-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathdatasets.py
187 lines (160 loc) · 16.1 KB
/
datasets.py
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
import os
import torch.nn as nn
from continuum import ClassIncremental, InstanceIncremental
from continuum.datasets import (
CIFAR100, ImageNet100, TinyImageNet200, ImageFolderDataset, Core50
)
from .utils import get_dataset_class_names
from torchvision import transforms
class ImageNet1000(ImageFolderDataset):
"""Continuum dataset for datasets with tree-like structure.
:param train_folder: The folder of the train data.
:param test_folder: The folder of the test data.
:param download: Dummy parameter.
"""
def __init__(
self,
data_path: str,
train: bool = True,
download: bool = False,
):
super().__init__(data_path=data_path, train=train, download=download)
@property
def transformations(self):
"""Default transformations if nothing is provided to the scenario."""
return [
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
]
def get_data(self):
if self.train:
self.data_path = os.path.join(self.data_path, "train")
else:
self.data_path = os.path.join(self.data_path, "val")
return super().get_data()
class ImageNet_R(ImageFolderDataset):
"""Continuum dataset for datasets with tree-like structure.
:param train_folder: The folder of the train data.
:param test_folder: The folder of the test data.
:param download: Dummy parameter.
"""
def __init__(
self,
data_path: str,
train: bool = True,
download: bool = False,
):
super().__init__(data_path=data_path, train=train, download=download)
@property
def transformations(self):
"""Default transformations if nothing is provided to the scenario."""
return [
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
]
def get_data(self):
if self.train:
self.data_path = os.path.join(self.data_path, "train")
else:
self.data_path = os.path.join(self.data_path, "test")
return super().get_data()
def get_dataset(cfg, is_train, transforms=None):
if cfg.dataset == "cifar100":
data_path = os.path.join(cfg.dataset_root, cfg.dataset)
dataset = CIFAR100(
data_path=data_path,
download=True,
train=is_train,
# transforms=transforms
)
classes_names = dataset.dataset.classes
elif cfg.dataset == "tinyimagenet":
data_path = os.path.join(cfg.dataset_root, cfg.dataset)
dataset = TinyImageNet200(
data_path,
train=is_train,
download=True
)
classes_names = get_dataset_class_names(cfg.workdir, cfg.dataset)
elif cfg.dataset == "tiny":
data_path = '/'
dataset = ImageNet100(
data_path,
train=is_train,
data_subset=os.path.join('/', "train_200_random_tiny.txt" if is_train else "val_200_random_tiny.txt")
)
