_00000002.jpg"]]
def create_demo():
with gr.Blocks() as demo:
gr.Markdown("# π§ Model Optimization Lab β Compare, Export, Benchmark")
device_opts = ["auto", "cpu"]
if torch.cuda.is_available():
device_opts.append("cuda")
if getattr(torch.backends, "mps", None) and torch.backends.mps.is_available():
device_opts.append("mps")
preset_opts = list(PRESETS.keys()) + ["custom"]
seg_model_options = [cfg.name for cfg in SEGMENTATION_MODEL_CONFIGS]
det_model_options = DETECTION_MODEL_OPTIONS.copy()
with gr.Tabs():
# ---- PRUNING TAB ----
with gr.Tab("Pruning-Classification"):
with gr.Row():
with gr.Column():
img_p = gr.Image(label="Upload Image")
model_p = gr.Dropdown(MODEL_OPTIONS, value=MODEL_OPTIONS[0], label="Base Model")
preset_p = gr.Dropdown(preset_opts, value="custom", label="Hardware Preset")
method_p = gr.Dropdown(["unstructured", "structured"], value="structured", label="Pruning Method")
amount_p = gr.Slider(minimum=0.1, maximum=0.9, step=0.1, value=0.4, label="Pruning Amount")
device_p = gr.Dropdown(device_opts, value=device_opts[0], label="Device")
channels_last_p = gr.Checkbox(label="Channels-last input (CUDA)", value=True)
amp_p = gr.Checkbox(label="Mixed precision (AMP)", value=True)
compile_p = gr.Checkbox(label="Torch compile (PyTorch 2)")
export_ts_p = gr.Checkbox(label="Export TorchScript")
export_onnx_p = gr.Checkbox(label="Export ONNX")
export_report_p = gr.Checkbox(label="Export JSON report", value=True)
btn_p = gr.Button("Run Pruned Model")
gr.Examples(examples=examples, inputs=img_p)
gr.Markdown(
"### π Classification Pruning Guide\n\n"
"**What is Pruning?**\n"
"Pruning removes less important weights from neural networks to reduce model size and potentially improve inference speed. "
"This tab applies pruning to ImageNet classification models.\n\n"
"**Options Explained:**\n"
"- **Base Model**: Select from 7 pretrained architectures (ResNet-50, MobileNetV3, EfficientNet-B0, ConvNeXt-Tiny, ViT-Base, RegNetY-016, EfficientNet-Lite0). Each has different size/accuracy tradeoffs.\n"
"- **Hardware Preset**: Quick configurations for common deployment scenarios:\n"
" - *Edge CPU*: Optimized for resource-constrained devices (CPU-only, 30% pruning, dynamic quantization)\n"
" - *Datacenter GPU*: Maximum performance on modern GPUs (CUDA, channels-last, compile, 20% pruning)\n"
" - *Apple MPS*: Tuned for Apple Silicon (M1/M2/M3 chips with Metal Performance Shaders)\n"
" - *Custom*: Manual control over all settings\n"
"- **Pruning Method**:\n"
" - *Structured*: Removes entire filters/channels; better hardware support and actual speedups\n"
" - *Unstructured*: Zeros individual weights; higher compression but needs specialized sparse kernels for speedup\n"
"- **Pruning Amount**: Percentage of weights to remove (0.1 = 10%, 0.9 = 90%). Higher values = smaller model but potential accuracy loss.\n"
"- **Device**: Inference hardware (auto-detects best available: CUDA β MPS β CPU)\n"
"- **Channels-last (CUDA only)**: Memory layout optimization for faster convolution operations on NVIDIA GPUs\n"
"- **Mixed Precision (AMP)**: Uses FP16 where safe, FP32 where needed; faster on modern GPUs with Tensor Cores\n"
"- **Torch Compile**: PyTorch 2.0+ graph optimization; can provide 20-40% speedup but adds compilation overhead\n\n"
"**Export Options:**\n"
"- *TorchScript*: Serialized model for C++ deployment or production serving\n"
