shell：cd eval_retrieval
--dataset lvis coco
extract embedding
torchrun --nproc-per-node=8 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=29500
extract_embedding.py --model wedetect
--wedetect_checkpoint wedetect_base.pth
--wedetect_uni_checkpoint wedetect_base_uni.pth
--dataset lvis # coco
retrieval
python3 retrieval_metric.py --model wedetect --dataset lvis --thre 0.2
results:
================================================================================
Class Prec Recall F1 Support Pred
--------------------------------------------------------------------------------
barbell 1.0 1.0 1.0 1 1
zebra 0.9726 0.9846 0.9786 325 329
elephant 0.9517 0.9753 0.9634 364 373
giraffe 0.9171 0.9888 0.9516 358 386
banana 0.8946 0.929 0.9115 338 351
fireplug 0.8503 0.9653 0.9042 259 294
broccoli 0.8163 0.9748 0.8885 278 332
frisbee 0.81 0.9701 0.8828 334 400
mouse_(computer_equipment) 0.8297 0.9273 0.8758 289 323
Ferris_wheel 1.0 0.7778 0.875 9 7
...........
.......
sled 0.0714 0.1667 0.1 6 14
matchbox 0.0556 0.5 0.1 2 18
bamboo 0.0714 0.1667 0.1 12 28
duffel_bag 0.0543 0.6098 0.0998 41 460
toaster_oven 0.0545 0.4615 0.0976 13 110
hook 0.0759 0.1364 0.0976 44 79
French_toast 0.0513 1.0 0.0976 2 39
thermostat 0.0563 0.36 0.0973 25 160
cork_(bottle_plug) 0.0602 0.25 0.0971 20 83
bead 0.0538 0.4545 0.0962 11 93
vest 0.052 0.617 0.0959 94 1116
dress_suit 0.0519 0.6154 0.0958 13 154
wok 0.0506 0.8 0.0952 5 79
gun 0.0556 0.3333 0.0952 3 18
television_camera 0.0588 0.25 0.0952 4 17
shampoo 0.0559 0.3214 0.0952 28 161
candy_bar 0.0526 0.5 0.0952 2 19
handkerchief 0.0556 0.2857 0.093 7 36
crucifix 0.0495 0.7143 0.0926 7 101
fire_hose 0.0833 0.1 0.0909 10 12
falcon 0.0476 1.0 0.0909 1 21
dress_hat 0.0479 0.8 0.0904 10 167
green_onion 0.0487 0.6111 0.0902 18 226
vending_machine 0.0492 0.5 0.0896 12 122
ladle 0.0474 0.7647 0.0893 17 274
reamer_(juicer) 0.05 0.4 0.0889 5 40
crescent_roll 0.0541 0.25 0.0889 8 37
brake_light 0.0469 0.6222 0.0872 45 597
boiled_egg 0.0465 0.6667 0.087 6 86
tachometer 0.0455 1.0 0.087 1 22
passenger_car_(part_of_a_train) 0.0468 0.6154 0.087 13 171
bagel 0.0459 0.7692 0.0866 13 218
bathrobe 0.0476 0.4286 0.0857 7 63
bookcase 0.0448 0.9286 0.0855 14 290
dispenser 0.0619 0.1364 0.0851 44 97
sponge 0.0588 0.1538 0.0851 13 34
calf 0.0506 0.2667 0.0851 15 79
kayak 0.0449 0.6667 0.0842 6 89
paperback_book 0.0435 1.0 0.0833 1 23
CD_player 0.0448 0.6 0.0833 5 67
cupboard 0.0436 0.9375 0.0833 48 1032
mallet 0.05 0.25 0.0833 4 20
paper_plate 0.0434 0.7333 0.0819 60 1014
