Commit bf491463 authored by limm's avatar limm
Browse files

add v0.19.1 release

parent e17f5ea2
{"0": ["n01440764", "tench"], "1": ["n01443537", "goldfish"], "2": ["n01484850", "great_white_shark"], "3": ["n01491361", "tiger_shark"], "4": ["n01494475", "hammerhead"], "5": ["n01496331", "electric_ray"], "6": ["n01498041", "stingray"], "7": ["n01514668", "cock"], "8": ["n01514859", "hen"], "9": ["n01518878", "ostrich"], "10": ["n01530575", "brambling"], "11": ["n01531178", "goldfinch"], "12": ["n01532829", "house_finch"], "13": ["n01534433", "junco"], "14": ["n01537544", "indigo_bunting"], "15": ["n01558993", "robin"], "16": ["n01560419", "bulbul"], "17": ["n01580077", "jay"], "18": ["n01582220", "magpie"], "19": ["n01592084", "chickadee"], "20": ["n01601694", "water_ouzel"], "21": ["n01608432", "kite"], "22": ["n01614925", "bald_eagle"], "23": ["n01616318", "vulture"], "24": ["n01622779", "great_grey_owl"], "25": ["n01629819", "European_fire_salamander"], "26": ["n01630670", "common_newt"], "27": ["n01631663", "eft"], "28": ["n01632458", "spotted_salamander"], "29": ["n01632777", "axolotl"], "30": ["n01641577", "bullfrog"], "31": ["n01644373", "tree_frog"], "32": ["n01644900", "tailed_frog"], "33": ["n01664065", "loggerhead"], "34": ["n01665541", "leatherback_turtle"], "35": ["n01667114", "mud_turtle"], "36": ["n01667778", "terrapin"], "37": ["n01669191", "box_turtle"], "38": ["n01675722", "banded_gecko"], "39": ["n01677366", "common_iguana"], "40": ["n01682714", "American_chameleon"], "41": ["n01685808", "whiptail"], "42": ["n01687978", "agama"], "43": ["n01688243", "frilled_lizard"], "44": ["n01689811", "alligator_lizard"], "45": ["n01692333", "Gila_monster"], "46": ["n01693334", "green_lizard"], "47": ["n01694178", "African_chameleon"], "48": ["n01695060", "Komodo_dragon"], "49": ["n01697457", "African_crocodile"], "50": ["n01698640", "American_alligator"], "51": ["n01704323", "triceratops"], "52": ["n01728572", "thunder_snake"], "53": ["n01728920", "ringneck_snake"], "54": ["n01729322", "hognose_snake"], "55": ["n01729977", "green_snake"], "56": ["n01734418", "king_snake"], "57": ["n01735189", "garter_snake"], "58": ["n01737021", "water_snake"], "59": ["n01739381", "vine_snake"], "60": ["n01740131", "night_snake"], "61": ["n01742172", "boa_constrictor"], "62": ["n01744401", "rock_python"], "63": ["n01748264", "Indian_cobra"], "64": ["n01749939", "green_mamba"], "65": ["n01751748", "sea_snake"], "66": ["n01753488", "horned_viper"], "67": ["n01755581", "diamondback"], "68": ["n01756291", "sidewinder"], "69": ["n01768244", "trilobite"], "70": ["n01770081", "harvestman"], "71": ["n01770393", "scorpion"], "72": ["n01773157", "black_and_gold_garden_spider"], "73": ["n01773549", "barn_spider"], "74": ["n01773797", "garden_spider"], "75": ["n01774384", "black_widow"], "76": ["n01774750", "tarantula"], "77": ["n01775062", "wolf_spider"], "78": ["n01776313", "tick"], "79": ["n01784675", "centipede"], "80": ["n01795545", "black_grouse"], "81": ["n01796340", "ptarmigan"], "82": ["n01797886", "ruffed_grouse"], "83": ["n01798484", "prairie_chicken"], "84": ["n01806143", "peacock"], "85": ["n01806567", "quail"], "86": ["n01807496", "partridge"], "87": ["n01817953", "African_grey"], "88": ["n01818515", "macaw"], "89": ["n01819313", "sulphur-crested_cockatoo"], "90": ["n01820546", "lorikeet"], "91": ["n01824575", "coucal"], "92": ["n01828970", "bee_eater"], "93": ["n01829413", "hornbill"], "94": ["n01833805", "hummingbird"], "95": ["n01843065", "jacamar"], "96": ["n01843383", "toucan"], "97": ["n01847000", "drake"], "98": ["n01855032", "red-breasted_merganser"], "99": ["n01855672", "goose"], "100": ["n01860187", "black_swan"], "101": ["n01871265", "tusker"], "102": ["n01872401", "echidna"], "103": ["n01873310", "platypus"], "104": ["n01877812", "wallaby"], "105": ["n01882714", "koala"], "106": ["n01883070", "wombat"], "107": ["n01910747", "jellyfish"], "108": ["n01914609", "sea_anemone"], "109": ["n01917289", "brain_coral"], "110": ["n01924916", "flatworm"], "111": ["n01930112", "nematode"], "112": ["n01943899", "conch"], "113": ["n01944390", "snail"], "114": ["n01945685", "slug"], "115": ["n01950731", "sea_slug"], "116": ["n01955084", "chiton"], "117": ["n01968897", "chambered_nautilus"], "118": ["n01978287", "Dungeness_crab"], "119": ["n01978455", "rock_crab"], "120": ["n01980166", "fiddler_crab"], "121": ["n01981276", "king_crab"], "122": ["n01983481", "American_lobster"], "123": ["n01984695", "spiny_lobster"], "124": ["n01985128", "crayfish"], "125": ["n01986214", "hermit_crab"], "126": ["n01990800", "isopod"], "127": ["n02002556", "white_stork"], "128": ["n02002724", "black_stork"], "129": ["n02006656", "spoonbill"], "130": ["n02007558", "flamingo"], "131": ["n02009229", "little_blue_heron"], "132": ["n02009912", "American_egret"], "133": ["n02011460", "bittern"], "134": ["n02012849", "crane"], "135": ["n02013706", "limpkin"], "136": ["n02017213", "European_gallinule"], "137": ["n02018207", "American_coot"], "138": ["n02018795", "bustard"], "139": ["n02025239", "ruddy_turnstone"], "140": ["n02027492", "red-backed_sandpiper"], "141": ["n02028035", "redshank"], "142": ["n02033041", "dowitcher"], "143": ["n02037110", "oystercatcher"], "144": ["n02051845", "pelican"], "145": ["n02056570", "king_penguin"], "146": ["n02058221", "albatross"], "147": ["n02066245", "grey_whale"], "148": ["n02071294", "killer_whale"], "149": ["n02074367", "dugong"], "150": ["n02077923", "sea_lion"], "151": ["n02085620", "Chihuahua"], "152": ["n02085782", "Japanese_spaniel"], "153": ["n02085936", "Maltese_dog"], "154": ["n02086079", "Pekinese"], "155": ["n02086240", "Shih-Tzu"], "156": ["n02086646", "Blenheim_spaniel"], "157": ["n02086910", "papillon"], "158": ["n02087046", "toy_terrier"], "159": ["n02087394", "Rhodesian_ridgeback"], "160": ["n02088094", "Afghan_hound"], "161": ["n02088238", "basset"], "162": ["n02088364", "beagle"], "163": ["n02088466", "bloodhound"], "164": ["n02088632", "bluetick"], "165": ["n02089078", "black-and-tan_coonhound"], "166": ["n02089867", "Walker_hound"], "167": ["n02089973", "English_foxhound"], "168": ["n02090379", "redbone"], "169": ["n02090622", "borzoi"], "170": ["n02090721", "Irish_wolfhound"], "171": ["n02091032", "Italian_greyhound"], "172": ["n02091134", "whippet"], "173": ["n02091244", "Ibizan_hound"], "174": ["n02091467", "Norwegian_elkhound"], "175": ["n02091635", "otterhound"], "176": ["n02091831", "Saluki"], "177": ["n02092002", "Scottish_deerhound"], "178": ["n02092339", "Weimaraner"], "179": ["n02093256", "Staffordshire_bullterrier"], "180": ["n02093428", "American_Staffordshire_terrier"], "181": ["n02093647", "Bedlington_terrier"], "182": ["n02093754", "Border_terrier"], "183": ["n02093859", "Kerry_blue_terrier"], "184": ["n02093991", "Irish_terrier"], "185": ["n02094114", "Norfolk_terrier"], "186": ["n02094258", "Norwich_terrier"], "187": ["n02094433", "Yorkshire_terrier"], "188": ["n02095314", "wire-haired_fox_terrier"], "189": ["n02095570", "Lakeland_terrier"], "190": ["n02095889", "Sealyham_terrier"], "191": ["n02096051", "Airedale"], "192": ["n02096177", "cairn"], "193": ["n02096294", "Australian_terrier"], "194": ["n02096437", "Dandie_Dinmont"], "195": ["n02096585", "Boston_bull"], "196": ["n02097047", "miniature_schnauzer"], "197": ["n02097130", "giant_schnauzer"], "198": ["n02097209", "standard_schnauzer"], "199": ["n02097298", "Scotch_terrier"], "200": ["n02097474", "Tibetan_terrier"], "201": ["n02097658", "silky_terrier"], "202": ["n02098105", "soft-coated_wheaten_terrier"], "203": ["n02098286", "West_Highland_white_terrier"], "204": ["n02098413", "Lhasa"], "205": ["n02099267", "flat-coated_retriever"], "206": ["n02099429", "curly-coated_retriever"], "207": ["n02099601", "golden_retriever"], "208": ["n02099712", "Labrador_retriever"], "209": ["n02099849", "Chesapeake_Bay_retriever"], "210": ["n02100236", "German_short-haired_pointer"], "211": ["n02100583", "vizsla"], "212": ["n02100735", "English_setter"], "213": ["n02100877", "Irish_setter"], "214": ["n02101006", "Gordon_setter"], "215": ["n02101388", "Brittany_spaniel"], "216": ["n02101556", "clumber"], "217": ["n02102040", "English_springer"], "218": ["n02102177", "Welsh_springer_spaniel"], "219": ["n02102318", "cocker_spaniel"], "220": ["n02102480", "Sussex_spaniel"], "221": ["n02102973", "Irish_water_spaniel"], "222": ["n02104029", "kuvasz"], "223": ["n02104365", "schipperke"], "224": ["n02105056", "groenendael"], "225": ["n02105162", "malinois"], "226": ["n02105251", "briard"], "227": ["n02105412", "kelpie"], "228": ["n02105505", "komondor"], "229": ["n02105641", "Old_English_sheepdog"], "230": ["n02105855", "Shetland_sheepdog"], "231": ["n02106030", "collie"], "232": ["n02106166", "Border_collie"], "233": ["n02106382", "Bouvier_des_Flandres"], "234": ["n02106550", "Rottweiler"], "235": ["n02106662", "German_shepherd"], "236": ["n02107142", "Doberman"], "237": ["n02107312", "miniature_pinscher"], "238": ["n02107574", "Greater_Swiss_Mountain_dog"], "239": ["n02107683", "Bernese_mountain_dog"], "240": ["n02107908", "Appenzeller"], "241": ["n02108000", "EntleBucher"], "242": ["n02108089", "boxer"], "243": ["n02108422", "bull_mastiff"], "244": ["n02108551", "Tibetan_mastiff"], "245": ["n02108915", "French_bulldog"], "246": ["n02109047", "Great_Dane"], "247": ["n02109525", "Saint_Bernard"], "248": ["n02109961", "Eskimo_dog"], "249": ["n02110063", "malamute"], "250": ["n02110185", "Siberian_husky"], "251": ["n02110341", "dalmatian"], "252": ["n02110627", "affenpinscher"], "253": ["n02110806", "basenji"], "254": ["n02110958", "pug"], "255": ["n02111129", "Leonberg"], "256": ["n02111277", "Newfoundland"], "257": ["n02111500", "Great_Pyrenees"], "258": ["n02111889", "Samoyed"], "259": ["n02112018", "Pomeranian"], "260": ["n02112137", "chow"], "261": ["n02112350", "keeshond"], "262": ["n02112706", "Brabancon_griffon"], "263": ["n02113023", "Pembroke"], "264": ["n02113186", "Cardigan"], "265": ["n02113624", "toy_poodle"], "266": ["n02113712", "miniature_poodle"], "267": ["n02113799", "standard_poodle"], "268": ["n02113978", "Mexican_hairless"], "269": ["n02114367", "timber_wolf"], "270": ["n02114548", "white_wolf"], "271": ["n02114712", "red_wolf"], "272": ["n02114855", "coyote"], "273": ["n02115641", "dingo"], "274": ["n02115913", "dhole"], "275": ["n02116738", "African_hunting_dog"], "276": ["n02117135", "hyena"], "277": ["n02119022", "red_fox"], "278": ["n02119789", "kit_fox"], "279": ["n02120079", "Arctic_fox"], "280": ["n02120505", "grey_fox"], "281": ["n02123045", "tabby"], "282": ["n02123159", "tiger_cat"], "283": ["n02123394", "Persian_cat"], "284": ["n02123597", "Siamese_cat"], "285": ["n02124075", "Egyptian_cat"], "286": ["n02125311", "cougar"], "287": ["n02127052", "lynx"], "288": ["n02128385", "leopard"], "289": ["n02128757", "snow_leopard"], "290": ["n02128925", "jaguar"], "291": ["n02129165", "lion"], "292": ["n02129604", "tiger"], "293": ["n02130308", "cheetah"], "294": ["n02132136", "brown_bear"], "295": ["n02133161", "American_black_bear"], "296": ["n02134084", "ice_bear"], "297": ["n02134418", "sloth_bear"], "298": ["n02137549", "mongoose"], "299": ["n02138441", "meerkat"], "300": ["n02165105", "tiger_beetle"], "301": ["n02165456", "ladybug"], "302": ["n02167151", "ground_beetle"], "303": ["n02168699", "long-horned_beetle"], "304": ["n02169497", "leaf_beetle"], "305": ["n02172182", "dung_beetle"], "306": ["n02174001", "rhinoceros_beetle"], "307": ["n02177972", "weevil"], "308": ["n02190166", "fly"], "309": ["n02206856", "bee"], "310": ["n02219486", "ant"], "311": ["n02226429", "grasshopper"], "312": ["n02229544", "cricket"], "313": ["n02231487", "walking_stick"], "314": ["n02233338", "cockroach"], "315": ["n02236044", "mantis"], "316": ["n02256656", "cicada"], "317": ["n02259212", "leafhopper"], "318": ["n02264363", "lacewing"], "319": ["n02268443", "dragonfly"], "320": ["n02268853", "damselfly"], "321": ["n02276258", "admiral"], "322": ["n02277742", "ringlet"], "323": ["n02279972", "monarch"], "324": ["n02280649", "cabbage_butterfly"], "325": ["n02281406", "sulphur_butterfly"], "326": ["n02281787", "lycaenid"], "327": ["n02317335", "starfish"], "328": ["n02319095", "sea_urchin"], "329": ["n02321529", "sea_cucumber"], "330": ["n02325366", "wood_rabbit"], "331": ["n02326432", "hare"], "332": ["n02328150", "Angora"], "333": ["n02342885", "hamster"], "334": ["n02346627", "porcupine"], "335": ["n02356798", "fox_squirrel"], "336": ["n02361337", "marmot"], "337": ["n02363005", "beaver"], "338": ["n02364673", "guinea_pig"], "339": ["n02389026", "sorrel"], "340": ["n02391049", "zebra"], "341": ["n02395406", "hog"], "342": ["n02396427", "wild_boar"], "343": ["n02397096", "warthog"], "344": ["n02398521", "hippopotamus"], "345": ["n02403003", "ox"], "346": ["n02408429", "water_buffalo"], "347": ["n02410509", "bison"], "348": ["n02412080", "ram"], "349": ["n02415577", "bighorn"], "350": ["n02417914", "ibex"], "351": ["n02422106", "hartebeest"], "352": ["n02422699", "impala"], "353": ["n02423022", "gazelle"], "354": ["n02437312", "Arabian_camel"], "355": ["n02437616", "llama"], "356": ["n02441942", "weasel"], "357": ["n02442845", "mink"], "358": ["n02443114", "polecat"], "359": ["n02443484", "black-footed_ferret"], "360": ["n02444819", "otter"], "361": ["n02445715", "skunk"], "362": ["n02447366", "badger"], "363": ["n02454379", "armadillo"], "364": ["n02457408", "three-toed_sloth"], "365": ["n02480495", "orangutan"], "366": ["n02480855", "gorilla"], "367": ["n02481823", "chimpanzee"], "368": ["n02483362", "gibbon"], "369": ["n02483708", "siamang"], "370": ["n02484975", "guenon"], "371": ["n02486261", "patas"], "372": ["n02486410", "baboon"], "373": ["n02487347", "macaque"], "374": ["n02488291", "langur"], "375": ["n02488702", "colobus"], "376": ["n02489166", "proboscis_monkey"], "377": ["n02490219", "marmoset"], "378": ["n02492035", "capuchin"], "379": ["n02492660", "howler_monkey"], "380": ["n02493509", "titi"], "381": ["n02493793", "spider_monkey"], "382": ["n02494079", "squirrel_monkey"], "383": ["n02497673", "Madagascar_cat"], "384": ["n02500267", "indri"], "385": ["n02504013", "Indian_elephant"], "386": ["n02504458", "African_elephant"], "387": ["n02509815", "lesser_panda"], "388": ["n02510455", "giant_panda"], "389": ["n02514041", "barracouta"], "390": ["n02526121", "eel"], "391": ["n02536864", "coho"], "392": ["n02606052", "rock_beauty"], "393": ["n02607072", "anemone_fish"], "394": ["n02640242", "sturgeon"], "395": ["n02641379", "gar"], "396": ["n02643566", "lionfish"], "397": ["n02655020", "puffer"], "398": ["n02666196", "abacus"], "399": ["n02667093", "abaya"], "400": ["n02669723", "academic_gown"], "401": ["n02672831", "accordion"], "402": ["n02676566", "acoustic_guitar"], "403": ["n02687172", "aircraft_carrier"], "404": ["n02690373", "airliner"], "405": ["n02692877", "airship"], "406": ["n02699494", "altar"], "407": ["n02701002", "ambulance"], "408": ["n02704792", "amphibian"], "409": ["n02708093", "analog_clock"], "410": ["n02727426", "apiary"], "411": ["n02730930", "apron"], "412": ["n02747177", "ashcan"], "413": ["n02749479", "assault_rifle"], "414": ["n02769748", "backpack"], "415": ["n02776631", "bakery"], "416": ["n02777292", "balance_beam"], "417": ["n02782093", "balloon"], "418": ["n02783161", "ballpoint"], "419": ["n02786058", "Band_Aid"], "420": ["n02787622", "banjo"], "421": ["n02788148", "bannister"], "422": ["n02790996", "barbell"], "423": ["n02791124", "barber_chair"], "424": ["n02791270", "barbershop"], "425": ["n02793495", "barn"], "426": ["n02794156", "barometer"], "427": ["n02795169", "barrel"], "428": ["n02797295", "barrow"], "429": ["n02799071", "baseball"], "430": ["n02802426", "basketball"], "431": ["n02804414", "bassinet"], "432": ["n02804610", "bassoon"], "433": ["n02807133", "bathing_cap"], "434": ["n02808304", "bath_towel"], "435": ["n02808440", "bathtub"], "436": ["n02814533", "beach_wagon"], "437": ["n02814860", "beacon"], "438": ["n02815834", "beaker"], "439": ["n02817516", "bearskin"], "440": ["n02823428", "beer_bottle"], "441": ["n02823750", "beer_glass"], "442": ["n02825657", "bell_cote"], "443": ["n02834397", "bib"], "444": ["n02835271", "bicycle-built-for-two"], "445": ["n02837789", "bikini"], "446": ["n02840245", "binder"], "447": ["n02841315", "binoculars"], "448": ["n02843684", "birdhouse"], "449": ["n02859443", "boathouse"], "450": ["n02860847", "bobsled"], "451": ["n02865351", "bolo_tie"], "452": ["n02869837", "bonnet"], "453": ["n02870880", "bookcase"], "454": ["n02871525", "bookshop"], "455": ["n02877765", "bottlecap"], "456": ["n02879718", "bow"], "457": ["n02883205", "bow_tie"], "458": ["n02892201", "brass"], "459": ["n02892767", "brassiere"], "460": ["n02894605", "breakwater"], "461": ["n02895154", "breastplate"], "462": ["n02906734", "broom"], "463": ["n02909870", "bucket"], "464": ["n02910353", "buckle"], "465": ["n02916936", "bulletproof_vest"], "466": ["n02917067", "bullet_train"], "467": ["n02927161", "butcher_shop"], "468": ["n02930766", "cab"], "469": ["n02939185", "caldron"], "470": ["n02948072", "candle"], "471": ["n02950826", "cannon"], "472": ["n02951358", "canoe"], "473": ["n02951585", "can_opener"], "474": ["n02963159", "cardigan"], "475": ["n02965783", "car_mirror"], "476": ["n02966193", "carousel"], "477": ["n02966687", "carpenter's_kit"], "478": ["n02971356", "carton"], "479": ["n02974003", "car_wheel"], "480": ["n02977058", "cash_machine"], "481": ["n02978881", "cassette"], "482": ["n02979186", "cassette_player"], "483": ["n02980441", "castle"], "484": ["n02981792", "catamaran"], "485": ["n02988304", "CD_player"], "486": ["n02992211", "cello"], "487": ["n02992529", "cellular_telephone"], "488": ["n02999410", "chain"], "489": ["n03000134", "chainlink_fence"], "490": ["n03000247", "chain_mail"], "491": ["n03000684", "chain_saw"], "492": ["n03014705", "chest"], "493": ["n03016953", "chiffonier"], "494": ["n03017168", "chime"], "495": ["n03018349", "china_cabinet"], "496": ["n03026506", "Christmas_stocking"], "497": ["n03028079", "church"], "498": ["n03032252", "cinema"], "499": ["n03041632", "cleaver"], "500": ["n03042490", "cliff_dwelling"], "501": ["n03045698", "cloak"], "502": ["n03047690", "clog"], "503": ["n03062245", "cocktail_shaker"], "504": ["n03063599", "coffee_mug"], "505": ["n03063689", "coffeepot"], "506": ["n03065424", "coil"], "507": ["n03075370", "combination_lock"], "508": ["n03085013", "computer_keyboard"], "509": ["n03089624", "confectionery"], "510": ["n03095699", "container_ship"], "511": ["n03100240", "convertible"], "512": ["n03109150", "corkscrew"], "513": ["n03110669", "cornet"], "514": ["n03124043", "cowboy_boot"], "515": ["n03124170", "cowboy_hat"], "516": ["n03125729", "cradle"], "517": ["n03126707", "crane"], "518": ["n03127747", "crash_helmet"], "519": ["n03127925", "crate"], "520": ["n03131574", "crib"], "521": ["n03133878", "Crock_Pot"], "522": ["n03134739", "croquet_ball"], "523": ["n03141823", "crutch"], "524": ["n03146219", "cuirass"], "525": ["n03160309", "dam"], "526": ["n03179701", "desk"], "527": ["n03180011", "desktop_computer"], "528": ["n03187595", "dial_telephone"], "529": ["n03188531", "diaper"], "530": ["n03196217", "digital_clock"], "531": ["n03197337", "digital_watch"], "532": ["n03201208", "dining_table"], "533": ["n03207743", "dishrag"], "534": ["n03207941", "dishwasher"], "535": ["n03208938", "disk_brake"], "536": ["n03216828", "dock"], "537": ["n03218198", "dogsled"], "538": ["n03220513", "dome"], "539": ["n03223299", "doormat"], "540": ["n03240683", "drilling_platform"], "541": ["n03249569", "drum"], "542": ["n03250847", "drumstick"], "543": ["n03255030", "dumbbell"], "544": ["n03259280", "Dutch_oven"], "545": ["n03271574", "electric_fan"], "546": ["n03272010", "electric_guitar"], "547": ["n03272562", "electric_locomotive"], "548": ["n03290653", "entertainment_center"], "549": ["n03291819", "envelope"], "550": ["n03297495", "espresso_maker"], "551": ["n03314780", "face_powder"], "552": ["n03325584", "feather_boa"], "553": ["n03337140", "file"], "554": ["n03344393", "fireboat"], "555": ["n03345487", "fire_engine"], "556": ["n03347037", "fire_screen"], "557": ["n03355925", "flagpole"], "558": ["n03372029", "flute"], "559": ["n03376595", "folding_chair"], "560": ["n03379051", "football_helmet"], "561": ["n03384352", "forklift"], "562": ["n03388043", "fountain"], "563": ["n03388183", "fountain_pen"], "564": ["n03388549", "four-poster"], "565": ["n03393912", "freight_car"], "566": ["n03394916", "French_horn"], "567": ["n03400231", "frying_pan"], "568": ["n03404251", "fur_coat"], "569": ["n03417042", "garbage_truck"], "570": ["n03424325", "gasmask"], "571": ["n03425413", "gas_pump"], "572": ["n03443371", "goblet"], "573": ["n03444034", "go-kart"], "574": ["n03445777", "golf_ball"], "575": ["n03445924", "golfcart"], "576": ["n03447447", "gondola"], "577": ["n03447721", "gong"], "578": ["n03450230", "gown"], "579": ["n03452741", "grand_piano"], "580": ["n03457902", "greenhouse"], "581": ["n03459775", "grille"], "582": ["n03461385", "grocery_store"], "583": ["n03467068", "guillotine"], "584": ["n03476684", "hair_slide"], "585": ["n03476991", "hair_spray"], "586": ["n03478589", "half_track"], "587": ["n03481172", "hammer"], "588": ["n03482405", "hamper"], "589": ["n03483316", "hand_blower"], "590": ["n03485407", "hand-held_computer"], "591": ["n03485794", "handkerchief"], "592": ["n03492542", "hard_disc"], "593": ["n03494278", "harmonica"], "594": ["n03495258", "harp"], "595": ["n03496892", "harvester"], "596": ["n03498962", "hatchet"], "597": ["n03527444", "holster"], "598": ["n03529860", "home_theater"], "599": ["n03530642", "honeycomb"], "600": ["n03532672", "hook"], "601": ["n03534580", "hoopskirt"], "602": ["n03535780", "horizontal_bar"], "603": ["n03538406", "horse_cart"], "604": ["n03544143", "hourglass"], "605": ["n03584254", "iPod"], "606": ["n03584829", "iron"], "607": ["n03590841", "jack-o'-lantern"], "608": ["n03594734", "jean"], "609": ["n03594945", "jeep"], "610": ["n03595614", "jersey"], "611": ["n03598930", "jigsaw_puzzle"], "612": ["n03599486", "jinrikisha"], "613": ["n03602883", "joystick"], "614": ["n03617480", "kimono"], "615": ["n03623198", "knee_pad"], "616": ["n03627232", "knot"], "617": ["n03630383", "lab_coat"], "618": ["n03633091", "ladle"], "619": ["n03637318", "lampshade"], "620": ["n03642806", "laptop"], "621": ["n03649909", "lawn_mower"], "622": ["n03657121", "lens_cap"], "623": ["n03658185", "letter_opener"], "624": ["n03661043", "library"], "625": ["n03662601", "lifeboat"], "626": ["n03666591", "lighter"], "627": ["n03670208", "limousine"], "628": ["n03673027", "liner"], "629": ["n03676483", "lipstick"], "630": ["n03680355", "Loafer"], "631": ["n03690938", "lotion"], "632": ["n03691459", "loudspeaker"], "633": ["n03692522", "loupe"], "634": ["n03697007", "lumbermill"], "635": ["n03706229", "magnetic_compass"], "636": ["n03709823", "mailbag"], "637": ["n03710193", "mailbox"], "638": ["n03710637", "maillot"], "639": ["n03710721", "maillot"], "640": ["n03717622", "manhole_cover"], "641": ["n03720891", "maraca"], "642": ["n03721384", "marimba"], "643": ["n03724870", "mask"], "644": ["n03729826", "matchstick"], "645": ["n03733131", "maypole"], "646": ["n03733281", "maze"], "647": ["n03733805", "measuring_cup"], "648": ["n03742115", "medicine_chest"], "649": ["n03743016", "megalith"], "650": ["n03759954", "microphone"], "651": ["n03761084", "microwave"], "652": ["n03763968", "military_uniform"], "653": ["n03764736", "milk_can"], "654": ["n03769881", "minibus"], "655": ["n03770439", "miniskirt"], "656": ["n03770679", "minivan"], "657": ["n03773504", "missile"], "658": ["n03775071", "mitten"], "659": ["n03775546", "mixing_bowl"], "660": ["n03776460", "mobile_home"], "661": ["n03777568", "Model_T"], "662": ["n03777754", "modem"], "663": ["n03781244", "monastery"], "664": ["n03782006", "monitor"], "665": ["n03785016", "moped"], "666": ["n03786901", "mortar"], "667": ["n03787032", "mortarboard"], "668": ["n03788195", "mosque"], "669": ["n03788365", "mosquito_net"], "670": ["n03791053", "motor_scooter"], "671": ["n03792782", "mountain_bike"], "672": ["n03792972", "mountain_tent"], "673": ["n03793489", "mouse"], "674": ["n03794056", "mousetrap"], "675": ["n03796401", "moving_van"], "676": ["n03803284", "muzzle"], "677": ["n03804744", "nail"], "678": ["n03814639", "neck_brace"], "679": ["n03814906", "necklace"], "680": ["n03825788", "nipple"], "681": ["n03832673", "notebook"], "682": ["n03837869", "obelisk"], "683": ["n03838899", "oboe"], "684": ["n03840681", "ocarina"], "685": ["n03841143", "odometer"], "686": ["n03843555", "oil_filter"], "687": ["n03854065", "organ"], "688": ["n03857828", "oscilloscope"], "689": ["n03866082", "overskirt"], "690": ["n03868242", "oxcart"], "691": ["n03868863", "oxygen_mask"], "692": ["n03871628", "packet"], "693": ["n03873416", "paddle"], "694": ["n03874293", "paddlewheel"], "695": ["n03874599", "padlock"], "696": ["n03876231", "paintbrush"], "697": ["n03877472", "pajama"], "698": ["n03877845", "palace"], "699": ["n03884397", "panpipe"], "700": ["n03887697", "paper_towel"], "701": ["n03888257", "parachute"], "702": ["n03888605", "parallel_bars"], "703": ["n03891251", "park_bench"], "704": ["n03891332", "parking_meter"], "705": ["n03895866", "passenger_car"], "706": ["n03899768", "patio"], "707": ["n03902125", "pay-phone"], "708": ["n03903868", "pedestal"], "709": ["n03908618", "pencil_box"], "710": ["n03908714", "pencil_sharpener"], "711": ["n03916031", "perfume"], "712": ["n03920288", "Petri_dish"], "713": ["n03924679", "photocopier"], "714": ["n03929660", "pick"], "715": ["n03929855", "pickelhaube"], "716": ["n03930313", "picket_fence"], "717": ["n03930630", "pickup"], "718": ["n03933933", "pier"], "719": ["n03935335", "piggy_bank"], "720": ["n03937543", "pill_bottle"], "721": ["n03938244", "pillow"], "722": ["n03942813", "ping-pong_ball"], "723": ["n03944341", "pinwheel"], "724": ["n03947888", "pirate"], "725": ["n03950228", "pitcher"], "726": ["n03954731", "plane"], "727": ["n03956157", "planetarium"], "728": ["n03958227", "plastic_bag"], "729": ["n03961711", "plate_rack"], "730": ["n03967562", "plow"], "731": ["n03970156", "plunger"], "732": ["n03976467", "Polaroid_camera"], "733": ["n03976657", "pole"], "734": ["n03977966", "police_van"], "735": ["n03980874", "poncho"], "736": ["n03982430", "pool_table"], "737": ["n03983396", "pop_bottle"], "738": ["n03991062", "pot"], "739": ["n03992509", "potter's_wheel"], "740": ["n03995372", "power_drill"], "741": ["n03998194", "prayer_rug"], "742": ["n04004767", "printer"], "743": ["n04005630", "prison"], "744": ["n04008634", "projectile"], "745": ["n04009552", "projector"], "746": ["n04019541", "puck"], "747": ["n04023962", "punching_bag"], "748": ["n04026417", "purse"], "749": ["n04033901", "quill"], "750": ["n04033995", "quilt"], "751": ["n04037443", "racer"], "752": ["n04039381", "racket"], "753": ["n04040759", "radiator"], "754": ["n04041544", "radio"], "755": ["n04044716", "radio_telescope"], "756": ["n04049303", "rain_barrel"], "757": ["n04065272", "recreational_vehicle"], "758": ["n04067472", "reel"], "759": ["n04069434", "reflex_camera"], "760": ["n04070727", "refrigerator"], "761": ["n04074963", "remote_control"], "762": ["n04081281", "restaurant"], "763": ["n04086273", "revolver"], "764": ["n04090263", "rifle"], "765": ["n04099969", "rocking_chair"], "766": ["n04111531", "rotisserie"], "767": ["n04116512", "rubber_eraser"], "768": ["n04118538", "rugby_ball"], "769": ["n04118776", "rule"], "770": ["n04120489", "running_shoe"], "771": ["n04125021", "safe"], "772": ["n04127249", "safety_pin"], "773": ["n04131690", "saltshaker"], "774": ["n04133789", "sandal"], "775": ["n04136333", "sarong"], "776": ["n04141076", "sax"], "777": ["n04141327", "scabbard"], "778": ["n04141975", "scale"], "779": ["n04146614", "school_bus"], "780": ["n04147183", "schooner"], "781": ["n04149813", "scoreboard"], "782": ["n04152593", "screen"], "783": ["n04153751", "screw"], "784": ["n04154565", "screwdriver"], "785": ["n04162706", "seat_belt"], "786": ["n04179913", "sewing_machine"], "787": ["n04192698", "shield"], "788": ["n04200800", "shoe_shop"], "789": ["n04201297", "shoji"], "790": ["n04204238", "shopping_basket"], "791": ["n04204347", "shopping_cart"], "792": ["n04208210", "shovel"], "793": ["n04209133", "shower_cap"], "794": ["n04209239", "shower_curtain"], "795": ["n04228054", "ski"], "796": ["n04229816", "ski_mask"], "797": ["n04235860", "sleeping_bag"], "798": ["n04238763", "slide_rule"], "799": ["n04239074", "sliding_door"], "800": ["n04243546", "slot"], "801": ["n04251144", "snorkel"], "802": ["n04252077", "snowmobile"], "803": ["n04252225", "snowplow"], "804": ["n04254120", "soap_dispenser"], "805": ["n04254680", "soccer_ball"], "806": ["n04254777", "sock"], "807": ["n04258138", "solar_dish"], "808": ["n04259630", "sombrero"], "809": ["n04263257", "soup_bowl"], "810": ["n04264628", "space_bar"], "811": ["n04265275", "space_heater"], "812": ["n04266014", "space_shuttle"], "813": ["n04270147", "spatula"], "814": ["n04273569", "speedboat"], "815": ["n04275548", "spider_web"], "816": ["n04277352", "spindle"], "817": ["n04285008", "sports_car"], "818": ["n04286575", "spotlight"], "819": ["n04296562", "stage"], "820": ["n04310018", "steam_locomotive"], "821": ["n04311004", "steel_arch_bridge"], "822": ["n04311174", "steel_drum"], "823": ["n04317175", "stethoscope"], "824": ["n04325704", "stole"], "825": ["n04326547", "stone_wall"], "826": ["n04328186", "stopwatch"], "827": ["n04330267", "stove"], "828": ["n04332243", "strainer"], "829": ["n04335435", "streetcar"], "830": ["n04336792", "stretcher"], "831": ["n04344873", "studio_couch"], "832": ["n04346328", "stupa"], "833": ["n04347754", "submarine"], "834": ["n04350905", "suit"], "835": ["n04355338", "sundial"], "836": ["n04355933", "sunglass"], "837": ["n04356056", "sunglasses"], "838": ["n04357314", "sunscreen"], "839": ["n04366367", "suspension_bridge"], "840": ["n04367480", "swab"], "841": ["n04370456", "sweatshirt"], "842": ["n04371430", "swimming_trunks"], "843": ["n04371774", "swing"], "844": ["n04372370", "switch"], "845": ["n04376876", "syringe"], "846": ["n04380533", "table_lamp"], "847": ["n04389033", "tank"], "848": ["n04392985", "tape_player"], "849": ["n04398044", "teapot"], "850": ["n04399382", "teddy"], "851": ["n04404412", "television"], "852": ["n04409515", "tennis_ball"], "853": ["n04417672", "thatch"], "854": ["n04418357", "theater_curtain"], "855": ["n04423845", "thimble"], "856": ["n04428191", "thresher"], "857": ["n04429376", "throne"], "858": ["n04435653", "tile_roof"], "859": ["n04442312", "toaster"], "860": ["n04443257", "tobacco_shop"], "861": ["n04447861", "toilet_seat"], "862": ["n04456115", "torch"], "863": ["n04458633", "totem_pole"], "864": ["n04461696", "tow_truck"], "865": ["n04462240", "toyshop"], "866": ["n04465501", "tractor"], "867": ["n04467665", "trailer_truck"], "868": ["n04476259", "tray"], "869": ["n04479046", "trench_coat"], "870": ["n04482393", "tricycle"], "871": ["n04483307", "trimaran"], "872": ["n04485082", "tripod"], "873": ["n04486054", "triumphal_arch"], "874": ["n04487081", "trolleybus"], "875": ["n04487394", "trombone"], "876": ["n04493381", "tub"], "877": ["n04501370", "turnstile"], "878": ["n04505470", "typewriter_keyboard"], "879": ["n04507155", "umbrella"], "880": ["n04509417", "unicycle"], "881": ["n04515003", "upright"], "882": ["n04517823", "vacuum"], "883": ["n04522168", "vase"], "884": ["n04523525", "vault"], "885": ["n04525038", "velvet"], "886": ["n04525305", "vending_machine"], "887": ["n04532106", "vestment"], "888": ["n04532670", "viaduct"], "889": ["n04536866", "violin"], "890": ["n04540053", "volleyball"], "891": ["n04542943", "waffle_iron"], "892": ["n04548280", "wall_clock"], "893": ["n04548362", "wallet"], "894": ["n04550184", "wardrobe"], "895": ["n04552348", "warplane"], "896": ["n04553703", "washbasin"], "897": ["n04554684", "washer"], "898": ["n04557648", "water_bottle"], "899": ["n04560804", "water_jug"], "900": ["n04562935", "water_tower"], "901": ["n04579145", "whiskey_jug"], "902": ["n04579432", "whistle"], "903": ["n04584207", "wig"], "904": ["n04589890", "window_screen"], "905": ["n04590129", "window_shade"], "906": ["n04591157", "Windsor_tie"], "907": ["n04591713", "wine_bottle"], "908": ["n04592741", "wing"], "909": ["n04596742", "wok"], "910": ["n04597913", "wooden_spoon"], "911": ["n04599235", "wool"], "912": ["n04604644", "worm_fence"], "913": ["n04606251", "wreck"], "914": ["n04612504", "yawl"], "915": ["n04613696", "yurt"], "916": ["n06359193", "web_site"], "917": ["n06596364", "comic_book"], "918": ["n06785654", "crossword_puzzle"], "919": ["n06794110", "street_sign"], "920": ["n06874185", "traffic_light"], "921": ["n07248320", "book_jacket"], "922": ["n07565083", "menu"], "923": ["n07579787", "plate"], "924": ["n07583066", "guacamole"], "925": ["n07584110", "consomme"], "926": ["n07590611", "hot_pot"], "927": ["n07613480", "trifle"], "928": ["n07614500", "ice_cream"], "929": ["n07615774", "ice_lolly"], "930": ["n07684084", "French_loaf"], "931": ["n07693725", "bagel"], "932": ["n07695742", "pretzel"], "933": ["n07697313", "cheeseburger"], "934": ["n07697537", "hotdog"], "935": ["n07711569", "mashed_potato"], "936": ["n07714571", "head_cabbage"], "937": ["n07714990", "broccoli"], "938": ["n07715103", "cauliflower"], "939": ["n07716358", "zucchini"], "940": ["n07716906", "spaghetti_squash"], "941": ["n07717410", "acorn_squash"], "942": ["n07717556", "butternut_squash"], "943": ["n07718472", "cucumber"], "944": ["n07718747", "artichoke"], "945": ["n07720875", "bell_pepper"], "946": ["n07730033", "cardoon"], "947": ["n07734744", "mushroom"], "948": ["n07742313", "Granny_Smith"], "949": ["n07745940", "strawberry"], "950": ["n07747607", "orange"], "951": ["n07749582", "lemon"], "952": ["n07753113", "fig"], "953": ["n07753275", "pineapple"], "954": ["n07753592", "banana"], "955": ["n07754684", "jackfruit"], "956": ["n07760859", "custard_apple"], "957": ["n07768694", "pomegranate"], "958": ["n07802026", "hay"], "959": ["n07831146", "carbonara"], "960": ["n07836838", "chocolate_sauce"], "961": ["n07860988", "dough"], "962": ["n07871810", "meat_loaf"], "963": ["n07873807", "pizza"], "964": ["n07875152", "potpie"], "965": ["n07880968", "burrito"], "966": ["n07892512", "red_wine"], "967": ["n07920052", "espresso"], "968": ["n07930864", "cup"], "969": ["n07932039", "eggnog"], "970": ["n09193705", "alp"], "971": ["n09229709", "bubble"], "972": ["n09246464", "cliff"], "973": ["n09256479", "coral_reef"], "974": ["n09288635", "geyser"], "975": ["n09332890", "lakeside"], "976": ["n09399592", "promontory"], "977": ["n09421951", "sandbar"], "978": ["n09428293", "seashore"], "979": ["n09468604", "valley"], "980": ["n09472597", "volcano"], "981": ["n09835506", "ballplayer"], "982": ["n10148035", "groom"], "983": ["n10565667", "scuba_diver"], "984": ["n11879895", "rapeseed"], "985": ["n11939491", "daisy"], "986": ["n12057211", "yellow_lady's_slipper"], "987": ["n12144580", "corn"], "988": ["n12267677", "acorn"], "989": ["n12620546", "hip"], "990": ["n12768682", "buckeye"], "991": ["n12985857", "coral_fungus"], "992": ["n12998815", "agaric"], "993": ["n13037406", "gyromitra"], "994": ["n13040303", "stinkhorn"], "995": ["n13044778", "earthstar"], "996": ["n13052670", "hen-of-the-woods"], "997": ["n13054560", "bolete"], "998": ["n13133613", "ear"], "999": ["n15075141", "toilet_tissue"]}