"- *ONNX*: Cross-framework format (TensorRT, OpenVINO, ONNX Runtime, CoreML)\n"
"- *JSON Report*: Detailed metrics, settings, and Top-5 predictions for both models\n"
"- *State Dict*: Always saved; PyTorch checkpoint for loading pruned weights later\n\n"
"**Reading the Results:**\n"
"- *Comparison Metrics*: Side-by-side accuracy, speed, and size\n"
"- *Top-5 Chart*: Visual comparison of prediction confidence across models\n"
"- *Layer Sparsity*: Per-layer breakdown showing which parts were pruned most"
)
with gr.Column():
metrics_p = gr.Dataframe(label="π Comparison Metrics", headers=["Metric", "Original Model", "Pruned Model"])
chart_p = gr.Plot(label="π― Top-5 Predictions Comparison")
sparsity_p = gr.Dataframe(label="Layer sparsity (%)")
downloads_p = gr.Files(label="Exports (state_dict / TorchScript / ONNX / report)")
btn_p.click(
fn=run_pruned,
inputs=[
img_p,
model_p,
method_p,
amount_p,
device_p,
channels_last_p,
compile_p,
amp_p,
export_ts_p,
export_onnx_p,
export_report_p,
gr.State(True),
preset_p,
],
outputs=[metrics_p, chart_p, sparsity_p, downloads_p],
)
# ---- QUANTIZATION TAB ----
with gr.Tab("Quantization-Classification"):
with gr.Row():
with gr.Column():
img_q = gr.Image(label="Upload Image")
model_q = gr.Dropdown(MODEL_OPTIONS, value=MODEL_OPTIONS[0], label="Base Model")
preset_q = gr.Dropdown(preset_opts, value="custom", label="Hardware Preset")
q_type = gr.Dropdown(["dynamic", "weight_only", "fp16"], value="dynamic", label="Quantization Type")
device_q = gr.Dropdown(device_opts, value=device_opts[0], label="Device")
channels_last_q = gr.Checkbox(label="Channels-last input (CUDA)", value=True)
amp_q = gr.Checkbox(label="Mixed precision (AMP)", value=True)
compile_q = gr.Checkbox(label="Torch compile (PyTorch 2)")
export_ts_q = gr.Checkbox(label="Export TorchScript")
export_onnx_q = gr.Checkbox(label="Export ONNX")
export_report_q = gr.Checkbox(label="Export JSON report", value=True)
btn_q = gr.Button("Run Quantized Model")
gr.Examples(examples=examples, inputs=img_q)
gr.Markdown(
"### π Classification Quantization Guide\n\n"
"**What is Quantization?**\n"
"Quantization reduces model precision from 32-bit floats to lower bit-widths (INT8, FP16), decreasing memory usage and "
"enabling faster inference on hardware with specialized low-precision instructions.\n\n"
"**Options Explained:**\n"
"- **Base Model**: Choose from 7 pretrained ImageNet classifiers with varying complexity.\n"
"- **Hardware Preset**: Same presets as pruning tab, but with quantization-specific defaults.\n"
"- **Quantization Type**:\n"
" - *Dynamic*: Post-training INT8 quantization on linear layers; activations quantized dynamically at runtime. **Forces CPU** (PyTorch's INT8 kernels are CPU-only). Best for transformers and MLP-heavy models.\n"
" - *Weight-only*: Stores weights as INT8, computes in FP32. Reduces memory bandwidth, smaller model files. **CPU-optimized**.\n"
" - *FP16*: Half-precision floating point; requires GPU with FP16 support (CUDA, MPS). Minimal accuracy loss, ~2x speedup on modern GPUs.\n"
"- **Device**: Hardware target (dynamic/weight-only auto-switch to CPU for kernel compatibility)\n"
"- **Channels-last**: CUDA memory layout optimization (ignored on CPU)\n"
"- **Mixed Precision (AMP)**: Can combine with FP16 quantization on GPUs\n"
"- **Torch Compile**: Graph-level optimizations from PyTorch 2.0+\n\n"
"**Export Options:** Same as pruning (TorchScript, ONNX, JSON report, state dict)\n\n"