mailbox_(at_home) 0.046 0.3636 0.0816 22 174
wristband 0.0427 0.84 0.0812 25 492
shopping_bag 0.0422 0.8333 0.0803 30 593
sour_cream 0.0426 0.6667 0.08 3 47
router_(computer_equipment) 0.0526 0.1667 0.08 6 19
file_cabinet 0.0455 0.3333 0.08 9 66
bib 0.0444 0.3333 0.0784 12 90
raincoat 0.043 0.4444 0.0784 9 93
wallet 0.0416 0.625 0.0779 24 361
tote_bag 0.0414 0.6429 0.0778 28 435
legging_(clothing) 0.044 0.3333 0.0777 12 91
atomizer 0.0588 0.1111 0.0769 9 17
figurine 0.0407 0.5333 0.0757 30 393
shoulder_bag 0.0399 0.7333 0.0756 30 552
cart 0.0403 0.6 0.0755 10 149
suspenders 0.0397 0.6667 0.0749 15 252
grizzly 0.0385 1.0 0.0741 6 156
igniter 0.0909 0.0625 0.0741 16 11
casserole 0.0435 0.25 0.0741 4 23
notepad 0.1111 0.0556 0.0741 18 9
coleslaw 0.039 0.6 0.0732 5 77
bobbin 0.0444 0.2 0.0727 10 45
postbox_(public) 0.0382 0.5 0.0709 10 131
legume 0.0364 0.8 0.0696 5 110
cube 0.0357 1.0 0.069 1 28
Sharpie 0.037 0.5 0.069 2 27
muffin 0.0379 0.3333 0.068 15 132
person 0.0351 0.9769 0.0678 390 10842
billboard 0.0351 0.8378 0.0675 37 882
bandanna 0.04 0.2083 0.0671 24 125
mandarin_orange 0.0347 0.9 0.0669 10 259
beachball 0.037 0.3333 0.0667 3 27
heater 0.0526 0.0909 0.0667 11 19
highchair 0.0377 0.2857 0.0667 7 53
volleyball 0.0345 0.6667 0.0656 3 58
walking_stick 0.0351 0.5 0.0656 4 57
webcam 0.05 0.0909 0.0645 11 20
bedspread 0.0344 0.5294 0.0645 17 262
Dixie_cup 0.0329 0.7895 0.0632 19 456
dining_table 0.0328 0.8571 0.0632 70 1830
milestone 0.0323 1.0 0.0625 1 31
sombrero 0.0323 1.0 0.0625 1 31
hairnet 0.0345 0.3333 0.0625 3 29
footstool 0.0326 0.6 0.0619 5 92
card 0.0349 0.2727 0.0619 11 86
chickpea 0.0317 1.0 0.0615 2 63
pepper 0.0357 0.2188 0.0614 32 196
sunhat 0.0318 0.7143 0.061 14 314
receipt 0.0314 0.8571 0.0606 7 191
freight_car 0.0306 0.6 0.0583 5 98
mixer_(kitchen_tool) 0.0377 0.125 0.058 16 53
trunk 0.03 0.6 0.0571 10 200
ice_maker 0.0357 0.1429 0.0571 7 28
barrel 0.0291 0.75 0.056 20 516
overalls_(clothing) 0.0301 0.3846 0.0559 13 166
garbage_truck 0.0303 0.3333 0.0556 3 33
chocolate_cake 0.0291 0.6 0.0556 5 103
tissue_paper 0.031 0.2292 0.0546 48 355
tapestry 0.028 0.75 0.0541 4 107
business_card 0.0323 0.1667 0.0541 6 31
manger 0.0526 0.0556 0.0541 18 19
battery 0.0303 0.2222 0.0533 9 66
comic_book 0.027 1.0 0.0526 1 37
flannel 0.027 0.5 0.0513 2 37
padlock 0.0476 0.0556 0.0513 18 21
sculpture 0.0267 0.5 0.0507 14 262
booklet 0.0274 0.2857 0.05 14 146
headstall_(for_horses) 0.0261 0.6 0.05 10 230
clip 0.0299 0.1429 0.0494 14 67
loveseat 0.0333 0.0909 0.0488 11 30
pony 0.025 0.8333 0.0485 12 400