\ No newline at end of file
{"0": ["n01440764", "tench"], "1": ["n01443537", "goldfish"], "2": ["n01484850", "great_white_shark"], "3": ["n01491361", "tiger_shark"], "4": ["n01494475", "hammerhead"], "5": ["n01496331", "electric_ray"], "6": ["n01498041", "stingray"], "7": ["n01514668", "cock"], "8": ["n01514859", "hen"], "9": ["n01518878", "ostrich"], "10": ["n01530575", "brambling"], "11": ["n01531178", "goldfinch"], "12": ["n01532829", "house_finch"], "13": ["n01534433", "junco"], "14": ["n01537544", "indigo_bunting"], "15": ["n01558993", "robin"], "16": ["n01560419", "bulbul"], "17": ["n01580077", "jay"], "18": ["n01582220", "magpie"], "19": ["n01592084", "chickadee"], "20": ["n01601694", "water_ouzel"], "21": ["n01608432", "kite"], "22": ["n01614925", "bald_eagle"], "23": ["n01616318", "vulture"], "24": ["n01622779", "great_grey_owl"], "25": ["n01629819", "European_fire_salamander"], "26": ["n01630670", "common_newt"], "27": ["n01631663", "eft"], "28": ["n01632458", "spotted_salamander"], "29": ["n01632777", "axolotl"], "30": ["n01641577", "bullfrog"], "31": ["n01644373", "tree_frog"], "32": ["n01644900", "tailed_frog"], "33": ["n01664065", "loggerhead"], "34": ["n01665541", "leatherback_turtle"], "35": ["n01667114", "mud_turtle"], "36": ["n01667778", "terrapin"], "37": ["n01669191", "box_turtle"], "38": ["n01675722", "banded_gecko"], "39": ["n01677366", "common_iguana"], "40": ["n01682714", "American_chameleon"], "41": ["n01685808", "whiptail"], "42": ["n01687978", "agama"], "43": ["n01688243", "frilled_lizard"], "44": ["n01689811", "alligator_lizard"], "45": ["n01692333", "Gila_monster"], "46": ["n01693334", "green_lizard"], "47": ["n01694178", "African_chameleon"], "48": ["n01695060", "Komodo_dragon"], "49": ["n01697457", "African_crocodile"], "50": ["n01698640", "American_alligator"], "51": ["n01704323", "triceratops"], "52": ["n01728572", "thunder_snake"], "53": ["n01728920", "ringneck_snake"], "54": ["n01729322", "hognose_snake"], "55": ["n01729977", "green_snake"], "56": ["n01734418", "king_snake"], "57": ["n01735189", "garter_snake"], "58": ["n01737021", "water_snake"], "59": ["n01739381", "vine_snake"], "60": ["n01740131", "night_snake"], "61": ["n01742172", "boa_constrictor"], "62": ["n01744401", "rock_python"], "63": ["n01748264", "Indian_cobra"], "64": ["n01749939", "green_mamba"], "65": ["n01751748", "sea_snake"], "66": ["n01753488", "horned_viper"], "67": ["n01755581", "diamondback"], "68": ["n01756291", "sidewinder"], "69": ["n01768244", "trilobite"], "70": ["n01770081", "harvestman"], "71": ["n01770393", "scorpion"], "72": ["n01773157", "black_and_gold_garden_spider"], "73": ["n01773549", "barn_spider"], "74": ["n01773797", "garden_spider"], "75": ["n01774384", "black_widow"], "76": ["n01774750", "tarantula"], "77": ["n01775062", "wolf_spider"], "78": ["n01776313", "tick"], "79": ["n01784675", "centipede"], "80": ["n01795545", "black_grouse"], "81": ["n01796340", "ptarmigan"], "82": ["n01797886", "ruffed_grouse"], "83": ["n01798484", "prairie_chicken"], "84": ["n01806143", "peacock"], "85": ["n01806567", "quail"], "86": ["n01807496", "partridge"], "87": ["n01817953", "African_grey"], "88": ["n01818515", "macaw"], "89": ["n01819313", "sulphur-crested_cockatoo"], "90": ["n01820546", "lorikeet"], "91": ["n01824575", "coucal"], "92": ["n01828970", "bee_eater"], "93": ["n01829413", "hornbill"], "94": ["n01833805", "hummingbird"], "95": ["n01843065", "jacamar"], "96": ["n01843383", "toucan"], "97": ["n01847000", "drake"], "98": ["n01855032", "red-breasted_merganser"], "99": ["n01855672", "goose"], "100": ["n01860187", "black_swan"], "101": ["n01871265", "tusker"], "102": ["n01872401", "echidna"], "103": ["n01873310", "platypus"], "104": ["n01877812", "wallaby"], "105": ["n01882714", "koala"], "106": ["n01883070", "wombat"], "107": ["n01910747", "jellyfish"], "108": ["n01914609", "sea_anemone"], "109": ["n01917289", "brain_coral"], "110": ["n01924916", "flatworm"], "111": ["n01930112", "nematode"], "112": ["n01943899", "conch"], "113": ["n01944390", "snail"], "114": ["n01945685", "slug"], "115": ["n01950731", "sea_slug"], "116": ["n01955084", "chiton"], "117": ["n01968897", "chambered_nautilus"], "118": ["n01978287", "Dungeness_crab"], "119": ["n01978455", "rock_crab"], "120": ["n01980166", "fiddler_crab"], "121": ["n01981276", "king_crab"], "122": ["n01983481", "American_lobster"], "123": ["n01984695", "spiny_lobster"], "124": ["n01985128", "crayfish"], "125": ["n01986214", "hermit_crab"], "126": ["n01990800", "isopod"], "127": ["n02002556", "white_stork"], "128": ["n02002724", "black_stork"], "129": ["n02006656", "spoonbill"], "130": ["n02007558", "flamingo"], "131": ["n02009229", "little_blue_heron"], "132": ["n02009912", "American_egret"], "133": ["n02011460", "bittern"], "134": ["n02012849", "crane"], "135": ["n02013706", "limpkin"], "136": ["n02017213", "European_gallinule"], "137": ["n02018207", "American_coot"], "138": ["n02018795", "bustard"], "139": ["n02025239", "ruddy_turnstone"], "140": ["n02027492", "red-backed_sandpiper"], "141": ["n02028035", "redshank"], "142": ["n02033041", "dowitcher"], "143": ["n02037110", "oystercatcher"], "144": ["n02051845", "pelican"], "145": ["n02056570", "king_penguin"], "146": ["n02058221", "albatross"], "147": ["n02066245", "grey_whale"], "148": ["n02071294", "killer_whale"], "149": ["n02074367", "dugong"], "150": ["n02077923", "sea_lion"], "151": ["n02085620", "Chihuahua"], "152": ["n02085782", "Japanese_spaniel"], "153": ["n02085936", "Maltese_dog"], "154": ["n02086079", "Pekinese"], "155": ["n02086240", "Shih-Tzu"], "156": ["n02086646", "Blenheim_spaniel"], "157": ["n02086910", "papillon"], "158": ["n02087046", "toy_terrier"], "159": ["n02087394", "Rhodesian_ridgeback"], "160": ["n02088094", "Afghan_hound"], "161": ["n02088238", "basset"], "162": ["n02088364", "beagle"], "163": ["n02088466", "bloodhound"], "164": ["n02088632", "bluetick"], "165": ["n02089078", "black-and-tan_coonhound"], "166": ["n02089867", "Walker_hound"], "167": ["n02089973", "English_foxhound"], "168": ["n02090379", "redbone"], "169": ["n02090622", "borzoi"], "170": ["n02090721", "Irish_wolfhound"], "171": ["n02091032", "Italian_greyhound"], "172": ["n02091134", "whippet"], "173": ["n02091244", "Ibizan_hound"], "174": ["n02091467", "Norwegian_elkhound"], "175": ["n02091635", "otterhound"], "176": ["n02091831", "Saluki"], "177": ["n02092002", "Scottish_deerhound"], "178": ["n02092339", "Weimaraner"], "179": ["n02093256", "Staffordshire_bullterrier"], "180": ["n02093428", "American_Staffordshire_terrier"], "181": ["n02093647", "Bedlington_terrier"], "182": ["n02093754", "Border_terrier"], "183": ["n02093859", "Kerry_blue_terrier"], "184": ["n02093991", "Irish_terrier"], "185": ["n02094114", "Norfolk_terrier"], "186": ["n02094258", "Norwich_terrier"], "187": ["n02094433", "Yorkshire_terrier"], "188": ["n02095314", "wire-haired_fox_terrier"], "189": ["n02095570", "Lakeland_terrier"], "190": ["n02095889", "Sealyham_terrier"], "191": ["n02096051", "Airedale"], "192": ["n02096177", "cairn"], "193": ["n02096294", "Australian_terrier"], "194": ["n02096437", "Dandie_Dinmont"], "195": ["n02096585", "Boston_bull"], "196": ["n02097047", "miniature_schnauzer"], "197": ["n02097130", "giant_schnauzer"], "198": ["n02097209", "standard_schnauzer"], "199": ["n02097298", "Scotch_terrier"], "200": ["n02097474", "Tibetan_terrier"], "201": ["n02097658", "silky_terrier"], "202": ["n02098105", "soft-coated_wheaten_terrier"], "203": ["n02098286", "West_Highland_white_terrier"], "204": ["n02098413", "Lhasa"], "205": ["n02099267", "flat-coated_retriever"], "206": ["n02099429", "curly-coated_retriever"], "207": ["n02099601", "golden_retriever"], "208": ["n02099712", "Labrador_retriever"], "209": ["n02099849", "Chesapeake_Bay_retriever"], "210": ["n02100236", "German_short-haired_pointer"], "211": ["n02100583", "vizsla"], "212": ["n02100735", "English_setter"], "213": ["n02100877", "Irish_setter"], "214": ["n02101006", "Gordon_setter"], "215": ["n02101388", "Brittany_spaniel"], "216": ["n02101556", "clumber"], "217": ["n02102040", "English_springer"], "218": ["n02102177", "Welsh_springer_spaniel"], "219": ["n02102318", "cocker_spaniel"], "220": ["n02102480", "Sussex_spaniel"], "221": ["n02102973", "Irish_water_spaniel"], "222": ["n02104029", "kuvasz"], "223": ["n02104365", "schipperke"], "224": ["n02105056", "groenendael"], "225": ["n02105162", "malinois"], "226": ["n02105251", "briard"], "227": ["n02105412", "kelpie"], "228": ["n02105505", "komondor"], "229": ["n02105641", "Old_English_sheepdog"], "230": ["n02105855", "Shetland_sheepdog"], "231": ["n02106030", "collie"], "232": ["n02106166", "Border_collie"], "233": ["n02106382", "Bouvier_des_Flandres"], "234": ["n02106550", "Rottweiler"], "235": ["n02106662", "German_shepherd"], "236": ["n02107142", "Doberman"], "237": ["n02107312", "miniature_pinscher"], "238": ["n02107574", "Greater_Swiss_Mountain_dog"], "239": ["n02107683", "Bernese_mountain_dog"], "240": ["n02107908", "Appenzeller"], "241": ["n02108000", "EntleBucher"], "242": ["n02108089", "boxer"], "243": ["n02108422", "bull_mastiff"], "244": ["n02108551", "Tibetan_mastiff"], "245": ["n02108915", "French_bulldog"], "246": ["n02109047", "Great_Dane"], "247": ["n02109525", "Saint_Bernard"], "248": ["n02109961", "Eskimo_dog"], "249": ["n02110063", "malamute"], "250": ["n02110185", "Siberian_husky"], "251": ["n02110341", "dalmatian"], "252": ["n02110627", "affenpinscher"], "253": ["n02110806", "basenji"], "254": ["n02110958", "pug"], "255": ["n02111129", "Leonberg"], "256": ["n02111277", "Newfoundland"], "257": ["n02111500", "Great_Pyrenees"], "258": ["n02111889", "Samoyed"], "259": ["n02112018", "Pomeranian"], "260": ["n02112137", "chow"], "261": ["n02112350", "keeshond"], "262": ["n02112706", "Brabancon_griffon"], "263": ["n02113023", "Pembroke"], "264": ["n02113186", "Cardigan"], "265": ["n02113624", "toy_poodle"], "266": ["n02113712", "miniature_poodle"], "267": ["n02113799", "standard_poodle"], "268": ["n02113978", "Mexican_hairless"], "269": ["n02114367", "timber_wolf"], "270": ["n02114548", "white_wolf"], "271": ["n02114712", "red_wolf"], "272": ["n02114855", "coyote"], "273": ["n02115641", "dingo"], "274": ["n02115913", "dhole"], "275": ["n02116738", "African_hunting_dog"], "276": ["n02117135", "hyena"], "277": ["n02119022", "red_fox"], "278": ["n02119789", "kit_fox"], "279": ["n02120079", "Arctic_fox"], "280": ["n02120505", "grey_fox"], "281": ["n02123045", "tabby"], "282": ["n02123159", "tiger_cat"], "283": ["n02123394", "Persian_cat"], "284": ["n02123597", "Siamese_cat"], "285": ["n02124075", "Egyptian_cat"], "286": ["n02125311", "cougar"], "287": ["n02127052", "lynx"], "288": ["n02128385", "leopard"], "289": ["n02128757", "snow_leopard"], "290": ["n02128925", "jaguar"], "291": ["n02129165", "lion"], "292": ["n02129604", "tiger"], "293": ["n02130308", "cheetah"], "294": ["n02132136", "brown_bear"], "295": ["n02133161", "American_black_bear"], "296": ["n02134084", "ice_bear"], "297": ["n02134418", "sloth_bear"], "298": ["n02137549", "mongoose"], "299": ["n02138441", "meerkat"], "300": ["n02165105", "tiger_beetle"], "301": ["n02165456", "ladybug"], "302": ["n02167151", "ground_beetle"], "303": ["n02168699", "long-horned_beetle"], "304": ["n02169497", "leaf_beetle"], "305": ["n02172182", "dung_beetle"], "306": ["n02174001", "rhinoceros_beetle"], "307": ["n02177972", "weevil"], "308": ["n02190166", "fly"], "309": ["n02206856", "bee"], "310": ["n02219486", "ant"], "311": ["n02226429", "grasshopper"], "312": ["n02229544", "cricket"], "313": ["n02231487", "walking_stick"], "314": ["n02233338", "cockroach"], "315": ["n02236044", "mantis"], "316": ["n02256656", "cicada"], "317": ["n02259212", "leafhopper"], "318": ["n02264363", "lacewing"], "319": ["n02268443", "dragonfly"], "320": ["n02268853", "damselfly"], "321": ["n02276258", "admiral"], "322": ["n02277742", "ringlet"], "323": ["n02279972", "monarch"], "324": ["n02280649", "cabbage_butterfly"], "325": ["n02281406", "sulphur_butterfly"], "326": ["n02281787", "lycaenid"], "327": ["n02317335", "starfish"], "328": ["n02319095", "sea_urchin"], "329": ["n02321529", "sea_cucumber"], "330": ["n02325366", "wood_rabbit"], "331": ["n02326432", "hare"], "332": ["n02328150", "Angora"], "333": ["n02342885", "hamster"], "334": ["n02346627", "porcupine"], "335": ["n02356798", "fox_squirrel"], "336": ["n02361337", "marmot"], "337": ["n02363005", "beaver"], "338": ["n02364673", "guinea_pig"], "339": ["n02389026", "sorrel"], "340": ["n02391049", "zebra"], "341": ["n02395406", "hog"], "342": ["n02396427", "wild_boar"], "343": ["n02397096", "warthog"], "344": ["n02398521", "hippopotamus"], "345": ["n02403003", "ox"], "346": ["n02408429", "water_buffalo"], "347": ["n02410509", "bison"], "348": ["n02412080", "ram"], "349": ["n02415577", "bighorn"], "350": ["n02417914", "ibex"], "351": ["n02422106", "hartebeest"], "352": ["n02422699", "impala"], "353": ["n02423022", "gazelle"], "354": ["n02437312", "Arabian_camel"], "355": ["n02437616", "llama"], "356": ["n02441942", "weasel"], "357": ["n02442845", "mink"], "358": ["n02443114", "polecat"], "359": ["n02443484", "black-footed_ferret"], "360": ["n02444819", "otter"], "361": ["n02445715", "skunk"], "362": ["n02447366", "badger"], "363": ["n02454379", "armadillo"], "364": ["n02457408", "three-toed_sloth"], "365": ["n02480495", "orangutan"], "366": ["n02480855", "gorilla"], "367": ["n02481823", "chimpanzee"], "368": ["n02483362", "gibbon"], "369": ["n02483708", "siamang"], "370": ["n02484975", "guenon"], "371": ["n02486261", "patas"], "372": ["n02486410", "baboon"], "373": ["n02487347", "macaque"], "374": ["n02488291", "langur"], "375": ["n02488702", "colobus"], "376": ["n02489166", "proboscis_monkey"], "377": ["n02490219", "marmoset"], "378": ["n02492035", "capuchin"], "379": ["n02492660", "howler_monkey"], "380": ["n02493509", "titi"], "381": ["n02493793", "spider_monkey"], "382": ["n02494079", "squirrel_monkey"], "383": ["n02497673", "Madagascar_cat"], "384": ["n02500267", "indri"], "385": ["n02504013", "Indian_elephant"], "386": ["n02504458", "African_elephant"], "387": ["n02509815", "lesser_panda"], "388": ["n02510455", "giant_panda"], "389": ["n02514041", "barracouta"], "390": ["n02526121", "eel"], "391": ["n02536864", "coho"], "392": ["n02606052", "rock_beauty"], "393": ["n02607072", "anemone_fish"], "394": ["n02640242", "sturgeon"], "395": ["n02641379", "gar"], "396": ["n02643566", "lionfish"], "397": ["n02655020", "puffer"], "398": ["n02666196", "abacus"], "399": ["n02667093", "abaya"], "400": ["n02669723", "academic_gown"], "401": ["n02672831", "accordion"], "402": ["n02676566", "acoustic_guitar"], "403": ["n02687172", "aircraft_carrier"], "404": ["n02690373", "airliner"], "405": ["n02692877", "airship"], "406": ["n02699494", "altar"], "407": ["n02701002", "ambulance"], "408": ["n02704792", "amphibian"], "409": ["n02708093", "analog_clock"], "410": ["n02727426", "apiary"], "411": ["n02730930", "apron"], "412": ["n02747177", "ashcan"], "413": ["n02749479", "assault_rifle"], "414": ["n02769748", "backpack"], "415": ["n02776631", "bakery"], "416": ["n02777292", "balance_beam"], "417": ["n02782093", "balloon"], "418": ["n02783161", "ballpoint"], "419": ["n02786058", "Band_Aid"], "420": ["n02787622", "banjo"], "421": ["n02788148", "bannister"], "422": ["n02790996", "barbell"], "423": ["n02791124", "barber_chair"], "424": ["n02791270", "barbershop"], "425": ["n02793495", "barn"], "426": ["n02794156", "barometer"], "427": ["n02795169", "barrel"], "428": ["n02797295", "barrow"], "429": ["n02799071", "baseball"], "430": ["n02802426", "basketball"], "431": ["n02804414", "bassinet"], "432": ["n02804610", "bassoon"], "433": ["n02807133", "bathing_cap"], "434": ["n02808304", "bath_towel"], "435": ["n02808440", "bathtub"], "436": ["n02814533", "beach_wagon"], "437": ["n02814860", "beacon"], "438": ["n02815834", "beaker"], "439": ["n02817516", "bearskin"], "440": ["n02823428", "beer_bottle"], "441": ["n02823750", "beer_glass"], "442": ["n02825657", "bell_cote"], "443": ["n02834397", "bib"], "444": ["n02835271", "bicycle-built-for-two"], "445": ["n02837789", "bikini"], "446": ["n02840245", "binder"], "447": ["n02841315", "binoculars"], "448": ["n02843684", "birdhouse"], "449": ["n02859443", "boathouse"], "450": ["n02860847", "bobsled"], "451": ["n02865351", "bolo_tie"], "452": ["n02869837", "bonnet"], "453": ["n02870880", "bookcase"], "454": ["n02871525", "bookshop"], "455": ["n02877765", "bottlecap"], "456": ["n02879718", "bow"], "457": ["n02883205", "bow_tie"], "458": ["n02892201", "brass"], "459": ["n02892767", "brassiere"], "460": ["n02894605", "breakwater"], "461": ["n02895154", "breastplate"], "462": ["n02906734", "broom"], "463": ["n02909870", "bucket"], "464": ["n02910353", "buckle"], "465": ["n02916936", "bulletproof_vest"], "466": ["n02917067", "bullet_train"], "467": ["n02927161", "butcher_shop"], "468": ["n02930766", "cab"], "469": ["n02939185", "caldron"], "470": ["n02948072", "candle"], "471": ["n02950826", "cannon"], "472": ["n02951358", "canoe"], "473": ["n02951585", "can_opener"], "474": ["n02963159", "cardigan"], "475": ["n02965783", "car_mirror"], "476": ["n02966193", "carousel"], "477": ["n02966687", "carpenter's_kit"], "478": ["n02971356", "carton"], "479": ["n02974003", "car_wheel"], "480": ["n02977058", "cash_machine"], "481": ["n02978881", "cassette"], "482": ["n02979186", "cassette_player"], "483": ["n02980441", "castle"], "484": ["n02981792", "catamaran"], "485": ["n02988304", "CD_player"], "486": ["n02992211", "cello"], "487": ["n02992529", "cellular_telephone"], "488": ["n02999410", "chain"], "489": ["n03000134", "chainlink_fence"], "490": ["n03000247", "chain_mail"], "491": ["n03000684", "chain_saw"], "492": ["n03014705", "chest"], "493": ["n03016953", "chiffonier"], "494": ["n03017168", "chime"], "495": ["n03018349", "china_cabinet"], "496": ["n03026506", "Christmas_stocking"], "497": ["n03028079", "church"], "498": ["n03032252", "cinema"], "499": ["n03041632", "cleaver"], "500": ["n03042490", "cliff_dwelling"], "501": ["n03045698", "cloak"], "502": ["n03047690", "clog"], "503": ["n03062245", "cocktail_shaker"], "504": ["n03063599", "coffee_mug"], "505": ["n03063689", "coffeepot"], "506": ["n03065424", "coil"], "507": ["n03075370", "combination_lock"], "508": ["n03085013", "computer_keyboard"], "509": ["n03089624", "confectionery"], "510": ["n03095699", "container_ship"], "511": ["n03100240", "convertible"], "512": ["n03109150", "corkscrew"], "513": ["n03110669", "cornet"], "514": ["n03124043", "cowboy_boot"], "515": ["n03124170", "cowboy_hat"], "516": ["n03125729", "cradle"], "517": ["n03126707", "crane"], "518": ["n03127747", "crash_helmet"], "519": ["n03127925", "crate"], "520": ["n03131574", "crib"], "521": ["n03133878", "Crock_Pot"], "522": ["n03134739", "croquet_ball"], "523": ["n03141823", "crutch"], "524": ["n03146219", "cuirass"], "525": ["n03160309", "dam"], "526": ["n03179701", "desk"], "527": ["n03180011", "desktop_computer"], "528": ["n03187595", "dial_telephone"], "529": ["n03188531", "diaper"], "530": ["n03196217", "digital_clock"], "531": ["n03197337", "digital_watch"], "532": ["n03201208", "dining_table"], "533": ["n03207743", "dishrag"], "534": ["n03207941", "dishwasher"], "535": ["n03208938", "disk_brake"], "536": ["n03216828", "dock"], "537": ["n03218198", "dogsled"], "538": ["n03220513", "dome"], "539": ["n03223299", "doormat"], "540": ["n03240683", "drilling_platform"], "541": ["n03249569", "drum"], "542": ["n03250847", "drumstick"], "543": ["n03255030", "dumbbell"], "544": ["n03259280", "Dutch_oven"], "545": ["n03271574", "electric_fan"], "546": ["n03272010", "electric_guitar"], "547": ["n03272562", "electric_locomotive"], "548": ["n03290653", "entertainment_center"], "549": ["n03291819", "envelope"], "550": ["n03297495", "espresso_maker"], "551": ["n03314780", "face_powder"], "552": ["n03325584", "feather_boa"], "553": ["n03337140", "file"], "554": ["n03344393", "fireboat"], "555": ["n03345487", "fire_engine"], "556": ["n03347037", "fire_screen"], "557": ["n03355925", "flagpole"], "558": ["n03372029", "flute"], "559": ["n03376595", "folding_chair"], "560": ["n03379051", "football_helmet"], "561": ["n03384352", "forklift"], "562": ["n03388043", "fountain"], "563": ["n03388183", "fountain_pen"], "564": ["n03388549", "four-poster"], "565": ["n03393912", "freight_car"], "566": ["n03394916", "French_horn"], "567": ["n03400231", "frying_pan"], "568": ["n03404251", "fur_coat"], "569": ["n03417042", "garbage_truck"], "570": ["n03424325", "gasmask"], "571": ["n03425413", "gas_pump"], "572": ["n03443371", "goblet"], "573": ["n03444034", "go-kart"], "574": ["n03445777", "golf_ball"], "575": ["n03445924", "golfcart"], "576": ["n03447447", "gondola"], "577": ["n03447721", "gong"], "578": ["n03450230", "gown"], "579": ["n03452741", "grand_piano"], "580": ["n03457902", "greenhouse"], "581": ["n03459775", "grille"], "582": ["n03461385", "grocery_store"], "583": ["n03467068", "guillotine"], "584": ["n03476684", "hair_slide"], "585": ["n03476991", "hair_spray"], "586": ["n03478589", "half_track"], "587": ["n03481172", "hammer"], "588": ["n03482405", "hamper"], "589": ["n03483316", "hand_blower"], "590": ["n03485407", "hand-held_computer"], "591": ["n03485794", "handkerchief"], "592": ["n03492542", "hard_disc"], "593": ["n03494278", "harmonica"], "594": ["n03495258", "harp"], "595": ["n03496892", "harvester"], "596": ["n03498962", "hatchet"], "597": ["n03527444", "holster"], "598": ["n03529860", "home_theater"], "599": ["n03530642", "honeycomb"], "600": ["n03532672", "hook"], "601": ["n03534580", "hoopskirt"], "602": ["n03535780", "horizontal_bar"], "603": ["n03538406", "horse_cart"], "604": ["n03544143", "hourglass"], "605": ["n03584254", "iPod"], "606": ["n03584829", "iron"], "607": ["n03590841", "jack-o'-lantern"], "608": ["n03594734", "jean"], "609": ["n03594945", "jeep"], "610": ["n03595614", "jersey"], "611": ["n03598930", "jigsaw_puzzle"], "612": ["n03599486", "jinrikisha"], "613": ["n03602883", "joystick"], "614": ["n03617480", "kimono"], "615": ["n03623198", "knee_pad"], "616": ["n03627232", "knot"], "617": ["n03630383", "lab_coat"], "618": ["n03633091", "ladle"], "619": ["n03637318", "lampshade"], "620": ["n03642806", "laptop"], "621": ["n03649909", "lawn_mower"], "622": ["n03657121", "lens_cap"], "623": ["n03658185", "letter_opener"], "624": ["n03661043", "library"], "625": ["n03662601", "lifeboat"], "626": ["n03666591", "lighter"], "627": ["n03670208", "limousine"], "628": ["n03673027", "liner"], "629": ["n03676483", "lipstick"], "630": ["n03680355", "Loafer"], "631": ["n03690938", "lotion"], "632": ["n03691459", "loudspeaker"], "633": ["n03692522", "loupe"], "634": ["n03697007", "lumbermill"], "635": ["n03706229", "magnetic_compass"], "636": ["n03709823", "mailbag"], "637": ["n03710193", "mailbox"], "638": ["n03710637", "maillot"], "639": ["n03710721", "maillot"], "640": ["n03717622", "manhole_cover"], "641": ["n03720891", "maraca"], "642": ["n03721384", "marimba"], "643": ["n03724870", "mask"], "644": ["n03729826", "matchstick"], "645": ["n03733131", "maypole"], "646": ["n03733281", "maze"], "647": ["n03733805", "measuring_cup"], "648": ["n03742115", "medicine_chest"], "649": ["n03743016", "megalith"], "650": ["n03759954", "microphone"], "651": ["n03761084", "microwave"], "652": ["n03763968", "military_uniform"], "653": ["n03764736", "milk_can"], "654": ["n03769881", "minibus"], "655": ["n03770439", "miniskirt"], "656": ["n03770679", "minivan"], "657": ["n03773504", "missile"], "658": ["n03775071", "mitten"], "659": ["n03775546", "mixing_bowl"], "660": ["n03776460", "mobile_home"], "661": ["n03777568", "Model_T"], "662": ["n03777754", "modem"], "663": ["n03781244", "monastery"], "664": ["n03782006", "monitor"], "665": ["n03785016", "moped"], "666": ["n03786901", "mortar"], "667": ["n03787032", "mortarboard"], "668": ["n03788195", "mosque"], "669": ["n03788365", "mosquito_net"], "670": ["n03791053", "motor_scooter"], "671": ["n03792782", "mountain_bike"], "672": ["n03792972", "mountain_tent"], "673": ["n03793489", "mouse"], "674": ["n03794056", "mousetrap"], "675": ["n03796401", "moving_van"], "676": ["n03803284", "muzzle"], "677": ["n03804744", "nail"], "678": ["n03814639", "neck_brace"], "679": ["n03814906", "necklace"], "680": ["n03825788", "nipple"], "681": ["n03832673", "notebook"], "682": ["n03837869", "obelisk"], "683": ["n03838899", "oboe"], "684": ["n03840681", "ocarina"], "685": ["n03841143", "odometer"], "686": ["n03843555", "oil_filter"], "687": ["n03854065", "organ"], "688": ["n03857828", "oscilloscope"], "689": ["n03866082", "overskirt"], "690": ["n03868242", "oxcart"], "691": ["n03868863", "oxygen_mask"], "692": ["n03871628", "packet"], "693": ["n03873416", "paddle"], "694": ["n03874293", "paddlewheel"], "695": ["n03874599", "padlock"], "696": ["n03876231", "paintbrush"], "697": ["n03877472", "pajama"], "698": ["n03877845", "palace"], "699": ["n03884397", "panpipe"], "700": ["n03887697", "paper_towel"], "701": ["n03888257", "parachute"], "702": ["n03888605", "parallel_bars"], "703": ["n03891251", "park_bench"], "704": ["n03891332", "parking_meter"], "705": ["n03895866", "passenger_car"], "706": ["n03899768", "patio"], "707": ["n03902125", "pay-phone"], "708": ["n03903868", "pedestal"], "709": ["n03908618", "pencil_box"], "710": ["n03908714", "pencil_sharpener"], "711": ["n03916031", "perfume"], "712": ["n03920288", "Petri_dish"], "713": ["n03924679", "photocopier"], "714": ["n03929660", "pick"], "715": ["n03929855", "pickelhaube"], "716": ["n03930313", "picket_fence"], "717": ["n03930630", "pickup"], "718": ["n03933933", "pier"], "719": ["n03935335", "piggy_bank"], "720": ["n03937543", "pill_bottle"], "721": ["n03938244", "pillow"], "722": ["n03942813", "ping-pong_ball"], "723": ["n03944341", "pinwheel"], "724": ["n03947888", "pirate"], "725": ["n03950228", "pitcher"], "726": ["n03954731", "plane"], "727": ["n03956157", "planetarium"], "728": ["n03958227", "plastic_bag"], "729": ["n03961711", "plate_rack"], "730": ["n03967562", "plow"], "731": ["n03970156", "plunger"], "732": ["n03976467", "Polaroid_camera"], "733": ["n03976657", "pole"], "734": ["n03977966", "police_van"], "735": ["n03980874", "poncho"], "736": ["n03982430", "pool_table"], "737": ["n03983396", "pop_bottle"], "738": ["n03991062", "pot"], "739": ["n03992509", "potter's_wheel"], "740": ["n03995372", "power_drill"], "741": ["n03998194", "prayer_rug"], "742": ["n04004767", "printer"], "743": ["n04005630", "prison"], "744": ["n04008634", "projectile"], "745": ["n04009552", "projector"], "746": ["n04019541", "puck"], "747": ["n04023962", "punching_bag"], "748": ["n04026417", "purse"], "749": ["n04033901", "quill"], "750": ["n04033995", "quilt"], "751": ["n04037443", "racer"], "752": ["n04039381", "racket"], "753": ["n04040759", "radiator"], "754": ["n04041544", "radio"], "755": ["n04044716", "radio_telescope"], "756": ["n04049303", "rain_barrel"], "757": ["n04065272", "recreational_vehicle"], "758": ["n04067472", "reel"], "759": ["n04069434", "reflex_camera"], "760": ["n04070727", "refrigerator"], "761": ["n04074963", "remote_control"], "762": ["n04081281", "restaurant"], "763": ["n04086273", "revolver"], "764": ["n04090263", "rifle"], "765": ["n04099969", "rocking_chair"], "766": ["n04111531", "rotisserie"], "767": ["n04116512", "rubber_eraser"], "768": ["n04118538", "rugby_ball"], "769": ["n04118776", "rule"], "770": ["n04120489", "running_shoe"], "771": ["n04125021", "safe"], "772": ["n04127249", "safety_pin"], "773": ["n04131690", "saltshaker"], "774": ["n04133789", "sandal"], "775": ["n04136333", "sarong"], "776": ["n04141076", "sax"], "777": ["n04141327", "scabbard"], "778": ["n04141975", "scale"], "779": ["n04146614", "school_bus"], "780": ["n04147183", "schooner"], "781": ["n04149813", "scoreboard"], "782": ["n04152593", "screen"], "783": ["n04153751", "screw"], "784": ["n04154565", "screwdriver"], "785": ["n04162706", "seat_belt"], "786": ["n04179913", "sewing_machine"], "787": ["n04192698", "shield"], "788": ["n04200800", "shoe_shop"], "789": ["n04201297", "shoji"], "790": ["n04204238", "shopping_basket"], "791": ["n04204347", "shopping_cart"], "792": ["n04208210", "shovel"], "793": ["n04209133", "shower_cap"], "794": ["n04209239", "shower_curtain"], "795": ["n04228054", "ski"], "796": ["n04229816", "ski_mask"], "797": ["n04235860", "sleeping_bag"], "798": ["n04238763", "slide_rule"], "799": ["n04239074", "sliding_door"], "800": ["n04243546", "slot"], "801": ["n04251144", "snorkel"], "802": ["n04252077", "snowmobile"], "803": ["n04252225", "snowplow"], "804": ["n04254120", "soap_dispenser"], "805": ["n04254680", "soccer_ball"], "806": ["n04254777", "sock"], "807": ["n04258138", "solar_dish"], "808": ["n04259630", "sombrero"], "809": ["n04263257", "soup_bowl"], "810": ["n04264628", "space_bar"], "811": ["n04265275", "space_heater"], "812": ["n04266014", "space_shuttle"], "813": ["n04270147", "spatula"], "814": ["n04273569", "speedboat"], "815": ["n04275548", "spider_web"], "816": ["n04277352", "spindle"], "817": ["n04285008", "sports_car"], "818": ["n04286575", "spotlight"], "819": ["n04296562", "stage"], "820": ["n04310018", "steam_locomotive"], "821": ["n04311004", "steel_arch_bridge"], "822": ["n04311174", "steel_drum"], "823": ["n04317175", "stethoscope"], "824": ["n04325704", "stole"], "825": ["n04326547", "stone_wall"], "826": ["n04328186", "stopwatch"], "827": ["n04330267", "stove"], "828": ["n04332243", "strainer"], "829": ["n04335435", "streetcar"], "830": ["n04336792", "stretcher"], "831": ["n04344873", "studio_couch"], "832": ["n04346328", "stupa"], "833": ["n04347754", "submarine"], "834": ["n04350905", "suit"], "835": ["n04355338", "sundial"], "836": ["n04355933", "sunglass"], "837": ["n04356056", "sunglasses"], "838": ["n04357314", "sunscreen"], "839": ["n04366367", "suspension_bridge"], "840": ["n04367480", "swab"], "841": ["n04370456", "sweatshirt"], "842": ["n04371430", "swimming_trunks"], "843": ["n04371774", "swing"], "844": ["n04372370", "switch"], "845": ["n04376876", "syringe"], "846": ["n04380533", "table_lamp"], "847": ["n04389033", "tank"], "848": ["n04392985", "tape_player"], "849": ["n04398044", "teapot"], "850": ["n04399382", "teddy"], "851": ["n04404412", "television"], "852": ["n04409515", "tennis_ball"], "853": ["n04417672", "thatch"], "854": ["n04418357", "theater_curtain"], "855": ["n04423845", "thimble"], "856": ["n04428191", "thresher"], "857": ["n04429376", "throne"], "858": ["n04435653", "tile_roof"], "859": ["n04442312", "toaster"], "860": ["n04443257", "tobacco_shop"], "861": ["n04447861", "toilet_seat"], "862": ["n04456115", "torch"], "863": ["n04458633", "totem_pole"], "864": ["n04461696", "tow_truck"], "865": ["n04462240", "toyshop"], "866": ["n04465501", "tractor"], "867": ["n04467665", "trailer_truck"], "868": ["n04476259", "tray"], "869": ["n04479046", "trench_coat"], "870": ["n04482393", "tricycle"], "871": ["n04483307", "trimaran"], "872": ["n04485082", "tripod"], "873": ["n04486054", "triumphal_arch"], "874": ["n04487081", "trolleybus"], "875": ["n04487394", "trombone"], "876": ["n04493381", "tub"], "877": ["n04501370", "turnstile"], "878": ["n04505470", "typewriter_keyboard"], "879": ["n04507155", "umbrella"], "880": ["n04509417", "unicycle"], "881": ["n04515003", "upright"], "882": ["n04517823", "vacuum"], "883": ["n04522168", "vase"], "884": ["n04523525", "vault"], "885": ["n04525038", "velvet"], "886": ["n04525305", "vending_machine"], "887": ["n04532106", "vestment"], "888": ["n04532670", "viaduct"], "889": ["n04536866", "violin"], "890": ["n04540053", "volleyball"], "891": ["n04542943", "waffle_iron"], "892": ["n04548280", "wall_clock"], "893": ["n04548362", "wallet"], "894": ["n04550184", "wardrobe"], "895": ["n04552348", "warplane"], "896": ["n04553703", "washbasin"], "897": ["n04554684", "washer"], "898": ["n04557648", "water_bottle"], "899": ["n04560804", "water_jug"], "900": ["n04562935", "water_tower"], "901": ["n04579145", "whiskey_jug"], "902": ["n04579432", "whistle"], "903": ["n04584207", "wig"], "904": ["n04589890", "window_screen"], "905": ["n04590129", "window_shade"], "906": ["n04591157", "Windsor_tie"], "907": ["n04591713", "wine_bottle"], "908": ["n04592741", "wing"], "909": ["n04596742", "wok"], "910": ["n04597913", "wooden_spoon"], "911": ["n04599235", "wool"], "912": ["n04604644", "worm_fence"], "913": ["n04606251", "wreck"], "914": ["n04612504", "yawl"], "915": ["n04613696", "yurt"], "916": ["n06359193", "web_site"], "917": ["n06596364", "comic_book"], "918": ["n06785654", "crossword_puzzle"], "919": ["n06794110", "street_sign"], "920": ["n06874185", "traffic_light"], "921": ["n07248320", "book_jacket"], "922": ["n07565083", "menu"], "923": ["n07579787", "plate"], "924": ["n07583066", "guacamole"], "925": ["n07584110", "consomme"], "926": ["n07590611", "hot_pot"], "927": ["n07613480", "trifle"], "928": ["n07614500", "ice_cream"], "929": ["n07615774", "ice_lolly"], "930": ["n07684084", "French_loaf"], "931": ["n07693725", "bagel"], "932": ["n07695742", "pretzel"], "933": ["n07697313", "cheeseburger"], "934": ["n07697537", "hotdog"], "935": ["n07711569", "mashed_potato"], "936": ["n07714571", "head_cabbage"], "937": ["n07714990", "broccoli"], "938": ["n07715103", "cauliflower"], "939": ["n07716358", "zucchini"], "940": ["n07716906", "spaghetti_squash"], "941": ["n07717410", "acorn_squash"], "942": ["n07717556", "butternut_squash"], "943": ["n07718472", "cucumber"], "944": ["n07718747", "artichoke"], "945": ["n07720875", "bell_pepper"], "946": ["n07730033", "cardoon"], "947": ["n07734744", "mushroom"], "948": ["n07742313", "Granny_Smith"], "949": ["n07745940", "strawberry"], "950": ["n07747607", "orange"], "951": ["n07749582", "lemon"], "952": ["n07753113", "fig"], "953": ["n07753275", "pineapple"], "954": ["n07753592", "banana"], "955": ["n07754684", "jackfruit"], "956": ["n07760859", "custard_apple"], "957": ["n07768694", "pomegranate"], "958": ["n07802026", "hay"], "959": ["n07831146", "carbonara"], "960": ["n07836838", "chocolate_sauce"], "961": ["n07860988", "dough"], "962": ["n07871810", "meat_loaf"], "963": ["n07873807", "pizza"], "964": ["n07875152", "potpie"], "965": ["n07880968", "burrito"], "966": ["n07892512", "red_wine"], "967": ["n07920052", "espresso"], "968": ["n07930864", "cup"], "969": ["n07932039", "eggnog"], "970": ["n09193705", "alp"], "971": ["n09229709", "bubble"], "972": ["n09246464", "cliff"], "973": ["n09256479", "coral_reef"], "974": ["n09288635", "geyser"], "975": ["n09332890", "lakeside"], "976": ["n09399592", "promontory"], "977": ["n09421951", "sandbar"], "978": ["n09428293", "seashore"], "979": ["n09468604", "valley"], "980": ["n09472597", "volcano"], "981": ["n09835506", "ballplayer"], "982": ["n10148035", "groom"], "983": ["n10565667", "scuba_diver"], "984": ["n11879895", "rapeseed"], "985": ["n11939491", "daisy"], "986": ["n12057211", "yellow_lady's_slipper"], "987": ["n12144580", "corn"], "988": ["n12267677", "acorn"], "989": ["n12620546", "hip"], "990": ["n12768682", "buckeye"], "991": ["n12985857", "coral_fungus"], "992": ["n12998815", "agaric"], "993": ["n13037406", "gyromitra"], "994": ["n13040303", "stinkhorn"], "995": ["n13044778", "earthstar"], "996": ["n13052670", "hen-of-the-woods"], "997": ["n13054560", "bolete"], "998": ["n13133613", "ear"], "999": ["n15075141", "toilet_tissue"]}