"**Important Notes:**\n"
"β οΈ Dynamic/weight-only quantization automatically uses CPU even if GPU is selected (PyTorch limitation)\n"
"β οΈ ResNet-50 and similar CNN-heavy models see modest INT8 speedups because only linear layers are quantized\n"
"β οΈ FP16 on CPU often reverts to FP32 internally, adding overhead instead of speedup\n\n"
"**Reading the Results:**\n"
"- *Latency*: Dynamic quantization may show higher latency due to runtime overhead; production deployments should use cached models\n"
"- *Model Size*: FP16 β 50% reduction, INT8 dynamic β 75% reduction (varies by architecture)\n"
"- *Accuracy*: Watch for confidence drops; quantization can shift predictions slightly"
)
with gr.Column():
metrics_q = gr.Dataframe(label="π Comparison Metrics", headers=["Metric", "Original Model", "Quantized Model"])
chart_q = gr.Plot(label="π― Top-5 Predictions Comparison")
downloads_q = gr.Files(label="Exports (state_dict / TorchScript / ONNX / report)")
btn_q.click(
fn=run_quantized,
inputs=[
img_q,
model_q,
q_type,
device_q,
channels_last_q,
compile_q,
amp_q,
export_ts_q,
export_onnx_q,
export_report_q,
gr.State(True),
preset_q,
],
outputs=[metrics_q, chart_q, downloads_q],
)
# ---- DETECTION PRUNING TAB ----
with gr.Tab("Pruning-Detection"):
with gr.Row():
with gr.Column():
img_dp = gr.Image(label="Upload Image")
model_dp = gr.Dropdown(det_model_options, value=det_model_options[0], label="Object Detector (COCO)")
preset_dp = gr.Dropdown(preset_opts, value="custom", label="Hardware Preset")
method_dp = gr.Dropdown(["unstructured", "structured"], value="structured", label="Pruning Method")
amount_dp = gr.Slider(minimum=0.1, maximum=0.9, step=0.1, value=0.3, label="Pruning Amount")
score_dp = gr.Slider(minimum=0.05, maximum=0.9, step=0.05, value=0.25, label="Score Threshold")
device_dp = gr.Dropdown(device_opts, value=device_opts[0], label="Device")
channels_last_dp = gr.Checkbox(label="Channels-last input (CUDA)", value=True)
amp_dp = gr.Checkbox(label="Mixed precision (AMP)", value=True)
compile_dp = gr.Checkbox(label="Torch compile (PyTorch 2)")
export_ts_dp = gr.Checkbox(label="Export TorchScript")
export_onnx_dp = gr.Checkbox(label="Export ONNX")
export_report_dp = gr.Checkbox(label="Export JSON report", value=True)
btn_dp = gr.Button("Run Detection Pruning")
gr.Examples(examples=examples, inputs=img_dp)
gr.Markdown(
"### π¦Ύ Detection Pruning Guide\n\n"
"**Models:**\n"
"- TorchVision: Faster R-CNN ResNet50 FPN, SSDlite320 MobileNetV3 (COCO pretrained)\n"
"- Ultralytics YOLO12: sizes n/s/m/l/x (COCO, auto-downloaded if missing)\n\n"
"**Core Options:**\n"
"- *Hardware Preset*: Same CPU/GPU defaults as classification; channels-last only applies on CUDA.\n"
"- *Pruning Method*: Structured is safest for detection heads; unstructured yields higher sparsity but rarely speeds up NMS.\n"
"- *Score Threshold*: Filters low-confidence boxes before metrics/overlays.\n"
"- *AMP / Torch Compile*: Only useful on GPU; compile adds startup cost but can speed up steady-state.\n"
"- *YOLO12 exports*: TorchScript/ONNX disabled here; state_dict still saved for the underlying torch model.\n\n"
"**Reading Results:**\n"
"- Metrics: latency, box count above threshold, mean score, model size.\n"
"- Overlay slider: drag to compare original vs pruned detections.\n"
"- Detections table: flattened list of boxes for quick scanning."