sweatband 0.028 0.1765 0.0484 17 107
jeep 0.0255 0.4444 0.0482 9 157
sweatshirt 0.0249 0.713 0.0481 115 3296
turnip 0.025 0.5 0.0476 2 40
cylinder 0.0244 1.0 0.0476 1 41
step_stool 0.025 0.4 0.0471 5 80
tape_(sticky_cloth_or_paper) 0.0455 0.0455 0.0455 22 22
salami 0.0233 0.6667 0.0449 3 86
toast_(food) 0.0263 0.1538 0.0449 13 76
parachute 0.0231 0.6 0.0444 5 130
headset 0.0226 1.0 0.0441 3 133
jet_plane 0.0225 0.7 0.0436 10 311
medicine 0.0233 0.3077 0.0432 13 172
water_faucet 0.0217 1.0 0.0426 15 690
coffeepot 0.0227 0.3333 0.0426 6 88
eggplant 0.022 0.5 0.0421 4 91
wall_clock 0.0215 0.8125 0.0419 16 605
shears 0.0213 0.6667 0.0412 3 94
toolbox 0.0238 0.1429 0.0408 7 42
food_processor 0.0213 0.5 0.0408 4 94
underwear 0.0227 0.1818 0.0404 11 88
bouquet 0.0208 0.6667 0.0404 3 96
latch 0.0211 0.4048 0.0402 42 804
bandage 0.022 0.2222 0.04 9 91
folding_chair 0.0207 0.5 0.0397 12 290
bowler_hat 0.0207 0.4286 0.0395 7 145
crouton 0.0202 0.6667 0.0392 3 99
flap 0.0217 0.1667 0.0385 6 46
pliers 0.0213 0.1667 0.0377 6 47
freshener 0.0196 0.3333 0.037 3 51
diary 0.0189 1.0 0.037 1 53
puppy 0.0188 1.0 0.0369 9 479
sugar_bowl 0.0196 0.25 0.0364 4 51
turban 0.0183 0.8 0.0359 5 218
identity_card 0.0196 0.2 0.0357 5 51
window_box_(for_plants) 0.0184 0.5 0.0354 14 381
tongs 0.0312 0.0385 0.0345 26 32
teacup 0.0174 0.7778 0.034 9 403
honey 0.0179 0.3333 0.0339 3 56
tow_truck 0.0174 0.5 0.0336 4 115
alcohol 0.0169 0.3571 0.0324 14 295
parchment 0.0164 1.0 0.0323 1 61
soupspoon 0.0161 1.0 0.0316 7 436
lab_coat 0.0164 0.3333 0.0312 3 61
turkey_(food) 0.0167 0.25 0.0312 8 120
camcorder 0.0169 0.1667 0.0308 6 59
thread 0.016 0.2308 0.03 13 187
clipboard 0.0167 0.125 0.0294 8 60
slipper_(footwear) 0.015 0.4615 0.0291 13 400
brass_plaque 0.0147 1.0 0.029 1 68
saddlebag 0.0159 0.1667 0.029 6 63
table_lamp 0.0146 0.8333 0.0288 12 683
armband 0.0146 0.875 0.0287 8 480
teakettle 0.0146 0.75 0.0287 4 205
armoire 0.0145 1.0 0.0286 1 69
barrow 0.0146 0.6667 0.0286 3 137
racket 0.0144 1.0 0.0285 7 485
basketball 0.0137 0.6667 0.0268 9 438
cincture 0.0139 0.3333 0.0267 3 72
seabird 0.0137 0.3333 0.0263 3 73
dinghy 0.0132 0.6667 0.026 3 151
pet 0.0135 0.3529 0.026 17 444
kitten 0.0132 0.75 0.0259 8 455
blouse 0.0132 0.5686 0.0258 51 2195
trench_coat 0.013 1.0 0.0256 1 77
pottery 0.0127 0.5 0.0248 6 236
newsstand 0.0127 0.3333 0.0244 3 79
chili_(vegetable) 0.0122 1.0 0.0242 3 245
bat_(animal) 0.0122 1.0 0.0241 1 82
hamburger 0.0122 0.6 0.024 5 245
cover 0.0125 0.24 0.0238 25 479
bulletin_board 0.012 0.3 0.0232 10 249