"""
=====================================================
Optical Flow: Predicting movement with the RAFT model
=====================================================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_optical_flow.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_optical_flow.py>` to download the full example code.
Optical flow is the task of predicting movement between two images, usually two
consecutive frames of a video. Optical flow models take two images as input, and
predict a flow: the flow indicates the displacement of every single pixel in the
first image, and maps it to its corresponding pixel in the second image. Flows
are (2, H, W)-dimensional tensors, where the first axis corresponds to the
predicted horizontal and vertical displacements.
The following example illustrates how torchvision can be used to predict flows
using our implementation of the RAFT model. We will also see how to convert the
predicted flows to RGB images for visualization.
"""
import numpy as np
import torch
import matplotlib.pyplot as plt
import torchvision.transforms.functional as F
plt.rcParams["savefig.bbox"] = "tight"
# sphinx_gallery_thumbnail_number = 2
def plot(imgs, **imshow_kwargs):
if not isinstance(imgs[0], list):
# Make a 2d grid even if there's just 1 row
imgs = [imgs]
num_rows = len(imgs)
num_cols = len(imgs[0])
_, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)
for row_idx, row in enumerate(imgs):
for col_idx, img in enumerate(row):
ax = axs[row_idx, col_idx]
img = F.to_pil_image(img.to("cpu"))
ax.imshow(np.asarray(img), **imshow_kwargs)
ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
plt.tight_layout()
# %%
# Reading Videos Using Torchvision
# --------------------------------
# We will first read a video using :func:`~torchvision.io.read_video`.
# Alternatively one can use the new :class:`~torchvision.io.VideoReader` API (if
# torchvision is built from source).
# The video we will use here is free of use from `pexels.com
# <https://www.pexels.com/video/a-man-playing-a-game-of-basketball-5192157/>`_,
# credits go to `Pavel Danilyuk <https://www.pexels.com/@pavel-danilyuk>`_.
import tempfile
from pathlib import Path
from urllib.request import urlretrieve
video_url = "https://download.pytorch.org/tutorial/pexelscom_pavel_danilyuk_basketball_hd.mp4"
video_path = Path(tempfile.mkdtemp()) / "basketball.mp4"
_ = urlretrieve(video_url, video_path)
# %%
# :func:`~torchvision.io.read_video` returns the video frames, audio frames and
# the metadata associated with the video. In our case, we only need the video
# frames.
#
# Here we will just make 2 predictions between 2 pre-selected pairs of frames,
# namely frames (100, 101) and (150, 151). Each of these pairs corresponds to a
# single model input.
from torchvision.io import read_video
frames, _, _ = read_video(str(video_path), output_format="TCHW")
img1_batch = torch.stack([frames[100], frames[150]])
img2_batch = torch.stack([frames[101], frames[151]])
plot(img1_batch)
# %%
# The RAFT model accepts RGB images. We first get the frames from
# :func:`~torchvision.io.read_video` and resize them to ensure their dimensions
# are divisible by 8. Note that we explicitly use ``antialias=False``, because
# this is how those models were trained. Then we use the transforms bundled into
# the weights in order to preprocess the input and rescale its values to the
# required ``[-1, 1]`` interval.
from torchvision.models.optical_flow import Raft_Large_Weights
weights = Raft_Large_Weights.DEFAULT
transforms = weights.transforms()
def preprocess(img1_batch, img2_batch):
img1_batch = F.resize(img1_batch, size=[520, 960], antialias=False)
img2_batch = F.resize(img2_batch, size=[520, 960], antialias=False)
return transforms(img1_batch, img2_batch)
img1_batch, img2_batch = preprocess(img1_batch, img2_batch)
print(f"shape = {img1_batch.shape}, dtype = {img1_batch.dtype}")
# %%
# Estimating Optical flow using RAFT
# ----------------------------------
# We will use our RAFT implementation from
# :func:`~torchvision.models.optical_flow.raft_large`, which follows the same
# architecture as the one described in the `original paper <https://arxiv.org/abs/2003.12039>`_.
# We also provide the :func:`~torchvision.models.optical_flow.raft_small` model
# builder, which is smaller and faster to run, sacrificing a bit of accuracy.
from torchvision.models.optical_flow import raft_large
# If you can, run this example on a GPU, it will be a lot faster.
device = "cuda" if torch.cuda.is_available() else "cpu"
model = raft_large(weights=Raft_Large_Weights.DEFAULT, progress=False).to(device)
model = model.eval()
list_of_flows = model(img1_batch.to(device), img2_batch.to(device))
print(f"type = {type(list_of_flows)}")
print(f"length = {len(list_of_flows)} = number of iterations of the model")
# %%
# The RAFT model outputs lists of predicted flows where each entry is a
# (N, 2, H, W) batch of predicted flows that corresponds to a given "iteration"
# in the model. For more details on the iterative nature of the model, please
# refer to the `original paper <https://arxiv.org/abs/2003.12039>`_. Here, we
# are only interested in the final predicted flows (they are the most accurate
# ones), so we will just retrieve the last item in the list.
#
# As described above, a flow is a tensor with dimensions (2, H, W) (or (N, 2, H,
# W) for batches of flows) where each entry corresponds to the horizontal and
# vertical displacement of each pixel from the first image to the second image.
# Note that the predicted flows are in "pixel" unit, they are not normalized
# w.r.t. the dimensions of the images.
predicted_flows = list_of_flows[-1]
print(f"dtype = {predicted_flows.dtype}")
print(f"shape = {predicted_flows.shape} = (N, 2, H, W)")
print(f"min = {predicted_flows.min()}, max = {predicted_flows.max()}")
# %%
# Visualizing predicted flows
# ---------------------------
# Torchvision provides the :func:`~torchvision.utils.flow_to_image` utility to
# convert a flow into an RGB image. It also supports batches of flows.
# each "direction" in the flow will be mapped to a given RGB color. In the
# images below, pixels with similar colors are assumed by the model to be moving
# in similar directions. The model is properly able to predict the movement of
# the ball and the player. Note in particular the different predicted direction
# of the ball in the first image (going to the left) and in the second image
# (going up).
from torchvision.utils import flow_to_image
flow_imgs = flow_to_image(predicted_flows)
# The images have been mapped into [-1, 1] but for plotting we want them in [0, 1]
img1_batch = [(img1 + 1) / 2 for img1 in img1_batch]
grid = [[img1, flow_img] for (img1, flow_img) in zip(img1_batch, flow_imgs)]
plot(grid)
# %%
# Bonus: Creating GIFs of predicted flows
# ---------------------------------------
# In the example above we have only shown the predicted flows of 2 pairs of
# frames. A fun way to apply the Optical Flow models is to run the model on an
# entire video, and create a new video from all the predicted flows. Below is a
# snippet that can get you started with this. We comment out the code, because
# this example is being rendered on a machine without a GPU, and it would take
# too long to run it.
# from torchvision.io import write_jpeg
# for i, (img1, img2) in enumerate(zip(frames, frames[1:])):
# # Note: it would be faster to predict batches of flows instead of individual flows
# img1, img2 = preprocess(img1, img2)
# list_of_flows = model(img1.to(device), img2.to(device))
# predicted_flow = list_of_flows[-1][0]
# flow_img = flow_to_image(predicted_flow).to("cpu")
# output_folder = "/tmp/" # Update this to the folder of your choice
# write_jpeg(flow_img, output_folder + f"predicted_flow_{i}.jpg")
# %%
# Once the .jpg flow images are saved, you can convert them into a video or a
# GIF using ffmpeg with e.g.:
#
# ffmpeg -f image2 -framerate 30 -i predicted_flow_%d.jpg -loop -1 flow.gif
"""
=====================================
Repurposing masks into bounding boxes
=====================================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_repurposing_annotations.py>` to download the full example code.
The following example illustrates the operations available
the :ref:`torchvision.ops <ops>` module for repurposing
segmentation masks into object localization annotations for different tasks
(e.g. transforming masks used by instance and panoptic segmentation
methods into bounding boxes used by object detection methods).
"""
# sphinx_gallery_thumbnail_path = "../../gallery/assets/repurposing_annotations_thumbnail.png"
import os
import numpy as np
import torch
import matplotlib.pyplot as plt
import torchvision.transforms.functional as F
ASSETS_DIRECTORY = "../assets"
plt.rcParams["savefig.bbox"] = "tight"
def show(imgs):
if not isinstance(imgs, list):
imgs = [imgs]
fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)
for i, img in enumerate(imgs):
img = img.detach()
img = F.to_pil_image(img)
axs[0, i].imshow(np.asarray(img))
axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
# %%
# Masks
# -----
# In tasks like instance and panoptic segmentation, masks are commonly defined, and are defined by this package,
# as a multi-dimensional array (e.g. a NumPy array or a PyTorch tensor) with the following shape:
#
# (num_objects, height, width)
#
# Where num_objects is the number of annotated objects in the image. Each (height, width) object corresponds to exactly
# one object. For example, if your input image has the dimensions 224 x 224 and has four annotated objects the shape
# of your masks annotation has the following shape:
#
# (4, 224, 224).
#
# A nice property of masks is that they can be easily repurposed to be used in methods to solve a variety of object
# localization tasks.
# %%
# Converting Masks to Bounding Boxes
# -----------------------------------------------
# For example, the :func:`~torchvision.ops.masks_to_boxes` operation can be used to
# transform masks into bounding boxes that can be
# used as input to detection models such as FasterRCNN and RetinaNet.
# We will take images and masks from the `PenFudan Dataset <https://www.cis.upenn.edu/~jshi/ped_html/>`_.
from torchvision.io import read_image
img_path = os.path.join(ASSETS_DIRECTORY, "FudanPed00054.png")
mask_path = os.path.join(ASSETS_DIRECTORY, "FudanPed00054_mask.png")
img = read_image(img_path)
mask = read_image(mask_path)
# %%
# Here the masks are represented as a PNG Image, with floating point values.
# Each pixel is encoded as different colors, with 0 being background.
# Notice that the spatial dimensions of image and mask match.
print(mask.size())
print(img.size())
print(mask)
# %%
# We get the unique colors, as these would be the object ids.
obj_ids = torch.unique(mask)
# first id is the background, so remove it.
obj_ids = obj_ids[1:]
# split the color-encoded mask into a set of boolean masks.
# Note that this snippet would work as well if the masks were float values instead of ints.
masks = mask == obj_ids[:, None, None]
# %%
# Now the masks are a boolean tensor.
# The first dimension in this case 3 and denotes the number of instances: there are 3 people in the image.
# The other two dimensions are height and width, which are equal to the dimensions of the image.
# For each instance, the boolean tensors represent if the particular pixel
# belongs to the segmentation mask of the image.
print(masks.size())
print(masks)
# %%
# Let us visualize an image and plot its corresponding segmentation masks.
# We will use the :func:`~torchvision.utils.draw_segmentation_masks` to draw the segmentation masks.
from torchvision.utils import draw_segmentation_masks
drawn_masks = []
for mask in masks:
drawn_masks.append(draw_segmentation_masks(img, mask, alpha=0.8, colors="blue"))
show(drawn_masks)
# %%
# To convert the boolean masks into bounding boxes.
# We will use the :func:`~torchvision.ops.masks_to_boxes` from the torchvision.ops module
# It returns the boxes in ``(xmin, ymin, xmax, ymax)`` format.
from torchvision.ops import masks_to_boxes
boxes = masks_to_boxes(masks)
print(boxes.size())
print(boxes)
# %%
# As the shape denotes, there are 3 boxes and in ``(xmin, ymin, xmax, ymax)`` format.
# These can be visualized very easily with :func:`~torchvision.utils.draw_bounding_boxes` utility
# provided in :ref:`torchvision.utils <utils>`.
from torchvision.utils import draw_bounding_boxes
drawn_boxes = draw_bounding_boxes(img, boxes, colors="red")
show(drawn_boxes)
# %%
# These boxes can now directly be used by detection models in torchvision.
# Here is demo with a Faster R-CNN model loaded from
# :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`
from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights
weights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT
model = fasterrcnn_resnet50_fpn(weights=weights, progress=False)
print(img.size())
tranforms = weights.transforms()
img = tranforms(img)
target = {}
target["boxes"] = boxes
target["labels"] = labels = torch.ones((masks.size(0),), dtype=torch.int64)
detection_outputs = model(img.unsqueeze(0), [target])
# %%
# Converting Segmentation Dataset to Detection Dataset
# ----------------------------------------------------
#
# With this utility it becomes very simple to convert a segmentation dataset to a detection dataset.
# With this we can now use a segmentation dataset to train a detection model.
# One can similarly convert panoptic dataset to detection dataset.
# Here is an example where we re-purpose the dataset from the
# `PenFudan Detection Tutorial <https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html>`_.
class SegmentationToDetectionDataset(torch.utils.data.Dataset):
def __init__(self, root, transforms):
self.root = root
self.transforms = transforms
# load all image files, sorting them to
# ensure that they are aligned
self.imgs = list(sorted(os.listdir(os.path.join(root, "PNGImages"))))
self.masks = list(sorted(os.listdir(os.path.join(root, "PedMasks"))))
def __getitem__(self, idx):
# load images and masks
img_path = os.path.join(self.root, "PNGImages", self.imgs[idx])
mask_path = os.path.join(self.root, "PedMasks", self.masks[idx])
img = read_image(img_path)
mask = read_image(mask_path)
img = F.convert_image_dtype(img, dtype=torch.float)
mask = F.convert_image_dtype(mask, dtype=torch.float)
# We get the unique colors, as these would be the object ids.
obj_ids = torch.unique(mask)
# first id is the background, so remove it.
obj_ids = obj_ids[1:]
# split the color-encoded mask into a set of boolean masks.
masks = mask == obj_ids[:, None, None]
boxes = masks_to_boxes(masks)
# there is only one class
labels = torch.ones((masks.shape[0],), dtype=torch.int64)
target = {}
target["boxes"] = boxes
target["labels"] = labels
if self.transforms is not None:
img, target = self.transforms(img, target)
return img, target
"""
=========================
Tensor transforms and JIT
=========================
This example illustrates various features that are now supported by the
:ref:`image transformations <transforms>` on Tensor images. In particular, we
show how image transforms can be performed on GPU, and how one can also script
them using JIT compilation.