)
with gr.Column():
metrics_dp = gr.Dataframe(label="π Detection Metrics", headers=["Metric", "Original Model", "Pruned Model"])
overlay_dp = gr.ImageSlider(label="Overlay Comparison", type="pil")
dets_dp = gr.Dataframe(label="Detections (Original vs Pruned)")
downloads_dp = gr.Files(label="Exports (state_dict / TorchScript / ONNX / report)")
btn_dp.click(
fn=run_pruned_detection,
inputs=[
img_dp,
model_dp,
method_dp,
amount_dp,
device_dp,
channels_last_dp,
compile_dp,
amp_dp,
export_ts_dp,
export_onnx_dp,
export_report_dp,
gr.State(True),
preset_dp,
score_dp,
],
outputs=[
metrics_dp,
overlay_dp,
dets_dp,
downloads_dp,
],
)
# ---- DETECTION QUANTIZATION TAB ----
with gr.Tab("Quantization-Detection"):
with gr.Row():
with gr.Column():
img_dq = gr.Image(label="Upload Image")
model_dq = gr.Dropdown(det_model_options, value=det_model_options[0], label="Object Detector (COCO)")
preset_dq = gr.Dropdown(preset_opts, value="custom", label="Hardware Preset")
q_type_dq = gr.Dropdown(["dynamic", "weight_only", "fp16"], value="dynamic", label="Quantization Type")
score_dq = gr.Slider(minimum=0.05, maximum=0.9, step=0.05, value=0.25, label="Score Threshold")
device_dq = gr.Dropdown(device_opts, value=device_opts[0], label="Device")
channels_last_dq = gr.Checkbox(label="Channels-last input (CUDA)", value=True)
amp_dq = gr.Checkbox(label="Mixed precision (AMP)", value=True)
compile_dq = gr.Checkbox(label="Torch compile (PyTorch 2)")
export_ts_dq = gr.Checkbox(label="Export TorchScript")
export_onnx_dq = gr.Checkbox(label="Export ONNX")
export_report_dq = gr.Checkbox(label="Export JSON report", value=True)
btn_dq = gr.Button("Run Detection Quantization")
gr.Examples(examples=examples, inputs=img_dq)
gr.Markdown(
"### β‘ Detection Quantization Guide\n\n"
"**Models:** TorchVision detectors and YOLO12 n/s/m/l/x (Ultralytics). YOLO12 uses its internal preprocessing; other models use TorchVision transforms.\n\n"
"**Quantization Modes:**\n"
"- *Dynamic / Weight-only*: INT8 linear layers on CPU. UI auto-switches to CPU even if GPU selected (PyTorch limitation).\n"
"- *FP16*: Half precision for CUDA/MPS; keeps CPU in FP32. Pair with AMP + channels-last for best GPU speed.\n\n"
"**Tips:**\n"
"- Score threshold trims noisy boxes before metrics/overlays.\n"
"- TorchScript/ONNX exports are skipped for YOLO12; state_dict still saved. TorchVision exports remain enabled.\n"
"- For fastest runs, keep AMP + channels-last on CUDA; disable compile if you only run a single image.\n\n"
"**Outputs:** Metrics table, overlay slider, detections table, and exports in `exports/` with `_det` suffix."