water_cooler 0.0125 0.1111 0.0225 9 80
heron 0.0115 0.5 0.0225 2 87
carton 0.0113 0.2222 0.0216 18 353
coloring_material 0.0108 1.0 0.0213 1 93
brownie 0.0107 0.8333 0.0211 6 468
pie 0.0107 0.3333 0.0207 9 281
vinegar 0.0103 1.0 0.0204 3 291
parka 0.0102 0.25 0.0196 4 98
spotlight 0.0099 0.2727 0.0191 11 303
wedding_ring 0.0098 0.3333 0.019 6 204
dumpster 0.0092 0.6667 0.0182 15 1086
parasol 0.009 0.6 0.0178 5 332
pouch 0.0083 0.1 0.0154 10 120
blazer 0.0078 0.2 0.0151 10 255
shaker 0.0078 0.2 0.015 5 128
packet 0.0077 0.25 0.0149 4 130
water_jug 0.007 1.0 0.014 2 284
detergent 0.007 0.3333 0.0138 3 142
dish 0.0067 0.4762 0.0132 21 1489
flute_glass 0.0068 0.3333 0.0132 6 296
clasp 0.0067 0.1111 0.0127 9 149
washbasin 0.0064 1.0 0.0127 2 312
canister 0.0062 0.4286 0.0122 7 483
garbage 0.0056 0.5 0.0111 4 357
saucepan 0.0055 1.0 0.011 2 361
gasmask 0.0056 1.0 0.011 1 180
grocery_bag 0.005 1.0 0.01 3 600
mat_(gym_equipment) 0.0047 1.0 0.0093 4 854
liquor 0.0044 1.0 0.0087 2 458
rat 0.0042 0.5 0.0084 2 237
runner_(carpet) 0.0037 0.6667 0.0073 3 545
platter 0.0036 0.7143 0.0071 7 1401
cayenne_(spice) 0.0034 0.3333 0.0068 3 292
softball 0.0034 1.0 0.0067 1 298
kitchen_table 0.0021 0.5714 0.0043 7 1864
keg 0.002 1.0 0.004 1 494
sweat_pants 0.0018 0.2 0.0036 5 548
cap_(headwear) 0.0016 1.0 0.0031 5 3191
hamper 0.0014 1.0 0.0029 1 693
gravy_boat 0.0009 0.5 0.0017 2 1173
costume 0.0008 0.4286 0.0016 7 3736
string_cheese 0.0 0.0 0.0 1 76
futon 0.0 0.0 0.0 4 49
nest 0.0 0.0 0.0 2 11
pirate_flag 0.0 0.0 0.0 1 17
curling_iron 0.0 0.0 0.0 1 17
deadbolt 0.0 0.0 0.0 7 51
cream_pitcher 0.0 0.0 0.0 2 55
binoculars 0.0 0.0 0.0 2 1
foal 0.0 0.0 0.0 4 0
gargoyle 0.0 0.0 0.0 2 3
peanut_butter 0.0 0.0 0.0 10 19
flipper_(footwear) 0.0 0.0 0.0 2 5
crape 0.0 0.0 0.0 2 3
tambourine 0.0 0.0 0.0 1 1
speaker_(stero_equipment) 0.0 0.0 0.0 210 0
handsaw 0.0 0.0 0.0 4 0
smoothie 0.0 0.0 0.0 2 20
wine_bucket 0.0 0.0 0.0 5 16
bow_(weapon) 0.0 0.0 0.0 2 113
pendulum 0.0 0.0 0.0 1 12
leather 0.0 0.0 0.0 1 4
quiche 0.0 0.0 0.0 1 17
lamb-chop 0.0 0.0 0.0 1 35
penny_(coin) 0.0 0.0 0.0 1 0
dollar 0.0 0.0 0.0 1 2
alligator 0.0 0.0 0.0 1 0
wolf 0.0 0.0 0.0 2 7
koala 0.0 0.0 0.0 1 3
gondola_(boat) 0.0 0.0 0.0 1 68
dumbbell 0.0 0.0 0.0 2 2
corset 0.0 0.0 0.0 2 1
compass 0.0 0.0 0.0 1 5
poker_(fire_stirring_tool) 0.0 0.0 0.0 2 26
eclair 0.0 0.0 0.0 1 100
poncho 0.0 0.0 0.0 2 4
salmon_(food) 0.0 0.0 0.0 2 6
shot_glass 0.0 0.0 0.0 4 55
noseband_(for_animals) 0.0 0.0 0.0 10 0
playpen 0.0 0.0 0.0 1 121
syringe 0.0 0.0 0.0 1 4