Prior to v0.8.0, transforms in torchvision have traditionally been PIL-centric
and presented multiple limitations due to that. Now, since v0.8.0, transforms
implementations are Tensor and PIL compatible and we can achieve the following
new features:
- transform multi-band torch tensor images (with more than 3-4 channels)
- torchscript transforms together with your model for deployment
- support for GPU acceleration
- batched transformation such as for videos
- read and decode data directly as torch tensor with torchscript support (for PNG and JPEG image formats)
===================
Torchscript support
===================
.. note::
These features are only possible with **Tensor** images.
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_scripted_tensor_transforms.py>` to download the full example code.
This example illustrates `torchscript
<https://pytorch.org/docs/stable/jit.html>`_ support of the torchvision
:ref:`transforms <transforms>` on Tensor images.
"""
# %%
from pathlib import Path
import matplotlib.pyplot as plt
import numpy as np
import torch
import torchvision.transforms as T
from torchvision.io import read_image
import torch.nn as nn
import torchvision.transforms as v1
from torchvision.io import read_image
plt.rcParams["savefig.bbox"] = 'tight'
torch.manual_seed(1)
# If you're trying to run that on collab, you can download the assets and the
# helpers from https://github.com/pytorch/vision/tree/main/gallery/
import sys
sys.path += ["../transforms"]
from helpers import plot
ASSETS_PATH = Path('../assets')
def show(imgs):
fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)
for i, img in enumerate(imgs):
img = T.ToPILImage()(img.to('cpu'))
axs[0, i].imshow(np.asarray(img))
axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
# %%
# Most transforms support torchscript. For composing transforms, we use
# :class:`torch.nn.Sequential` instead of
# :class:`~torchvision.transforms.v2.Compose`:
####################################
# The :func:`~torchvision.io.read_image` function allows to read an image and
# directly load it as a tensor
dog1 = read_image(str(Path('assets') / 'dog1.jpg'))
dog2 = read_image(str(Path('assets') / 'dog2.jpg'))
show([dog1, dog2])
####################################
# Transforming images on GPU
# --------------------------
# Most transforms natively support tensors on top of PIL images (to visualize
# the effect of the transforms, you may refer to see
# :ref:`sphx_glr_auto_examples_plot_transforms.py`).
# Using tensor images, we can run the transforms on GPUs if cuda is available!
import torch.nn as nn
dog1 = read_image(str(ASSETS_PATH / 'dog1.jpg'))
dog2 = read_image(str(ASSETS_PATH / 'dog2.jpg'))
transforms = torch.nn.Sequential(
T.RandomCrop(224),
T.RandomHorizontalFlip(p=0.3),
v1.RandomCrop(224),
v1.RandomHorizontalFlip(p=0.3),
)
device = 'cuda' if torch.cuda.is_available() else 'cpu'
dog1 = dog1.to(device)
dog2 = dog2.to(device)
scripted_transforms = torch.jit.script(transforms)
transformed_dog1 = transforms(dog1)
transformed_dog2 = transforms(dog2)
show([transformed_dog1, transformed_dog2])
plot([dog1, scripted_transforms(dog1), dog2, scripted_transforms(dog2)])
####################################
# Scriptable transforms for easier deployment via torchscript
# -----------------------------------------------------------
# We now show how to combine image transformations and a model forward pass,
# while using ``torch.jit.script`` to obtain a single scripted module.
# %%
# .. warning::
#
# Above we have used transforms from the ``torchvision.transforms``
# namespace, i.e. the "v1" transforms. The v2 transforms from the
# ``torchvision.transforms.v2`` namespace are the :ref:`recommended
# <v1_or_v2>` way to use transforms in your code.
#
# The v2 transforms also support torchscript, but if you call
# ``torch.jit.script()`` on a v2 **class** transform, you'll actually end up
# with its (scripted) v1 equivalent. This may lead to slightly different
# results between the scripted and eager executions due to implementation
# differences between v1 and v2.
#
# If you really need torchscript support for the v2 transforms, **we
# recommend scripting the functionals** from the
# ``torchvision.transforms.v2.functional`` namespace to avoid surprises.
#
# Below we now show how to combine image transformations and a model forward
# pass, while using ``torch.jit.script`` to obtain a single scripted module.
#
# Let's define a ``Predictor`` module that transforms the input tensor and then
# applies an ImageNet model on it.
from torchvision.models import resnet18
from torchvision.models import resnet18, ResNet18_Weights
class Predictor(nn.Module):
def __init__(self):
super().__init__()
self.resnet18 = resnet18(pretrained=True, progress=False).eval()
self.transforms = nn.Sequential(
T.Resize([256, ]), # We use single int value inside a list due to torchscript type restrictions
T.CenterCrop(224),
T.ConvertImageDtype(torch.float),
T.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
)
weights = ResNet18_Weights.DEFAULT
self.resnet18 = resnet18(weights=weights, progress=False).eval()
self.transforms = weights.transforms(antialias=True)
def forward(self, x: torch.Tensor) -> torch.Tensor:
with torch.no_grad():
......@@ -107,10 +94,12 @@ class Predictor(nn.Module):
return y_pred.argmax(dim=1)
####################################
# %%
# Now, let's define scripted and non-scripted instances of ``Predictor`` and
# apply it on multiple tensor images of the same size
device = "cuda" if torch.cuda.is_available() else "cpu"
predictor = Predictor().to(device)
scripted_predictor = torch.jit.script(predictor).to(device)
......@@ -119,21 +108,21 @@ batch = torch.stack([dog1, dog2]).to(device)
res = predictor(batch)
res_scripted = scripted_predictor(batch)
####################################
# %%
# We can verify that the prediction of the scripted and non-scripted models are
# the same:
import json
with open(Path('assets') / 'imagenet_class_index.json', 'r') as labels_file:
with open(Path('../assets') / 'imagenet_class_index.json') as labels_file:
labels = json.load(labels_file)
for i, (pred, pred_scripted) in enumerate(zip(res, res_scripted)):
assert pred == pred_scripted
print(f"Prediction for Dog {i + 1}: {labels[str(pred.item())]}")
####################################
# Since the model is scripted, it can be easily dumped on disk an re-used
# %%
# Since the model is scripted, it can be easily dumped on disk and re-used
import tempfile
......@@ -143,3 +132,5 @@ with tempfile.NamedTemporaryFile() as f:
dumped_scripted_predictor = torch.jit.load(f.name)
res_scripted_dumped = dumped_scripted_predictor(batch)
assert (res_scripted_dumped == res_scripted).all()
# %%
"""
=========
Video API
=========
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_video_api.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_video_api.py>` to download the full example code.
This example illustrates some of the APIs that torchvision offers for
videos, together with the examples on how to build datasets and more.
"""
# %%
# 1. Introduction: building a new video object and examining the properties
# -------------------------------------------------------------------------
# First we select a video to test the object out. For the sake of argument
# we're using one from kinetics400 dataset.
# To create it, we need to define the path and the stream we want to use.
# %%
# Chosen video statistics:
#
# - WUzgd7C1pWA.mp4
# - source:
# - kinetics-400
# - video:
# - H-264
# - MPEG-4 AVC (part 10) (avc1)
# - fps: 29.97
# - audio:
# - MPEG AAC audio (mp4a)
# - sample rate: 48K Hz
#
import torch
import torchvision
from torchvision.datasets.utils import download_url
torchvision.set_video_backend("video_reader")
# Download the sample video
download_url(
"https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true",
".",
"WUzgd7C1pWA.mp4"
)
video_path = "./WUzgd7C1pWA.mp4"
# %%
# Streams are defined in a similar fashion as torch devices. We encode them as strings in a form
# of ``stream_type:stream_id`` where ``stream_type`` is a string and ``stream_id`` a long int.
# The constructor accepts passing a ``stream_type`` only, in which case the stream is auto-discovered.
# Firstly, let's get the metadata for our particular video:
stream = "video"
video = torchvision.io.VideoReader(video_path, stream)
video.get_metadata()
# %%
# Here we can see that video has two streams - a video and an audio stream.
# Currently available stream types include ['video', 'audio'].
# Each descriptor consists of two parts: stream type (e.g. 'video') and a unique stream id
# (which are determined by video encoding).
# In this way, if the video container contains multiple streams of the same type,
# users can access the one they want.
# If only stream type is passed, the decoder auto-detects first stream of that type and returns it.
# %%
# Let's read all the frames from the video stream. By default, the return value of
# ``next(video_reader)`` is a dict containing the following fields.
#
# The return fields are:
#
# - ``data``: containing a torch.tensor
# - ``pts``: containing a float timestamp of this particular frame
metadata = video.get_metadata()
video.set_current_stream("audio")
frames = [] # we are going to save the frames here.
ptss = [] # pts is a presentation timestamp in seconds (float) of each frame
for frame in video:
frames.append(frame['data'])
ptss.append(frame['pts'])
print("PTS for first five frames ", ptss[:5])
print("Total number of frames: ", len(frames))
approx_nf = metadata['audio']['duration'][0] * metadata['audio']['framerate'][0]
print("Approx total number of datapoints we can expect: ", approx_nf)
print("Read data size: ", frames[0].size(0) * len(frames))
# %%
# But what if we only want to read certain time segment of the video?
# That can be done easily using the combination of our ``seek`` function, and the fact that each call
# to next returns the presentation timestamp of the returned frame in seconds.
#
# Given that our implementation relies on python iterators,
# we can leverage itertools to simplify the process and make it more pythonic.
#
# For example, if we wanted to read ten frames from second second:
import itertools
video.set_current_stream("video")
frames = [] # we are going to save the frames here.
# We seek into a second second of the video and use islice to get 10 frames since
for frame, pts in itertools.islice(video.seek(2), 10):
frames.append(frame)
print("Total number of frames: ", len(frames))
# %%
# Or if we wanted to read from 2nd to 5th second,
# We seek into a second second of the video,
# then we utilize the itertools takewhile to get the
# correct number of frames:
video.set_current_stream("video")
frames = [] # we are going to save the frames here.
video = video.seek(2)
for frame in itertools.takewhile(lambda x: x['pts'] <= 5, video):
frames.append(frame['data'])
print("Total number of frames: ", len(frames))
approx_nf = (5 - 2) * video.get_metadata()['video']['fps'][0]
print("We can expect approx: ", approx_nf)
print("Tensor size: ", frames[0].size())
# %%
# 2. Building a sample read_video function
# ----------------------------------------------------------------------------------------
# We can utilize the methods above to build the read video function that follows
# the same API to the existing ``read_video`` function.
def example_read_video(video_object, start=0, end=None, read_video=True, read_audio=True):
if end is None:
end = float("inf")
if end < start:
raise ValueError(
"end time should be larger than start time, got "
f"start time={start} and end time={end}"
)
video_frames = torch.empty(0)
video_pts = []
if read_video:
video_object.set_current_stream("video")
frames = []
for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):
frames.append(frame['data'])
video_pts.append(frame['pts'])
if len(frames) > 0:
video_frames = torch.stack(frames, 0)
audio_frames = torch.empty(0)
audio_pts = []
if read_audio:
video_object.set_current_stream("audio")
frames = []
for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):
frames.append(frame['data'])
audio_pts.append(frame['pts'])
if len(frames) > 0:
audio_frames = torch.cat(frames, 0)
return video_frames, audio_frames, (video_pts, audio_pts), video_object.get_metadata()
# Total number of frames should be 327 for video and 523264 datapoints for audio
vf, af, info, meta = example_read_video(video)
print(vf.size(), af.size())
# %%
# 3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)
# -------------------------------------------------------------------------------------------------------
# Cool, so now we can use the same principle to make the sample dataset.
# We suggest trying out iterable dataset for this purpose.
# Here, we are going to build an example dataset that reads randomly selected 10 frames of video.
# %%
# Make sample dataset
import os
os.makedirs("./dataset", exist_ok=True)
os.makedirs("./dataset/1", exist_ok=True)
os.makedirs("./dataset/2", exist_ok=True)
# %%
# Download the videos
from torchvision.datasets.utils import download_url
download_url(
"https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true",
"./dataset/1", "WUzgd7C1pWA.mp4"
)
download_url(
"https://github.com/pytorch/vision/blob/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi?raw=true",
"./dataset/1",
"RATRACE_wave_f_nm_np1_fr_goo_37.avi"
)
download_url(
"https://github.com/pytorch/vision/blob/main/test/assets/videos/SOX5yA1l24A.mp4?raw=true",
"./dataset/2",
"SOX5yA1l24A.mp4"
)
download_url(
"https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi?raw=true",
"./dataset/2",
"v_SoccerJuggling_g23_c01.avi"
)
download_url(
"https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi?raw=true",
"./dataset/2",
"v_SoccerJuggling_g24_c01.avi"
)
# %%
# Housekeeping and utilities
import os
import random
from torchvision.datasets.folder import make_dataset
from torchvision import transforms as t
def _find_classes(dir):
classes = [d.name for d in os.scandir(dir) if d.is_dir()]
classes.sort()
class_to_idx = {cls_name: i for i, cls_name in enumerate(classes)}
return classes, class_to_idx
def get_samples(root, extensions=(".mp4", ".avi")):
_, class_to_idx = _find_classes(root)
return make_dataset(root, class_to_idx, extensions=extensions)
# %%
# We are going to define the dataset and some basic arguments.
# We assume the structure of the FolderDataset, and add the following parameters:
#
# - ``clip_len``: length of a clip in frames
# - ``frame_transform``: transform for every frame individually
# - ``video_transform``: transform on a video sequence
#
# .. note::
# We actually add epoch size as using :func:`~torch.utils.data.IterableDataset`
# class allows us to naturally oversample clips or images from each video if needed.
class RandomDataset(torch.utils.data.IterableDataset):
def __init__(self, root, epoch_size=None, frame_transform=None, video_transform=None, clip_len=16):
super(RandomDataset).__init__()
self.samples = get_samples(root)
# Allow for temporal jittering
if epoch_size is None:
epoch_size = len(self.samples)
self.epoch_size = epoch_size
self.clip_len = clip_len
self.frame_transform = frame_transform
self.video_transform = video_transform
def __iter__(self):
for i in range(self.epoch_size):
# Get random sample
path, target = random.choice(self.samples)
# Get video object
vid = torchvision.io.VideoReader(path, "video")
metadata = vid.get_metadata()
video_frames = [] # video frame buffer
# Seek and return frames
max_seek = metadata["video"]['duration'][0] - (self.clip_len / metadata["video"]['fps'][0])
start = random.uniform(0., max_seek)
for frame in itertools.islice(vid.seek(start), self.clip_len):
video_frames.append(self.frame_transform(frame['data']))
current_pts = frame['pts']
# Stack it into a tensor
video = torch.stack(video_frames, 0)
if self.video_transform:
video = self.video_transform(video)
output = {
'path': path,
'video': video,
'target': target,
'start': start,
'end': current_pts}
yield output
# %%
# Given a path of videos in a folder structure, i.e:
#
# - dataset
# - class 1
# - file 0
# - file 1
# - ...
# - class 2
# - file 0
# - file 1
# - ...
# - ...
#
# We can generate a dataloader and test the dataset.
transforms = [t.Resize((112, 112))]
frame_transform = t.Compose(transforms)
dataset = RandomDataset("./dataset", epoch_size=None, frame_transform=frame_transform)
# %%
from torch.utils.data import DataLoader
loader = DataLoader(dataset, batch_size=12)
data = {"video": [], 'start': [], 'end': [], 'tensorsize': []}
for batch in loader:
for i in range(len(batch['path'])):
data['video'].append(batch['path'][i])
data['start'].append(batch['start'][i].item())
data['end'].append(batch['end'][i].item())
data['tensorsize'].append(batch['video'][i].size())
print(data)
# %%
# 4. Data Visualization
# ----------------------------------
# Example of visualized video
import matplotlib.pyplot as plt
plt.figure(figsize=(12, 12))
for i in range(16):
plt.subplot(4, 4, i + 1)
plt.imshow(batch["video"][0, i, ...].permute(1, 2, 0))
plt.axis("off")
# %%
# Cleanup the video and dataset:
import os
import shutil
os.remove("./WUzgd7C1pWA.mp4")
shutil.rmtree("./dataset")
......@@ -3,10 +3,15 @@
Visualization utilities
=======================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_visualization_utils.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_visualization_utils.py>` to download the full example code.
This example illustrates some of the utilities that torchvision offers for
visualizing images, bounding boxes, and segmentation masks.
visualizing images, bounding boxes, segmentation masks and keypoints.