)
with gr.Column():
metrics_dq = gr.Dataframe(label="π Detection Metrics", headers=["Metric", "Original Model", "Quantized Model"])
overlay_dq = gr.ImageSlider(label="Overlay Comparison", type="pil")
dets_dq = gr.Dataframe(label="Detections (Original vs Quantized)")
downloads_dq = gr.Files(label="Exports (state_dict / TorchScript / ONNX / report)")
btn_dq.click(
fn=run_quantized_detection,
inputs=[
img_dq,
model_dq,
q_type_dq,
device_dq,
channels_last_dq,
compile_dq,
amp_dq,
export_ts_dq,
export_onnx_dq,
export_report_dq,
gr.State(True),
preset_dq,
score_dq,
],
outputs=[
metrics_dq,
overlay_dq,
dets_dq,
downloads_dq,
],
)
# ---- SEGMENTATION PRUNING TAB ----
with gr.Tab("Pruning-Segmentation"):
with gr.Row():
with gr.Column():
img_sp = gr.Image(label="Upload Image")
model_sp = gr.Dropdown(seg_model_options, value=seg_model_options[0], label="Pretrained ADE20K Model")
preset_sp = gr.Dropdown(preset_opts, value="custom", label="Hardware Preset")
method_sp = gr.Dropdown(["unstructured", "structured"], value="structured", label="Pruning Method")
amount_sp = gr.Slider(minimum=0.1, maximum=0.9, step=0.1, value=0.4, label="Pruning Amount")
device_sp = gr.Dropdown(device_opts, value=device_opts[0], label="Device")
channels_last_sp = gr.Checkbox(label="Channels-last input (CUDA)", value=True)
compile_sp = gr.Checkbox(label="Torch compile (PyTorch 2)")
amp_sp = gr.Checkbox(label="Mixed precision (AMP)", value=True)
export_ts_sp = gr.Checkbox(label="Export TorchScript")
export_onnx_sp = gr.Checkbox(label="Export ONNX")
export_report_sp = gr.Checkbox(label="Export JSON report", value=True)
btn_sp = gr.Button("Run Segmentation Pruning")
gr.Examples(examples=ade_examples, inputs=img_sp, label="ADE20K Samples")
gr.Markdown(
"### π¨ Segmentation Pruning Guide\n\n"
"**What is Semantic Segmentation?**\n"
"Semantic segmentation assigns a class label to every pixel in an image (e.g., sky, road, person, car). "
"This tab uses ADE20K-pretrained models that recognize 150 scene categories.\n\n"
"**Available Models:**\n"
"- **SegFormer B0** (512x512): Lightweight transformer-based segmenter; efficient for edge deployment\n"
"- **SegFormer B4** (512x512): Larger variant with better accuracy; ~4x B0 parameters\n"
"- **DPT Large**: Vision-transformer-based dense prediction; state-of-the-art accuracy but slower\n"
"- **UPerNet ConvNeXt-Tiny**: Unified perceptual parsing with modern CNN backbone; balanced speed/accuracy\n\n"
"**Segmentation-Specific Options:**\n"
"- All pruning/device/compile options work the same as classification\n"
"- Models use [smp-hub](https://huggingface.co/smp-hub) pretrained checkpoints via `segmentation-models-pytorch`\n"
"- Preprocessing pipelines are model-specific (loaded from Hugging Face metadata)\n"
"- Images are resized based on model training resolution (usually 512x512 or 640x640)\n\n"
"**Understanding Segmentation Outputs:**\n"
"1. **Comparison Metrics Table**:\n"
" - *Latency*: Inference time for full-image segmentation\n"
" - *Mean Confidence*: Average softmax probability across all pixels\n"
" - *Model Size*: State dict size in MB\n"
" - *Mask Agreement*: % of pixels with identical class predictions (100% = perfect match)\n"
"2. **Class Distribution Table**:\n"
" - Top 25 most prevalent classes by pixel coverage\n"
" - Shows percentage and pixel counts for both models\n"
" - Helps identify which objects dominate the scene\n"
"3. **Overlay Comparison Slider**:\n"
" - Original image blended with colored segmentation masks\n"