classes_names = ['seashore', 'scoreboard', 'plunger', 'chest', 'Persian_cat', 'candle', 'steel_arch_bridge', 'bathtub', 'fur_coat', 'gondola', 'remote_control', 'oboe', 'barrel', 'Egyptian_cat', 'beach_wagon', 'wok', 'pretzel', 'lesser_panda', 'iPod', 'koala', 'cardigan', 'punching_bag', 'albatross', 'abacus', 'snail', 'convertible', 'chimpanzee', 'mantis', 'pomegranate', 'Labrador_retriever', 'jellyfish', 'dumbbell', 'academic_gown', 'wooden_spoon', 'German_shepherd', 'space_heater', 'pill_bottle', 'kimono', 'sea_slug', 'vestment', 'fountain', 'gasmask', 'brain_coral', 'sea_cucumber', 'espresso', 'lawn_mower', 'sombrero', 'sunglasses', 'stopwatch', 'cockroach', 'sandal', 'refrigerator', 'tarantula', 'Christmas_stocking', 'banana', 'American_lobster', 'cougar', 'potpie', 'torch', 'poncho', 'beacon', 'gazelle', 'go-kart', 'black_widow', 'hog', 'sock', 'bighorn', 'monarch', 'sports_car', 'umbrella', 'altar', 'king_penguin', 'cash_machine', 'tractor', 'fly', 'bell_pepper', 'teddy', 'barbershop', 'moving_van', 'European_fire_salamander', 'birdhouse', 'guacamole', 'hourglass', 'bucket', 'orange', 'comic_book', 'bannister', 'backpack', 'dragonfly', 'crane', 'school_bus', 'brown_bear', 'snorkel', 'thatch', 'picket_fence', 'bullfrog', 'drumstick', 'golden_retriever', 'black_stork', 'goldfish', 'lemon', 'alp', 'trilobite', 'dugong', 'grasshopper', 'tabby', 'cliff', 'police_van', 'scorpion', 'pizza', 'meat_loaf', 'basketball', 'boa_constrictor', 'standard_poodle', 'mushroom', 'African_elephant', 'walking_stick', 'teapot', 'water_tower', 'spider_web', 'binoculars', 'cannon', 'bullet_train', 'lifeboat', 'guinea_pig', 'sulphur_butterfly', 'frying_pan', 'pay-phone', 'flagpole', 'acorn', 'ladybug', 'jinrikisha', 'military_uniform', 'freight_car', 'sewing_machine', 'lakeside', 'bison', 'suspension_bridge', 'beer_bottle', 'lion', 'desk', 'parking_meter', 'broom', 'rugby_ball', 'beaker', 'baboon', 'centipede', 'coral_reef', 'miniskirt', 'projectile', 'swimming_trunks', 'confectionery', 'tailed_frog', 'slug', 'dining_table', 'pop_bottle', 'mashed_potato', 'reel', 'Yorkshire_terrier', 'apron', 'cauliflower', 'Chihuahua', 'computer_keyboard', 'goose', 'spiny_lobster', 'dam', 'butcher_shop', 'pole', 'ox', 'volleyball', 'orangutan', 'triumphal_arch', 'bee', 'barn', 'water_jug', 'ice_lolly', 'turnstile', 'trolleybus', 'cliff_dwelling', 'Arabian_camel', 'bow_tie', 'CD_player', 'nail', 'American_alligator', 'lampshade', 'neck_brace', 'syringe', 'viaduct', 'ice_cream', 'rocking_chair', 'obelisk', 'chain', 'brass', 'magnetic_compass', 'maypole', 'limousine', 'bikini', 'plate', "potter's_wheel", 'organ']
elif cfg.dataset == "imagenet100":
data_path = '/'
dataset = ImageNet100(
data_path,
train=is_train,
data_subset=os.path.join('/', "train_100_dytox.txt" if is_train else "val_100_dytox.txt")
)
classes_names = get_dataset_class_names()
classes_names = ['beer_glass', 'oxcart', 'bearskin', 'drake', 'brass', 'acorn_squash', 'turnstile', 'harvester', 'studio_couch', 'Pomeranian', 'pole', 'cliff_dwelling', 'leaf_beetle', 'titi', 'microphone', 'parachute', 'tobacco_shop', 'bicycle-built-for-two', 'snowplow', 'bassinet', 'Lakeland_terrier', 'flute', 'vacuum', 'jacamar', 'borzoi', 'fire_screen', 'rubber_eraser', 'confectionery', 'tile_roof', 'accordion', 'sidewinder', 'tape_player', 'hand-held_computer', 'school_bus', 'golden_retriever', 'sarong', 'dowitcher', 'ram', 'fireboat', 'birdhouse', 'megalith', 'Italian_greyhound', 'banjo', 'pinwheel', 'Siberian_husky', 