"""
# sphinx_gallery_thumbnail_path = "../../gallery/assets/visualization_utils_thumbnail2.png"
import torch
import numpy as np
......@@ -21,7 +26,7 @@ plt.rcParams["savefig.bbox"] = 'tight'
def show(imgs):
if not isinstance(imgs, list):
imgs = [imgs]
fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)
fig, axs = plt.subplots(ncols=len(imgs), squeeze=False)
for i, img in enumerate(imgs):
img = img.detach()
img = F.to_pil_image(img)
......@@ -29,7 +34,7 @@ def show(imgs):
axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
####################################
# %%
# Visualizing a grid of images
# ----------------------------
# The :func:`~torchvision.utils.make_grid` function can be used to create a
......@@ -40,13 +45,14 @@ from torchvision.utils import make_grid
from torchvision.io import read_image
from pathlib import Path
dog1_int = read_image(str(Path('assets') / 'dog1.jpg'))
dog2_int = read_image(str(Path('assets') / 'dog2.jpg'))
dog1_int = read_image(str(Path('../assets') / 'dog1.jpg'))
dog2_int = read_image(str(Path('../assets') / 'dog2.jpg'))
dog_list = [dog1_int, dog2_int]
grid = make_grid([dog1_int, dog2_int, dog1_int, dog2_int])
grid = make_grid(dog_list)
show(grid)
####################################
# %%
# Visualizing bounding boxes
# --------------------------
# We can use :func:`~torchvision.utils.draw_bounding_boxes` to draw boxes on an
......@@ -62,41 +68,39 @@ result = draw_bounding_boxes(dog1_int, boxes, colors=colors, width=5)
show(result)
#####################################
# %%
# Naturally, we can also plot bounding boxes produced by torchvision detection
# models. Here is demo with a Faster R-CNN model loaded from
# models. Here is a demo with a Faster R-CNN model loaded from
# :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`
# model. You can also try using a RetinaNet with
# :func:`~torchvision.models.detection.retinanet_resnet50_fpn`, an SSDlite with
# :func:`~torchvision.models.detection.ssdlite320_mobilenet_v3_large` or an SSD with
# :func:`~torchvision.models.detection.ssd300_vgg16`. For more details
# on the output of such models, you may refer to :ref:`instance_seg_output`.
# model. For more details on the output of such models, you may
# refer to :ref:`instance_seg_output`.
from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights
from torchvision.models.detection import fasterrcnn_resnet50_fpn
from torchvision.transforms.functional import convert_image_dtype
weights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT
transforms = weights.transforms()
batch_int = torch.stack([dog1_int, dog2_int])
batch = convert_image_dtype(batch_int, dtype=torch.float)
images = [transforms(d) for d in dog_list]
model = fasterrcnn_resnet50_fpn(pretrained=True, progress=False)
model = fasterrcnn_resnet50_fpn(weights=weights, progress=False)
model = model.eval()
outputs = model(batch)
outputs = model(images)
print(outputs)
#####################################
# %%
# Let's plot the boxes detected by our model. We will only plot the boxes with a
# score greater than a given threshold.
score_threshold = .8
dogs_with_boxes = [
draw_bounding_boxes(dog_int, boxes=output['boxes'][output['scores'] > score_threshold], width=4)
for dog_int, output in zip(batch_int, outputs)
for dog_int, output in zip(dog_list, outputs)
]
show(dogs_with_boxes)
#####################################
# %%
# Visualizing segmentation masks
# ------------------------------
# The :func:`~torchvision.utils.draw_segmentation_masks` function can be used to
......@@ -110,26 +114,22 @@ show(dogs_with_boxes)
# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
#
# We will see how to use it with torchvision's FCN Resnet-50, loaded with
# :func:`~torchvision.models.segmentation.fcn_resnet50`. You can also try using
# DeepLabv3 (:func:`~torchvision.models.segmentation.deeplabv3_resnet50`) or
# lraspp mobilenet models
# (:func:`~torchvision.models.segmentation.lraspp_mobilenet_v3_large`).
#
# Let's start by looking at the ouput of the model. Remember that in general,
# images must be normalized before they're passed to a semantic segmentation
# model.
# :func:`~torchvision.models.segmentation.fcn_resnet50`. Let's start by looking
# at the output of the model.
from torchvision.models.segmentation import fcn_resnet50
from torchvision.models.segmentation import fcn_resnet50, FCN_ResNet50_Weights
weights = FCN_ResNet50_Weights.DEFAULT
transforms = weights.transforms(resize_size=None)
model = fcn_resnet50(pretrained=True, progress=False)
model = fcn_resnet50(weights=weights, progress=False)
model = model.eval()
normalized_batch = F.normalize(batch, mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))
output = model(normalized_batch)['out']
batch = torch.stack([transforms(d) for d in dog_list])
output = model(batch)['out']
print(output.shape, output.min().item(), output.max().item())
#####################################
# %%
# As we can see above, the output of the segmentation model is a tensor of shape
# ``(batch_size, num_classes, H, W)``. Each value is a non-normalized score, and
# we can normalize them into ``[0, 1]`` by using a softmax. After the softmax,
......@@ -139,24 +139,19 @@ print(output.shape, output.min().item(), output.max().item())
# Let's plot the masks that have been detected for the dog class and for the
# boat class:
sem_classes = [
'__background__', 'aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus',
'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', 'horse', 'motorbike',
'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor'
]
sem_class_to_idx = {cls: idx for (idx, cls) in enumerate(sem_classes)}
sem_class_to_idx = {cls: idx for (idx, cls) in enumerate(weights.meta["categories"])}
normalized_masks = torch.nn.functional.softmax(output, dim=1)
dog_and_boat_masks = [
normalized_masks[img_idx, sem_class_to_idx[cls]]
for img_idx in range(batch.shape[0])
for img_idx in range(len(dog_list))
for cls in ('dog', 'boat')
]
show(dog_and_boat_masks)
#####################################
# %%
# As expected, the model is confident about the dog class, but not so much for
# the boat class.
#
......@@ -171,7 +166,7 @@ print(f"shape = {boolean_dog_masks.shape}, dtype = {boolean_dog_masks.dtype}")
show([m.float() for m in boolean_dog_masks])
#####################################
# %%
# The line above where we define ``boolean_dog_masks`` is a bit cryptic, but you
# can read it as the following query: "For which pixels is 'dog' the most likely
# class?"
......@@ -189,15 +184,15 @@ from torchvision.utils import draw_segmentation_masks
dogs_with_masks = [
draw_segmentation_masks(img, masks=mask, alpha=0.7)
for img, mask in zip(batch_int, boolean_dog_masks)
for img, mask in zip(dog_list, boolean_dog_masks)
]
show(dogs_with_masks)
#####################################
# %%
# We can plot more than one mask per image! Remember that the model returned as
# many masks as there are classes. Let's ask the same query as above, but this
# time for *all* classes, not just the dog class: "For each pixel and each class
# C, is class C the most most likely class?"
# C, is class C the most likely class?"
#
# This one is a bit more involved, so we'll first show how to do it with a
# single image, and then we'll generalize to the batch
......@@ -213,7 +208,7 @@ print(f"dog1_all_classes_masks = {dog1_all_classes_masks.shape}, dtype = {dog1_a
dog_with_all_masks = draw_segmentation_masks(dog1_int, masks=dog1_all_classes_masks, alpha=.6)
show(dog_with_all_masks)
#####################################
# %%
# We can see in the image above that only 2 masks were drawn: the mask for the
# background and the mask for the dog. This is because the model thinks that
# only these 2 classes are the most likely ones across all the pixels. If the
......@@ -235,12 +230,12 @@ all_classes_masks = all_classes_masks.swapaxes(0, 1)
dogs_with_masks = [
draw_segmentation_masks(img, masks=mask, alpha=.6)
for img, mask in zip(batch_int, all_classes_masks)
for img, mask in zip(dog_list, all_classes_masks)
]
show(dogs_with_masks)
#####################################
# %%
# .. _instance_seg_output:
#
# Instance segmentation models
......@@ -261,14 +256,20 @@ show(dogs_with_masks)
# of them may not have masks, like
# :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`.
from torchvision.models.detection import maskrcnn_resnet50_fpn
model = maskrcnn_resnet50_fpn(pretrained=True, progress=False)
from torchvision.models.detection import maskrcnn_resnet50_fpn, MaskRCNN_ResNet50_FPN_Weights
weights = MaskRCNN_ResNet50_FPN_Weights.DEFAULT
transforms = weights.transforms()
images = [transforms(d) for d in dog_list]
model = maskrcnn_resnet50_fpn(weights=weights, progress=False)
model = model.eval()
output = model(batch)
output = model(images)
print(output)
#####################################
# %%
# Let's break this down. For each image in the batch, the model outputs some
# detections (or instances). The number of detections varies for each input
# image. Each instance is described by its bounding box, its label, its score
......@@ -291,33 +292,16 @@ dog1_masks = dog1_output['masks']
print(f"shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}, "
f"min = {dog1_masks.min()}, max = {dog1_masks.max()}")
#####################################
# Here the masks corresponds to probabilities indicating, for each pixel, how
# %%
# Here the masks correspond to probabilities indicating, for each pixel, how
# likely it is to belong to the predicted label of that instance. Those
# predicted labels correspond to the 'labels' element in the same output dict.
# Let's see which labels were predicted for the instances of the first image.
inst_classes = [
'__background__', 'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus',
'train', 'truck', 'boat', 'traffic light', 'fire hydrant', 'N/A', 'stop sign',
'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
'elephant', 'bear', 'zebra', 'giraffe', 'N/A', 'backpack', 'umbrella', 'N/A', 'N/A',
'handbag', 'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball',
'kite', 'baseball bat', 'baseball glove', 'skateboard', 'surfboard', 'tennis racket',
'bottle', 'N/A', 'wine glass', 'cup', 'fork', 'knife', 'spoon', 'bowl',
'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza',
'donut', 'cake', 'chair', 'couch', 'potted plant', 'bed', 'N/A', 'dining table',
'N/A', 'N/A', 'toilet', 'N/A', 'tv', 'laptop', 'mouse', 'remote', 'keyboard', 'cell phone',
'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'N/A', 'book',
'clock', 'vase', 'scissors', 'teddy bear', 'hair drier', 'toothbrush'
]
inst_class_to_idx = {cls: idx for (idx, cls) in enumerate(inst_classes)}
print("For the first dog, the following instances were detected:")
print([inst_classes[label] for label in dog1_output['labels']])
print([weights.meta["categories"][label] for label in dog1_output['labels']])
#####################################
# %%
# Interestingly, the model detects two persons in the image. Let's go ahead and
# plot those masks. Since :func:`~torchvision.utils.draw_segmentation_masks`
# expects boolean masks, we need to convert those probabilities into boolean
......@@ -335,15 +319,15 @@ dog1_bool_masks = dog1_bool_masks.squeeze(1)
show(draw_segmentation_masks(dog1_int, dog1_bool_masks, alpha=0.9))
#####################################
# %%
# The model seems to have properly detected the dog, but it also confused trees
# with people. Looking more closely at the scores will help us plotting more
# with people. Looking more closely at the scores will help us plot more
# relevant masks:
print(dog1_output['scores'])
#####################################
# Clearly the model is less confident about the dog detection than it is about
# %%
# Clearly the model is more confident about the dog detection than it is about
# the people detections. That's good news. When plotting the masks, we can ask
# for only those that have a good score. Let's use a score threshold of .75
# here, and also plot the masks of the second dog.
......@@ -357,11 +341,182 @@ boolean_masks = [
dogs_with_masks = [
draw_segmentation_masks(img, mask.squeeze(1))
for img, mask in zip(batch_int, boolean_masks)
for img, mask in zip(dog_list, boolean_masks)
]
show(dogs_with_masks)
#####################################
# %%
# The two 'people' masks in the first image where not selected because they have
# a lower score than the score threshold. Similarly in the second image, the
# a lower score than the score threshold. Similarly, in the second image, the
# instance with class 15 (which corresponds to 'bench') was not selected.
# %%
# .. _keypoint_output:
#
# Visualizing keypoints
# ------------------------------
# The :func:`~torchvision.utils.draw_keypoints` function can be used to
# draw keypoints on images. We will see how to use it with
# torchvision's KeypointRCNN loaded with :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`.
# We will first have a look at output of the model.
#
from torchvision.models.detection import keypointrcnn_resnet50_fpn, KeypointRCNN_ResNet50_FPN_Weights
from torchvision.io import read_image
person_int = read_image(str(Path("../assets") / "person1.jpg"))
weights = KeypointRCNN_ResNet50_FPN_Weights.DEFAULT
transforms = weights.transforms()
person_float = transforms(person_int)
model = keypointrcnn_resnet50_fpn(weights=weights, progress=False)
model = model.eval()
outputs = model([person_float])
print(outputs)
# %%
# As we see the output contains a list of dictionaries.
# The output list is of length batch_size.
# We currently have just a single image so length of list is 1.
# Each entry in the list corresponds to an input image,
# and it is a dict with keys `boxes`, `labels`, `scores`, `keypoints` and `keypoint_scores`.
# Each value associated to those keys has `num_instances` elements in it.
# In our case above there are 2 instances detected in the image.
kpts = outputs[0]['keypoints']
scores = outputs[0]['scores']
print(kpts)
print(scores)
# %%
# The KeypointRCNN model detects there are two instances in the image.
# If you plot the boxes by using :func:`~draw_bounding_boxes`
# you would recognize they are the person and the surfboard.
# If we look at the scores, we will realize that the model is much more confident about the person than surfboard.
# We could now set a threshold confidence and plot instances which we are confident enough.
# Let us set a threshold of 0.75 and filter out the keypoints corresponding to the person.
detect_threshold = 0.75
idx = torch.where(scores > detect_threshold)
keypoints = kpts[idx]
print(keypoints)
# %%
# Great, now we have the keypoints corresponding to the person.
# Each keypoint is represented by x, y coordinates and the visibility.
# We can now use the :func:`~torchvision.utils.draw_keypoints` function to draw keypoints.
# Note that the utility expects uint8 images.
from torchvision.utils import draw_keypoints
res = draw_keypoints(person_int, keypoints, colors="blue", radius=3)
show(res)
# %%
# As we see, the keypoints appear as colored circles over the image.
# The coco keypoints for a person are ordered and represent the following list.\
coco_keypoints = [
"nose", "left_eye", "right_eye", "left_ear", "right_ear",
"left_shoulder", "right_shoulder", "left_elbow", "right_elbow",
"left_wrist", "right_wrist", "left_hip", "right_hip",
"left_knee", "right_knee", "left_ankle", "right_ankle",
]
# %%
# What if we are interested in joining the keypoints?
# This is especially useful in creating pose detection or action recognition.
# We can join the keypoints easily using the `connectivity` parameter.
# A close observation would reveal that we would need to join the points in below
# order to construct human skeleton.
#
# nose -> left_eye -> left_ear. (0, 1), (1, 3)
#
# nose -> right_eye -> right_ear. (0, 2), (2, 4)
#
# nose -> left_shoulder -> left_elbow -> left_wrist. (0, 5), (5, 7), (7, 9)
#
# nose -> right_shoulder -> right_elbow -> right_wrist. (0, 6), (6, 8), (8, 10)
#
# left_shoulder -> left_hip -> left_knee -> left_ankle. (5, 11), (11, 13), (13, 15)
#
# right_shoulder -> right_hip -> right_knee -> right_ankle. (6, 12), (12, 14), (14, 16)
#
# We will create a list containing these keypoint ids to be connected.
connect_skeleton = [
(0, 1), (0, 2), (1, 3), (2, 4), (0, 5), (0, 6), (5, 7), (6, 8),
(7, 9), (8, 10), (5, 11), (6, 12), (11, 13), (12, 14), (13, 15), (14, 16)
]
# %%
# We pass the above list to the connectivity parameter to connect the keypoints.
#
res = draw_keypoints(person_int, keypoints, connectivity=connect_skeleton, colors="blue", radius=4, width=3)
show(res)
# %%
# That looks pretty good.
#
# .. _draw_keypoints_with_visibility:
#
# Drawing Keypoints with Visibility
# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
# Let's have a look at the results, another keypoint prediction module produced, and show the connectivity:
prediction = torch.tensor(
[[[208.0176, 214.2409, 1.0000],
[000.0000, 000.0000, 0.0000],
[197.8246, 210.6392, 1.0000],
[000.0000, 000.0000, 0.0000],
[178.6378, 217.8425, 1.0000],
[221.2086, 253.8591, 1.0000],
[160.6502, 269.4662, 1.0000],
[243.9929, 304.2822, 1.0000],
[138.4654, 328.8935, 1.0000],
[277.5698, 340.8990, 1.0000],
[153.4551, 374.5145, 1.0000],
[000.0000, 000.0000, 0.0000],
[226.0053, 370.3125, 1.0000],
[221.8081, 455.5516, 1.0000],
[273.9723, 448.9486, 1.0000],
[193.6275, 546.1933, 1.0000],
[273.3727, 545.5930, 1.0000]]]
)
res = draw_keypoints(person_int, prediction, connectivity=connect_skeleton, colors="blue", radius=4, width=3)
show(res)
# %%
# What happened there?
# The model, which predicted the new keypoints,
# can't detect the three points that are hidden on the upper left body of the skateboarder.
# More precisely, the model predicted that `(x, y, vis) = (0, 0, 0)` for the left_eye, left_ear, and left_hip.
# So we definitely don't want to display those keypoints and connections, and you don't have to.
# Looking at the parameters of :func:`~torchvision.utils.draw_keypoints`,
# we can see that we can pass a visibility tensor as an additional argument.
# Given the models' prediction, we have the visibility as the third keypoint dimension, we just need to extract it.
# Let's split the ``prediction`` into the keypoint coordinates and their respective visibility,
# and pass both of them as arguments to :func:`~torchvision.utils.draw_keypoints`.
coordinates, visibility = prediction.split([2, 1], dim=-1)
visibility = visibility.bool()
res = draw_keypoints(
person_int, coordinates, visibility=visibility, connectivity=connect_skeleton, colors="blue", radius=4, width=3
)
show(res)
# %%
# We can see that the undetected keypoints are not draw and the invisible keypoint connections were skipped.
# This can reduce the noise on images with multiple detections, or in cases like ours,
# when the keypoint-prediction model missed some detections.
# Most torch keypoint-prediction models return the visibility for every prediction, ready for you to use it.
# The :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn` model,
# which we used in the first case, does so too.
.. _transforms_gallery:
Transforms
----------
import matplotlib.pyplot as plt
import torch
from torchvision.utils import draw_bounding_boxes, draw_segmentation_masks
from torchvision import tv_tensors
from torchvision.transforms.v2 import functional as F
def plot(imgs, row_title=None, **imshow_kwargs):
if not isinstance(imgs[0], list):
# Make a 2d grid even if there's just 1 row
imgs = [imgs]
num_rows = len(imgs)
num_cols = len(imgs[0])
_, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)
for row_idx, row in enumerate(imgs):
for col_idx, img in enumerate(row):
boxes = None
masks = None
if isinstance(img, tuple):
img, target = img
if isinstance(target, dict):
boxes = target.get("boxes")
masks = target.get("masks")
elif isinstance(target, tv_tensors.BoundingBoxes):
boxes = target
else:
raise ValueError(f"Unexpected target type: {type(target)}")
img = F.to_image(img)
if img.dtype.is_floating_point and img.min() < 0:
# Poor man's re-normalization for the colors to be OK-ish. This
# is useful for images coming out of Normalize()
img -= img.min()
img /= img.max()
img = F.to_dtype(img, torch.uint8, scale=True)
if boxes is not None:
img = draw_bounding_boxes(img, boxes, colors="yellow", width=3)
if masks is not None:
img = draw_segmentation_masks(img, masks.to(torch.bool), colors=["green"] * masks.shape[0], alpha=.65)
ax = axs[row_idx, col_idx]
ax.imshow(img.permute(1, 2, 0).numpy(), **imshow_kwargs)
ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
if row_title is not None:
for row_idx in range(num_rows):
axs[row_idx, 0].set(ylabel=row_title[row_idx])
plt.tight_layout()
"""
===================================
How to write your own v2 transforms
===================================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_transforms.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_transforms.py>` to download the full example code.