" - Drag slider to compare original vs. pruned predictions\n"
" - Colors map to specific ADE20K classes (150 categories)\n"
"4. **Mask Comparison Slider**:\n"
" - Raw segmentation masks without image overlay\n"
" - Easier to spot subtle prediction differences\n"
"5. **Layer Sparsity Table**:\n"
" - Per-layer pruning statistics showing compression levels\n\n"
"**Export Options:**\n"
"Files saved with `_seg` suffix: `pruned_seg_model.ts`, `pruned_seg_report.json`, etc.\n\n"
"**Tips:**\n"
"- Use ADE20K validation images (provided examples) for meaningful class diversity\n"
"- High mask agreement (>95%) indicates pruning preserved segmentation quality\n"
"- Check class distribution to ensure dominant objects aren't misclassified\n"
"- Structured pruning typically maintains better segmentation quality than unstructured"
)
with gr.Column():
metrics_sp = gr.Dataframe(label="π Comparison Metrics")
class_sp = gr.Dataframe(label="π Class Distribution")
overlay_slider_sp = gr.ImageSlider(label="Overlay Comparison", type="pil")
mask_slider_sp = gr.ImageSlider(label="Mask Comparison", type="pil")
sparsity_sp = gr.Dataframe(label="Layer sparsity (%)")
downloads_sp = gr.Files(label="Exports (state_dict / TorchScript / ONNX / report)")
btn_sp.click(
fn=run_pruned_segmentation,
inputs=[
img_sp,
model_sp,
method_sp,
amount_sp,
device_sp,
channels_last_sp,
compile_sp,
amp_sp,
export_ts_sp,
export_onnx_sp,
export_report_sp,
gr.State(True),
preset_sp,
],
outputs=[
metrics_sp,
class_sp,
overlay_slider_sp,
mask_slider_sp,
sparsity_sp,
downloads_sp,
],
)
# ---- SEGMENTATION QUANTIZATION TAB ----
with gr.Tab("Quantization-Segmentation"):
with gr.Row():
with gr.Column():
img_sq = gr.Image(label="Upload Image")
model_sq = gr.Dropdown(seg_model_options, value=seg_model_options[0], label="Pretrained ADE20K Model")
preset_sq = gr.Dropdown(preset_opts, value="custom", label="Hardware Preset")
q_type_sq = gr.Dropdown(["dynamic", "weight_only", "fp16"], value="dynamic", label="Quantization Type")
device_sq = gr.Dropdown(device_opts, value=device_opts[0], label="Device")
channels_last_sq = gr.Checkbox(label="Channels-last input (CUDA)", value=True)
compile_sq = gr.Checkbox(label="Torch compile (PyTorch 2)")
amp_sq = gr.Checkbox(label="Mixed precision (AMP)", value=True)
export_ts_sq = gr.Checkbox(label="Export TorchScript")
export_onnx_sq = gr.Checkbox(label="Export ONNX")
export_report_sq = gr.Checkbox(label="Export JSON report", value=True)
btn_sq = gr.Button("Run Segmentation Quantization")
gr.Examples(examples=ade_examples, inputs=img_sq, label="ADE20K Samples")
gr.Markdown(
"### π¨ Segmentation Quantization Guide\n\n"
"**Quantization for Dense Prediction:**\n"
"Semantic segmentation models are typically larger and slower than classifiers, making quantization especially valuable. "
"This tab applies the same quantization techniques as classification but evaluates pixel-level prediction quality.\n\n"
"**Available Models & Quantization:**\n"
"- **SegFormer B0/B4**: Transformer-based; dynamic quantization helps with attention/MLP layers (CPU-only)\n"
"- **DPT Large**: Vision-transformer backbone; benefits significantly from FP16 on GPU (~2x speedup)\n"
"- **UPerNet ConvNeXt-Tiny**: CNN-based; FP16 quantization provides best GPU acceleration\n\n"
"**Quantization Type Selection:**\n"
"- **Dynamic/Weight-only**: β οΈ Automatically uses CPU (PyTorch INT8 limitation). Best for: \n"
" - Transformer-heavy models (SegFormer, DPT)\n"