'Rottweiler', 'miniature_pinscher', 'swab', 'leafhopper', 'tow_truck', 'sea_snake', 'ice_cream', 'black_and_gold_garden_spider', 'buckeye', 'rocking_chair', 'pelican', 'green_snake', 'English_springer', 'Pekinese', 'patas', 'sleeping_bag', 'vine_snake', 'chain_saw', 'three-toed_sloth', 'beer_bottle', 'Japanese_spaniel', 'king_crab', 'ladle', 'banded_gecko', 'common_newt', 'Norwegian_elkhound', 'rain_barrel', 'balloon', 'walking_stick', 'black-and-tan_coonhound', 'house_finch', 'oxygen_mask', 'acoustic_guitar', 'plate', 'sandbar', 'cock', 'paddlewheel', 'pickup', 'toaster', 'ptarmigan', 'Boston_bull', 'hotdog', 'groom', 'sea_cucumber', 'black-footed_ferret', 'redshank', 'police_van', 'marmoset', 'planetarium', 'cliff', 'totem_pole', 'Great_Pyrenees', 'orange', 'face_powder', 'monarch']
classes_names = ['tench', 'goldfish', 'great_white_shark', 'tiger_shark', 'hammerhead', 'electric_ray', 'stingray', 'cock', 'hen', 'ostrich', 'brambling', 'goldfinch', 'house_finch', 'junco', 'indigo_bunting', 'robin', 'bulbul', 'jay', 'magpie', 'chickadee', 'water_ouzel', 'kite', 'bald_eagle', 'vulture', 'great_grey_owl', 'European_fire_salamander', 'common_newt', 'eft', 'spotted_salamander', 'axolotl', 'bullfrog', 'tree_frog', 'tailed_frog', 'loggerhead', 'leatherback_turtle', 'mud_turtle', 'terrapin', 'box_turtle', 'banded_gecko', 'common_iguana', 'American_chameleon', 'whiptail', 'agama', 'frilled_lizard', 'alligator_lizard', 'Gila_monster', 'green_lizard', 'African_chameleon', 'Komodo_dragon', 'African_crocodile', 'American_alligator', 'triceratops', 'thunder_snake', 'ringneck_snake', 'hognose_snake', 'green_snake', 'king_snake', 'garter_snake', 'water_snake', 'vine_snake', 'night_snake', 'boa_constrictor', 'rock_python', 'Indian_cobra', 'green_mamba', 'sea_snake', 'horned_viper', 'diamondback', 'sidewinder', 'trilobite', 'harvestman', 'scorpion', 'black_and_gold_garden_spider', 'barn_spider', 'garden_spider', 'black_widow', 'tarantula', 'wolf_spider', 'tick', 'centipede', 'black_grouse', 'ptarmigan', 'ruffed_grouse', 'prairie_chicken', 'peacock', 'quail', 'partridge', 'African_grey', 'macaw', 'sulphur-crested_cockatoo', 'lorikeet', 'coucal', 'bee_eater', 'hornbill', 'hummingbird', 'jacamar', 'toucan', 'drake', 'red-breasted_merganser', 'goose']
# import pdb; pdb.set_trace()
elif cfg.dataset == "imageneta":
data_path = '/'
dataset = ImageNet100(
data_path,
train=is_train,
data_subset=os.path.join('/', "train_100_dytox.txt" if is_train else "val_imga_200.txt")
)
imga100 = ['scorpion', 'common_iguana', 'parking_meter', 'schooner', 'lion', 'tarantula', 'basketball', 'mask', 'reel', 'guacamole', 'American_black_bear', 'puffer', 'steam_locomotive', 'wreck', 'Chihuahua', 'sea_lion', 'rapeseed', 'broccoli', 'canoe', 'spider_web', 'bikini', 'hotdog', 'oystercatcher', 'corn', 'marimba', 'flagpole', 'eft', 'puck', 'spatula', 'sundial', 'stingray', 'cello', 'tank', 'cheeseburger', 'bald_eagle', 'leafhopper', 'African_chameleon', 'robin', 'ocarina', 'rhinoceros_beetle', 'vulture', 'pelican', 'chest', "yellow_lady's_slipper", 