shawl 0.0 0.0 0.0 1 24
pocketknife 0.0 0.0 0.0 1 10
windsock 0.0 0.0 0.0 6 8
jumpsuit 0.0 0.0 0.0 1 8
pistol 0.0 0.0 0.0 1 7
knocker_(on_a_door) 0.0 0.0 0.0 2 6
broach 0.0 0.0 0.0 1 1
pin_(non_jewelry) 0.0 0.0 0.0 4 12
mint_candy 0.0 0.0 0.0 3 2
corkboard 0.0 0.0 0.0 1 16
chocolate_mousse 0.0 0.0 0.0 1 13
gargle 0.0 0.0 0.0 3 18
timer 0.0 0.0 0.0 11 69
sportswear 0.0 0.0 0.0 1 8
hand_glass 0.0 0.0 0.0 2 5
river_boat 0.0 0.0 0.0 1 0
crab_(animal) 0.0 0.0 0.0 4 0
wind_chime 0.0 0.0 0.0 2 2
shark 0.0 0.0 0.0 5 3
nut 0.0 0.0 0.0 11 28
golf_club 0.0 0.0 0.0 2 11
sword 0.0 0.0 0.0 11 3
pantyhose 0.0 0.0 0.0 3 23
candy_cane 0.0 0.0 0.0 2 6
forklift 0.0 0.0 0.0 4 24
cardigan 0.0 0.0 0.0 1 22
tobacco_pipe 0.0 0.0 0.0 1 4
fruit_juice 0.0 0.0 0.0 1 69
ambulance 0.0 0.0 0.0 2 18
boom_microphone 0.0 0.0 0.0 2 24
ginger 0.0 0.0 0.0 4 2
bobby_pin 0.0 0.0 0.0 1 44
bookmark 0.0 0.0 0.0 1 13
checkerboard 0.0 0.0 0.0 1 14
lasagna 0.0 0.0 0.0 1 2
cassette 0.0 0.0 0.0 1 23
scarecrow 0.0 0.0 0.0 1 1
sparkler_(fireworks) 0.0 0.0 0.0 1 3
mitten 0.0 0.0 0.0 12 28
lawn_mower 0.0 0.0 0.0 3 2
underdrawers 0.0 0.0 0.0 4 19
typewriter 0.0 0.0 0.0 2 5
bullhorn 0.0 0.0 0.0 2 6
duct_tape 0.0 0.0 0.0 3 20
pad 0.0 0.0 0.0 13 1069
seahorse 0.0 0.0 0.0 1 5
dartboard 0.0 0.0 0.0 1 7
razorblade 0.0 0.0 0.0 5 2
cabin_car 0.0 0.0 0.0 3 12
coverall 0.0 0.0 0.0 1 80
bonnet 0.0 0.0 0.0 4 40
jam 0.0 0.0 0.0 2 7
shovel 0.0 0.0 0.0 14 44
ax 0.0 0.0 0.0 2 11
sawhorse 0.0 0.0 0.0 1 11
vulture 0.0 0.0 0.0 1 0
chessboard 0.0 0.0 0.0 1 5
pencil_sharpener 0.0 0.0 0.0 3 1
sharpener 0.0 0.0 0.0 1 12
whistle 0.0 0.0 0.0 1 1
fig_(fruit) 0.0 0.0 0.0 2 23
dustpan 0.0 0.0 0.0 3 10
oil_lamp 0.0 0.0 0.0 1 27
phonebook 0.0 0.0 0.0 1 46
flashlight 0.0 0.0 0.0 6 4
cornet 0.0 0.0 0.0 10 1
cornbread 0.0 0.0 0.0 2 10
roller_skate 0.0 0.0 0.0 1 15
brassiere 0.0 0.0 0.0 20 18
eyepatch 0.0 0.0 0.0 1 12
violin 0.0 0.0 0.0 2 3
subwoofer 0.0 0.0 0.0 3 7
martini 0.0 0.0 0.0 1 99
joystick 0.0 0.0 0.0 1 23
applesauce 0.0 0.0 0.0 1 5
mail_slot 0.0 0.0 0.0 4 2
musical_instrument 0.0 0.0 0.0 1 24
lip_balm 0.0 0.0 0.0 2 16
olive_oil 0.0 0.0 0.0 6 10
birdbath 0.0 0.0 0.0 3 296
gourd 0.0 0.0 0.0 1 28
armor 0.0 0.0 0.0 2 1
space_shuttle 0.0 0.0 0.0 2 6
dental_floss 0.0 0.0 0.0 1 2
flash 0.0 0.0 0.0 2 17
pan_(metal_container) 0.0 0.0 0.0 1 358
sleeping_bag 0.0 0.0 0.0 3 20
limousine 0.0 0.0 0.0 2 1
dropper 0.0 0.0 0.0 2 3
road_map 0.0 0.0 0.0 1 10
strainer 0.0 0.0 0.0 13 17
earplug 0.0 0.0 0.0 2 18
peeler_(tool_for_fruit_and_vegetables) 0.0 0.0 0.0 2 6