" - CPU-only deployment scenarios\n"
" - Memory-constrained environments\n"
"- **FP16**: Recommended for GPU deployment (CUDA, MPS). Provides:\n"
" - ~2x inference speedup on modern GPUs\n"
" - 50% memory reduction\n"
" - Minimal segmentation quality loss (<1% mIoU typically)\n\n"
"**Segmentation-Specific Metrics:**\n"
"1. **Mask Agreement**: Critical metric for segmentation; >95% is good, >98% is excellent\n"
"2. **Mean Confidence**: Should remain similar; large drops indicate quantization instability\n"
"3. **Class Distribution**: Compare pixel percentages; mismatches show which objects are affected\n\n"
"**Understanding the Outputs:**\n"
"- **Overlay Slider**: Drag to compare original vs. quantized predictions on the actual image\n"
"- **Mask Slider**: Raw segmentation masks for detailed comparison\n"
"- **Class Distribution**: Top 25 classes help identify systematic errors (e.g., 'road' β 'sidewalk' confusion)\n\n"
"**Performance Expectations:**\n"
"- **FP16 on CUDA**: Expect 1.5-2x speedup with <1% accuracy loss\n"
"- **Dynamic on CPU**: Model size β 75%, latency may increase (first-run overhead)\n"
"- **Weight-only on CPU**: Model size β 50%, latency similar to FP32\n\n"
"**Export Options:**\n"
"Files saved with `_seg` suffix: `quant_seg_model.onnx`, `quant_seg_state_dict.pth`, etc.\n\n"
"**Best Practices:**\n"
"β Use FP16 for GPU deployment (CUDA, MPS)\n"
"β Use dynamic quantization for CPU-bound transformer models\n"
"β Check mask agreement before deploying; <90% needs investigation\n"
"β Validate on multiple images; some scenes may be more sensitive to quantization\n"
"β Avoid FP16 on CPU (performance penalty, not benefit)\n"
"β Don't expect large speedups from dynamic quantization on CNN-heavy models (most layers are Conv2d, not Linear)"
)
with gr.Column():
metrics_sq = gr.Dataframe(label="π Comparison Metrics")
class_sq = gr.Dataframe(label="π Class Distribution")
overlay_slider_sq = gr.ImageSlider(label="Overlay Comparison", type="pil")
mask_slider_sq = gr.ImageSlider(label="Mask Comparison", type="pil")
downloads_sq = gr.Files(label="Exports (state_dict / TorchScript / ONNX / report)")
btn_sq.click(
fn=run_quantized_segmentation,
inputs=[
img_sq,
model_sq,
q_type_sq,
device_sq,
channels_last_sq,
compile_sq,
amp_sq,
export_ts_sq,
export_onnx_sq,
export_report_sq,
gr.State(True),
preset_sq,
],
outputs=[
metrics_sq,
class_sq,
overlay_slider_sq,
mask_slider_sq,
downloads_sq,
],
)
return demo
def main():
parser = argparse.ArgumentParser(description="Model optimization lab")
parser.add_argument("--cli", action="store_true", help="Run in CLI mode (no UI)")
parser.add_argument("--mode", choices=["prune", "quant"], default="prune", help="Optimization mode in CLI")
parser.add_argument("--image", type=str, help="Path to an image for CLI mode")
parser.add_argument("--model", type=str, default=MODEL_OPTIONS[0], choices=MODEL_OPTIONS)
parser.add_argument("--device", type=str, default="auto")
args = parser.parse_args()
if args.cli:
if not args.image:
raise SystemExit("--image is required in CLI mode")
img = Image.open(args.image)
if args.mode == "prune":
metrics, _, downloads = run_pruned(img, args.model, "structured", 0.4, device_choice=args.device)
else:
metrics, _, downloads = run_quantized(img, args.model, "dynamic", device_choice=args.device)
print(metrics)
print("Exports:", downloads)
return
demo = create_demo()
demo.launch(theme = gr.themes.Soft())
if __name__ == "__main__":
main()
|