'grand_piano', 'submarine', 'sleeping_bag', 'sandal', 'envelope', 'bell_pepper', 'sewing_machine', 'fox_squirrel', 'balance_beam', 'kimono', 'banjo', 'bullfrog', 'red_fox', 'beacon', 'mongoose', 'Persian_cat', 'feather_boa', 'go-kart', 'pretzel', 'jay', 'airliner', 'torch', 'fly', 'stethoscope', 'iron', 'suspension_bridge', 'drake', 'junco', 'goldfinch', 'viaduct', 'walking_stick', 'manhole_cover', 'parachute', 'studio_couch', 'mosque', 'cabbage_butterfly', 'forklift', 'doormat', 'American_egret', 'ballplayer', 'chain', 'pool_table', 'harvestman', 'limousine', 'teddy', 'lighter', 'bow_tie', 'ant', 'skunk', 'soap_dispenser', 'acorn', 'goblet', 'bison', 'beaker', 'barn', 'bow', 'goose', 'volcano', 'rugby_ball', 'academic_gown', 'African_elephant', 'mushroom', 'revolver', 'toaster', 'hermit_crab', 'jellyfish', 'organ', 'barrow', 'ladybug', 'snowplow', 'German_shepherd', 'obelisk', 'snowmobile', 'balloon', 'baboon', 'marmot', 'mitten', 'dragonfly', 'fountain', 'piggy_bank', 'jeep', 'apron', 'box_turtle', "jack-o'-lantern", 'dumbbell', 'mantis', 'water_tower', 'volleyball', 'unicycle', 'starfish', 'cowboy_boot', 'cockroach', 'capuchin', 'armadillo', 'custard_apple', 'sulphur-crested_cockatoo', 'agama', 'lemon', 'hand_blower', 'Rottweiler', 'snail', 'porcupine', 'acoustic_guitar', 'flamingo', 'broom', 'banana', 'wood_rabbit', 'nail', 'garter_snake', 'crayfish', 'bee', 'umbrella', 'school_bus', 'rocking_chair', 'washer', 'sax', 'sea_anemone', 'maraca', 'cliff', 'pomegranate', 'golden_retriever', 'American_alligator', 'pug', 'lorikeet', 'ambulance', 'golfcart', 'garbage_truck', 'accordion', 'wine_bottle', 'toucan', 'racket', 'hummingbird', 'centipede', 'koala', 'castle', 'grasshopper', 'monarch', 'Christmas_stocking', 'cradle', 'bubble', 'candle', 'lynx', 'shovel', 'weevil', 'dial_telephone', 'digital_clock', 'lycaenid', 'carbonara', 'breastplate', 'saltshaker', 'flatworm', 'tricycle', 'cucumber', 'drumstick', 'syringe', 'quill']
classes_names = imga100
# import pdb; pdb.set_trace()
elif cfg.dataset == "imagenet1000":
data_path = '/'
dataset = ImageNet1000(
data_path,
train=is_train
)
classes_names = get_dataset_class_names(cfg.workdir, cfg.dataset)
elif cfg.dataset == "imagenet_R":
data_path = '/path/to/imagenet-r/'
dataset = ImageNet_R(
data_path,
train=is_train
)
classes_names = ['goldfish', 'great_white_shark', 'hammerhead', 'stingray', 'hen', 'ostrich', 'goldfinch', 'junco', 'bald_eagle', 'vulture', 'newt', 'axolotl', 'tree_frog', 'iguana', 'African_chameleon', 'cobra', 'scorpion', 'tarantula', 'centipede', 'peacock', 'lorikeet', 'hummingbird', 'toucan', 'duck', 'goose', 'black_swan', 'koala', 'jellyfish', 'snail', 'lobster', 'hermit_crab', 'flamingo', 'american_egret', 'pelican', 'king_penguin', 'grey_whale', 'killer_whale', 'sea_lion', 'chihuahua', 'shih_tzu', 'afghan_hound', 'basset_hound', 'beagle', 'bloodhound', 'italian_greyhound', 'whippet', 'weimaraner', 'yorkshire_terrier', 'boston_terrier', 'scottish_terrier', 'west_highland_white_terrier', 'golden_retriever', 