elevator_car 0.0 0.0 0.0 1 7
shepherd_dog 0.0 0.0 0.0 1 66
urn 0.0 0.0 0.0 2 40
thermometer 0.0 0.0 0.0 5 22
cigarette_case 0.0 0.0 0.0 4 6
bottle_opener 0.0 0.0 0.0 3 9
water_heater 0.0 0.0 0.0 2 34
recliner 0.0 0.0 0.0 4 53
shaver_(electric) 0.0 0.0 0.0 2 4
blinder_(for_horses) 0.0 0.0 0.0 21 15
cocoa_(beverage) 0.0 0.0 0.0 2 1
nightshirt 0.0 0.0 0.0 3 105
spear 0.0 0.0 0.0 1 5
spider 0.0 0.0 0.0 1 3
clothes_hamper 0.0 0.0 0.0 7 8
needle 0.0 0.0 0.0 3 18
water_scooter 0.0 0.0 0.0 3 0
ferret 0.0 0.0 0.0 1 2
can_opener 0.0 0.0 0.0 4 2
Tabasco_sauce 0.0 0.0 0.0 1 21
lollipop 0.0 0.0 0.0 5 8
thumbtack 0.0 0.0 0.0 3 14
yoke_(animal_equipment) 0.0 0.0 0.0 3 2
cougar 0.0 0.0 0.0 1 7
pretzel 0.0 0.0 0.0 2 11
coatrack 0.0 0.0 0.0 2 19
Bible 0.0 0.0 0.0 1 28
salsa 0.0 0.0 0.0 2 31
fish_(food) 0.0 0.0 0.0 3 1
nosebag_(for_animals) 0.0 0.0 0.0 1 15
potholder 0.0 0.0 0.0 7 41
cufflink 0.0 0.0 0.0 4 0
rabbit 0.0 0.0 0.0 9 10
ironing_board 0.0 0.0 0.0 4 0
handcuff 0.0 0.0 0.0 2 1
shaving_cream 0.0 0.0 0.0 5 4
amplifier 0.0 0.0 0.0 2 7
beret 0.0 0.0 0.0 1 47
clementine 0.0 0.0 0.0 1 65
videotape 0.0 0.0 0.0 2 60
tux 0.0 0.0 0.0 1 8
iron_(for_clothing) 0.0 0.0 0.0 6 5
stirrer 0.0 0.0 0.0 4 33
satchel 0.0 0.0 0.0 1 320
griddle 0.0 0.0 0.0 1 110
pita_(bread) 0.0 0.0 0.0 2 23
bubble_gum 0.0 0.0 0.0 1 5
funnel 0.0 0.0 0.0 2 3
turtleneck_(clothing) 0.0 0.0 0.0 1 15
record_player 0.0 0.0 0.0 4 11
bolo_tie 0.0 0.0 0.0 1 288
tartan 0.0 0.0 0.0 4 65
salmon_(fish) 0.0 0.0 0.0 2 4
stepladder 0.0 0.0 0.0 5 43
hairpin 0.0 0.0 0.0 5 67
machine_gun 0.0 0.0 0.0 1 4
plow_(farm_equipment) 0.0 0.0 0.0 3 0
tinsel 0.0 0.0 0.0 5 33
robe 0.0 0.0 0.0 5 22
cowbell 0.0 0.0 0.0 4 2
hockey_stick 0.0 0.0 0.0 2 13
chap 0.0 0.0 0.0 2 740
piggy_bank 0.0 0.0 0.0 1 12
drumstick 0.0 0.0 0.0 2 5
cast 0.0 0.0 0.0 2 7
dolphin 0.0 0.0 0.0 1 1
kennel 0.0 0.0 0.0 1 8
milkshake 0.0 0.0 0.0 1 12
pennant 0.0 0.0 0.0 1 404
table-tennis_table 0.0 0.0 0.0 2 2
bulldozer 0.0 0.0 0.0 1 16
--------------------------------------------------------------------------------
Macro Avg 0.1703 0.5150 0.2201
================================================================================
shell：cd eval_retrieval
--dataset lvis coco
extract embedding
torchrun --nproc-per-node=8 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=29500
extract_embedding.py --model wedetect
--wedetect_checkpoint wedetect_base.pth
--wedetect_uni_checkpoint wedetect_base_uni.pth
--dataset lvis # coco
retrieval
python3 retrieval_metric.py --model wedetect --dataset lvis --thre 0.2