'labrador_retriever', 'cocker_spaniels', 'collie', 'border_collie', 'rottweiler', 'german_shepherd_dog', 'boxer', 'french_bulldog', 'saint_bernard', 'husky', 'dalmatian', 'pug', 'pomeranian', 'chow_chow', 'pembroke_welsh_corgi', 'toy_poodle', 'standard_poodle', 'timber_wolf', 'hyena', 'red_fox', 'tabby_cat', 'leopard', 'snow_leopard', 'lion', 'tiger', 'cheetah', 'polar_bear', 'meerkat', 'ladybug', 'fly', 'bee', 'ant', 'grasshopper', 'cockroach', 'mantis', 'dragonfly', 'monarch_butterfly', 'starfish', 'wood_rabbit', 'porcupine', 'fox_squirrel', 'beaver', 'guinea_pig', 'zebra', 'pig', 'hippopotamus', 'bison', 'gazelle', 'llama', 'skunk', 'badger', 'orangutan', 'gorilla', 'chimpanzee', 'gibbon', 'baboon', 'panda', 'eel', 'clown_fish', 'puffer_fish', 'accordion', 'ambulance', 'assault_rifle', 'backpack', 'barn', 'wheelbarrow', 'basketball', 'bathtub', 'lighthouse', 'beer_glass', 'binoculars', 'birdhouse', 'bow_tie', 'broom', 'bucket', 'cauldron', 'candle', 'cannon', 'canoe', 'carousel', 'castle', 'mobile_phone', 'cowboy_hat', 'electric_guitar', 'fire_engine', 'flute', 'gasmask', 'grand_piano', 'guillotine', 'hammer', 'harmonica', 'harp', 'hatchet', 'jeep', 'joystick', 'lab_coat', 'lawn_mower', 'lipstick', 'mailbox', 'missile', 'mitten', 'parachute', 'pickup_truck', 'pirate_ship', 'revolver', 'rugby_ball', 'sandal', 'saxophone', 'school_bus', 'schooner', 'shield', 'soccer_ball', 'space_shuttle', 'spider_web', 'steam_locomotive', 'scarf', 'submarine', 'tank', 'tennis_ball', 'tractor', 'trombone', 'vase', 'violin', 'military_aircraft', 'wine_bottle', 'ice_cream', 'bagel', 'pretzel', 'cheeseburger', 'hotdog', 'cabbage', 'broccoli', 'cucumber', 'bell_pepper', 'mushroom', 'Granny_Smith', 'strawberry', 'lemon', 'pineapple', 'banana', 'pomegranate', 'pizza', 'burrito', 'espresso', 'volcano', 'baseball_player', 'scuba_diver', 'acorn']
elif cfg.dataset == "core50":
data_path = os.path.join(cfg.dataset_root, cfg.dataset)
dataset = dataset = Core50(
data_path,
scenario="domains",
classification="category",
train=is_train
)
classes_names = [
"plug adapters", "mobile phones", "scissors", "light bulbs", "cans",
"glasses", "balls", "markers", "cups", "remote controls"
]
else:
ValueError(f"'{cfg.dataset}' is a invalid dataset.")
return dataset, classes_names
def build_cl_scenarios(cfg, is_train, transforms) -> nn.Module:
dataset, classes_names = get_dataset(cfg, is_train)
# import pdb; pdb.set_trace()
if cfg.scenario == "class":
scenario = ClassIncremental(
dataset,
initial_increment=cfg.initial_increment,
increment=cfg.increment,
transformations=transforms.transforms, # Convert Compose into list
class_order=cfg.class_order,
)
elif cfg.scenario == "domain":
scenario = InstanceIncremental(
dataset,
transformations=transforms.transforms,
)
elif cfg.scenario == "task-agnostic":
NotImplementedError("Method has not been implemented. Soon be added.")
else:
ValueError(f"You have entered `{cfg.scenario}` which is not a defined scenario, "
"please choose from {{'class', 'domain', 'task-agnostic'}}.")
return scenario, classes_names