This guide explains how to write transforms that are compatible with the
torchvision transforms V2 API.
"""
# %%
import torch
from torchvision import tv_tensors
from torchvision.transforms import v2
# %%
# Just create a ``nn.Module`` and override the ``forward`` method
# ===============================================================
#
# In most cases, this is all you're going to need, as long as you already know
# the structure of the input that your transform will expect. For example if
# you're just doing image classification, your transform will typically accept a
# single image as input, or a ``(img, label)`` input. So you can just hard-code
# your ``forward`` method to accept just that, e.g.
#
# .. code:: python
#
# class MyCustomTransform(torch.nn.Module):
# def forward(self, img, label):
# # Do some transformations
# return new_img, new_label
#
# .. note::
#
# This means that if you have a custom transform that is already compatible
# with the V1 transforms (those in ``torchvision.transforms``), it will
# still work with the V2 transforms without any change!
#
# We will illustrate this more completely below with a typical detection case,
# where our samples are just images, bounding boxes and labels:
class MyCustomTransform(torch.nn.Module):
def forward(self, img, bboxes, label): # we assume inputs are always structured like this
print(
f"I'm transforming an image of shape {img.shape} "
f"with bboxes = {bboxes}\n{label = }"
)
# Do some transformations. Here, we're just passing though the input
return img, bboxes, label
transforms = v2.Compose([
MyCustomTransform(),
v2.RandomResizedCrop((224, 224), antialias=True),
v2.RandomHorizontalFlip(p=1),
v2.Normalize(mean=[0, 0, 0], std=[1, 1, 1])
])
H, W = 256, 256
img = torch.rand(3, H, W)
bboxes = tv_tensors.BoundingBoxes(
torch.tensor([[0, 10, 10, 20], [50, 50, 70, 70]]),
format="XYXY",
canvas_size=(H, W)
)
label = 3
out_img, out_bboxes, out_label = transforms(img, bboxes, label)
# %%
print(f"Output image shape: {out_img.shape}\nout_bboxes = {out_bboxes}\n{out_label = }")
# %%
# .. note::
# While working with TVTensor classes in your code, make sure to
# familiarize yourself with this section:
# :ref:`tv_tensor_unwrapping_behaviour`
#
# Supporting arbitrary input structures
# =====================================
#
# In the section above, we have assumed that you already know the structure of
# your inputs and that you're OK with hard-coding this expected structure in
# your code. If you want your custom transforms to be as flexible as possible,
# this can be a bit limiting.
#
# A key feature of the builtin Torchvision V2 transforms is that they can accept
# arbitrary input structure and return the same structure as output (with
# transformed entries). For example, transforms can accept a single image, or a
# tuple of ``(img, label)``, or an arbitrary nested dictionary as input:
structured_input = {
"img": img,
"annotations": (bboxes, label),
"something_that_will_be_ignored": (1, "hello")
}
structured_output = v2.RandomHorizontalFlip(p=1)(structured_input)
assert isinstance(structured_output, dict)
assert structured_output["something_that_will_be_ignored"] == (1, "hello")
print(f"The transformed bboxes are:\n{structured_output['annotations'][0]}")
# %%
# If you want to reproduce this behavior in your own transform, we invite you to
# look at our `code
# <https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/_transform.py>`_
# and adapt it to your needs.
#
# In brief, the core logic is to unpack the input into a flat list using `pytree
# <https://github.com/pytorch/pytorch/blob/main/torch/utils/_pytree.py>`_, and
# then transform only the entries that can be transformed (the decision is made
# based on the **class** of the entries, as all TVTensors are
# tensor-subclasses) plus some custom logic that is out of score here - check the
# code for details. The (potentially transformed) entries are then repacked and
# returned, in the same structure as the input.
#
# We do not provide public dev-facing tools to achieve that at this time, but if
# this is something that would be valuable to you, please let us know by opening
# an issue on our `GitHub repo <https://github.com/pytorch/vision/issues>`_.
"""
====================================
How to write your own TVTensor class
====================================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_tv_tensors.py>` to download the full example code.
This guide is intended for advanced users and downstream library maintainers. We explain how to
write your own TVTensor class, and how to make it compatible with the built-in
Torchvision v2 transforms. Before continuing, make sure you have read
:ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.
"""
# %%
import torch
from torchvision import tv_tensors
from torchvision.transforms import v2
# %%
# We will create a very simple class that just inherits from the base
# :class:`~torchvision.tv_tensors.TVTensor` class. It will be enough to cover
# what you need to know to implement your more elaborate uses-cases. If you need
# to create a class that carries meta-data, take a look at how the
# :class:`~torchvision.tv_tensors.BoundingBoxes` class is `implemented
# <https://github.com/pytorch/vision/blob/main/torchvision/tv_tensors/_bounding_box.py>`_.
class MyTVTensor(tv_tensors.TVTensor):
pass
my_dp = MyTVTensor([1, 2, 3])
my_dp
# %%
# Now that we have defined our custom TVTensor class, we want it to be
# compatible with the built-in torchvision transforms, and the functional API.
# For that, we need to implement a kernel which performs the core of the
# transformation, and then "hook" it to the functional that we want to support
# via :func:`~torchvision.transforms.v2.functional.register_kernel`.
#
# We illustrate this process below: we create a kernel for the "horizontal flip"
# operation of our MyTVTensor class, and register it to the functional API.
from torchvision.transforms.v2 import functional as F
@F.register_kernel(functional="hflip", tv_tensor_cls=MyTVTensor)
def hflip_my_tv_tensor(my_dp, *args, **kwargs):
print("Flipping!")
out = my_dp.flip(-1)
return tv_tensors.wrap(out, like=my_dp)
# %%
# To understand why :func:`~torchvision.tv_tensors.wrap` is used, see
# :ref:`tv_tensor_unwrapping_behaviour`. Ignore the ``*args, **kwargs`` for now,
# we will explain it below in :ref:`param_forwarding`.
#
# .. note::
#
# In our call to ``register_kernel`` above we used a string
# ``functional="hflip"`` to refer to the functional we want to hook into. We
# could also have used the functional *itself*, i.e.
# ``@register_kernel(functional=F.hflip, ...)``.
#
# Now that we have registered our kernel, we can call the functional API on a
# ``MyTVTensor`` instance:
my_dp = MyTVTensor(torch.rand(3, 256, 256))
_ = F.hflip(my_dp)
# %%
# And we can also use the
# :class:`~torchvision.transforms.v2.RandomHorizontalFlip` transform, since it relies on :func:`~torchvision.transforms.v2.functional.hflip` internally:
t = v2.RandomHorizontalFlip(p=1)
_ = t(my_dp)
# %%
# .. note::
#
# We cannot register a kernel for a transform class, we can only register a
# kernel for a **functional**. The reason we can't register a transform
# class is because one transform may internally rely on more than one
# functional, so in general we can't register a single kernel for a given
# class.
#
# .. _param_forwarding:
#
# Parameter forwarding, and ensuring future compatibility of your kernels
# -----------------------------------------------------------------------
#
# The functional API that you're hooking into is public and therefore
# **backward** compatible: we guarantee that the parameters of these functionals
# won't be removed or renamed without a proper deprecation cycle. However, we
# don't guarantee **forward** compatibility, and we may add new parameters in
# the future.
#
# Imagine that in a future version, Torchvision adds a new ``inplace`` parameter
# to its :func:`~torchvision.transforms.v2.functional.hflip` functional. If you
# already defined and registered your own kernel as
def hflip_my_tv_tensor(my_dp): # noqa
print("Flipping!")
out = my_dp.flip(-1)
return tv_tensors.wrap(out, like=my_dp)
# %%
# then calling ``F.hflip(my_dp)`` will **fail**, because ``hflip`` will try to
# pass the new ``inplace`` parameter to your kernel, but your kernel doesn't
# accept it.
#
# For this reason, we recommend to always define your kernels with
# ``*args, **kwargs`` in their signature, as done above. This way, your kernel
# will be able to accept any new parameter that we may add in the future.
# (Technically, adding `**kwargs` only should be enough).
"""
===========================
How to use CutMix and MixUp
===========================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_cutmix_mixup.py>` to download the full example code.
:class:`~torchvision.transforms.v2.CutMix` and
:class:`~torchvision.transforms.v2.MixUp` are popular augmentation strategies
that can improve classification accuracy.
These transforms are slightly different from the rest of the Torchvision
transforms, because they expect
**batches** of samples as input, not individual images. In this example we'll
explain how to use them: after the ``DataLoader``, or as part of a collation
function.
"""
# %%
import torch
from torchvision.datasets import FakeData
from torchvision.transforms import v2
NUM_CLASSES = 100
# %%
# Pre-processing pipeline
# -----------------------
#
# We'll use a simple but typical image classification pipeline:
preproc = v2.Compose([
v2.PILToTensor(),
v2.RandomResizedCrop(size=(224, 224), antialias=True),
v2.RandomHorizontalFlip(p=0.5),
v2.ToDtype(torch.float32, scale=True), # to float32 in [0, 1]
v2.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)), # typically from ImageNet
])
dataset = FakeData(size=1000, num_classes=NUM_CLASSES, transform=preproc)
img, label = dataset[0]
print(f"{type(img) = }, {img.dtype = }, {img.shape = }, {label = }")
# %%
#
# One important thing to note is that neither CutMix nor MixUp are part of this
# pre-processing pipeline. We'll add them a bit later once we define the
# DataLoader. Just as a refresher, this is what the DataLoader and training loop
# would look like if we weren't using CutMix or MixUp:
from torch.utils.data import DataLoader
dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
for images, labels in dataloader:
print(f"{images.shape = }, {labels.shape = }")
print(labels.dtype)
# <rest of the training loop here>
break
# %%
# %%
# Where to use MixUp and CutMix
# -----------------------------
#
# After the DataLoader
# ^^^^^^^^^^^^^^^^^^^^
#
# Now let's add CutMix and MixUp. The simplest way to do this right after the
# DataLoader: the Dataloader has already batched the images and labels for us,
# and this is exactly what these transforms expect as input:
dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
cutmix = v2.CutMix(num_classes=NUM_CLASSES)
mixup = v2.MixUp(num_classes=NUM_CLASSES)
cutmix_or_mixup = v2.RandomChoice([cutmix, mixup])
for images, labels in dataloader:
print(f"Before CutMix/MixUp: {images.shape = }, {labels.shape = }")
images, labels = cutmix_or_mixup(images, labels)
print(f"After CutMix/MixUp: {images.shape = }, {labels.shape = }")
# <rest of the training loop here>
break
# %%
#
# Note how the labels were also transformed: we went from a batched label of
# shape (batch_size,) to a tensor of shape (batch_size, num_classes). The
# transformed labels can still be passed as-is to a loss function like
# :func:`torch.nn.functional.cross_entropy`.
#
# As part of the collation function
# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
#
# Passing the transforms after the DataLoader is the simplest way to use CutMix
# and MixUp, but one disadvantage is that it does not take advantage of the
# DataLoader multi-processing. For that, we can pass those transforms as part of
# the collation function (refer to the `PyTorch docs
# <https://pytorch.org/docs/stable/data.html#dataloader-collate-fn>`_ to learn
# more about collation).
from torch.utils.data import default_collate
def collate_fn(batch):
return cutmix_or_mixup(*default_collate(batch))
dataloader = DataLoader(dataset, batch_size=4, shuffle=True, num_workers=2, collate_fn=collate_fn)
for images, labels in dataloader:
print(f"{images.shape = }, {labels.shape = }")
# No need to call cutmix_or_mixup, it's already been called as part of the DataLoader!
# <rest of the training loop here>
break
# %%
# Non-standard input format
# -------------------------
#
# So far we've used a typical sample structure where we pass ``(images,
# labels)`` as inputs. MixUp and CutMix will magically work by default with most
# common sample structures: tuples where the second parameter is a tensor label,
# or dict with a "label[s]" key. Look at the documentation of the
# ``labels_getter`` parameter for more details.
#
# If your samples have a different structure, you can still use CutMix and MixUp
# by passing a callable to the ``labels_getter`` parameter. For example:
batch = {
"imgs": torch.rand(4, 3, 224, 224),
"target": {
"classes": torch.randint(0, NUM_CLASSES, size=(4,)),
"some_other_key": "this is going to be passed-through"
}
}
def labels_getter(batch):
return batch["target"]["classes"]
out = v2.CutMix(num_classes=NUM_CLASSES, labels_getter=labels_getter)(batch)
print(f"{out['imgs'].shape = }, {out['target']['classes'].shape = }")
"""
===============================================================
Transforms v2: End-to-end object detection/segmentation example
===============================================================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_e2e.py>` to download the full example code.
Object detection and segmentation tasks are natively supported:
``torchvision.transforms.v2`` enables jointly transforming images, videos,
bounding boxes, and masks.
This example showcases an end-to-end instance segmentation training case using
Torchvision utils from ``torchvision.datasets``, ``torchvision.models`` and
``torchvision.transforms.v2``. Everything covered here can be applied similarly
to object detection or semantic segmentation tasks.
"""
# %%
import pathlib
import torch
import torch.utils.data
from torchvision import models, datasets, tv_tensors
from torchvision.transforms import v2
torch.manual_seed(0)
# This loads fake data for illustration purposes of this example. In practice, you'll have
# to replace this with the proper data.
# If you're trying to run that on collab, you can download the assets and the
# helpers from https://github.com/pytorch/vision/tree/main/gallery/
ROOT = pathlib.Path("../assets") / "coco"
IMAGES_PATH = str(ROOT / "images")
ANNOTATIONS_PATH = str(ROOT / "instances.json")
from helpers import plot
# %%
# Dataset preparation
# -------------------
#
# We start off by loading the :class:`~torchvision.datasets.CocoDetection` dataset to have a look at what it currently
# returns.
dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH)
sample = dataset[0]
img, target = sample
print(f"{type(img) = }\n{type(target) = }\n{type(target[0]) = }\n{target[0].keys() = }")
# %%
# Torchvision datasets preserve the data structure and types as it was intended
# by the datasets authors. So by default, the output structure may not always be
# compatible with the models or the transforms.
#
# To overcome that, we can use the
# :func:`~torchvision.datasets.wrap_dataset_for_transforms_v2` function. For
# :class:`~torchvision.datasets.CocoDetection`, this changes the target
# structure to a single dictionary of lists:
dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=("boxes", "labels", "masks"))
sample = dataset[0]
img, target = sample
print(f"{type(img) = }\n{type(target) = }\n{target.keys() = }")
print(f"{type(target['boxes']) = }\n{type(target['labels']) = }\n{type(target['masks']) = }")
# %%
# We used the ``target_keys`` parameter to specify the kind of output we're
# interested in. Our dataset now returns a target which is dict where the values
# are :ref:`TVTensors <what_are_tv_tensors>` (all are :class:`torch.Tensor`
# subclasses). We're dropped all unncessary keys from the previous output, but
# if you need any of the original keys e.g. "image_id", you can still ask for
# it.
#
# .. note::
#
# If you just want to do detection, you don't need and shouldn't pass
# "masks" in ``target_keys``: if masks are present in the sample, they will
# be transformed, slowing down your transformations unnecessarily.
#
# As baseline, let's have a look at a sample without transformations:
plot([dataset[0], dataset[1]])
# %%
# Transforms
# ----------
#
# Let's now define our pre-processing transforms. All the transforms know how
# to handle images, bounding boxes and masks when relevant.
#
# Transforms are typically passed as the ``transforms`` parameter of the
# dataset so that they can leverage multi-processing from the
# :class:`torch.utils.data.DataLoader`.
transforms = v2.Compose(
[
v2.ToImage(),
v2.RandomPhotometricDistort(p=1),
v2.RandomZoomOut(fill={tv_tensors.Image: (123, 117, 104), "others": 0}),
v2.RandomIoUCrop(),
v2.RandomHorizontalFlip(p=1),
v2.SanitizeBoundingBoxes(),
v2.ToDtype(torch.float32, scale=True),
]
)
dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH, transforms=transforms)
dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=["boxes", "labels", "masks"])
# %%
# A few things are worth noting here:
#
# - We're converting the PIL image into a
# :class:`~torchvision.transforms.v2.Image` object. This isn't strictly
# necessary, but relying on Tensors (here: a Tensor subclass) will
# :ref:`generally be faster <transforms_perf>`.
# - We are calling :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` to
# make sure we remove degenerate bounding boxes, as well as their
# corresponding labels and masks.
# :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` should be placed
# at least once at the end of a detection pipeline; it is particularly
# critical if :class:`~torchvision.transforms.v2.RandomIoUCrop` was used.
#
# Let's look how the sample looks like with our augmentation pipeline in place:
# sphinx_gallery_thumbnail_number = 2
plot([dataset[0], dataset[1]])
# %%
# We can see that the color of the images were distorted, zoomed in or out, and flipped.
# The bounding boxes and the masks were transformed accordingly. And without any further ado, we can start training.
#
# Data loading and training loop
# ------------------------------
#
# Below we're using Mask-RCNN which is an instance segmentation model, but
# everything we've covered in this tutorial also applies to object detection and
# semantic segmentation tasks.
data_loader = torch.utils.data.DataLoader(
dataset,
batch_size=2,
# We need a custom collation function here, since the object detection
# models expect a sequence of images and target dictionaries. The default
# collation function tries to torch.stack() the individual elements,
# which fails in general for object detection, because the number of bounding
# boxes varies between the images of the same batch.
collate_fn=lambda batch: tuple(zip(*batch)),
)
model = models.get_model("maskrcnn_resnet50_fpn_v2", weights=None, weights_backbone=None).train()
for imgs, targets in data_loader:
loss_dict = model(imgs, targets)
# Put your training logic here
print(f"{[img.shape for img in imgs] = }")
print(f"{[type(target) for target in targets] = }")
for name, loss_val in loss_dict.items():
print(f"{name:<20}{loss_val:.3f}")
# %%
# Training References
# -------------------
#
# From there, you can check out the `torchvision references
# <https://github.com/pytorch/vision/tree/main/references>`_ where you'll find
# the actual training scripts we use to train our models.
#
# **Disclaimer** The code in our references is more complex than what you'll
# need for your own use-cases: this is because we're supporting different
# backends (PIL, tensors, TVTensors) and different transforms namespaces (v1 and
# v2). So don't be afraid to simplify and only keep what you need.
"""
==================================
Getting started with transforms v2
==================================
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_getting_started.py>` to download the full example code.
This example illustrates all of what you need to know to get started with the
new :mod:`torchvision.transforms.v2` API. We'll cover simple tasks like
image classification, and more advanced ones like object detection /
segmentation.
"""
# %%
# First, a bit of setup
from pathlib import Path
import torch
import matplotlib.pyplot as plt
plt.rcParams["savefig.bbox"] = 'tight'
from torchvision.transforms import v2
from torchvision.io import read_image
torch.manual_seed(1)
# If you're trying to run that on collab, you can download the assets and the
# helpers from https://github.com/pytorch/vision/tree/main/gallery/
from helpers import plot
img = read_image(str(Path('../assets') / 'astronaut.jpg'))
print(f"{type(img) = }, {img.dtype = }, {img.shape = }")
# %%
# The basics
# ----------
#
# The Torchvision transforms behave like a regular :class:`torch.nn.Module` (in
# fact, most of them are): instantiate a transform, pass an input, get a
# transformed output:
transform = v2.RandomCrop(size=(224, 224))
out = transform(img)
plot([img, out])
# %%
# I just want to do image classification
# --------------------------------------
#
# If you just care about image classification, things are very simple. A basic
# classification pipeline may look like this:
transforms = v2.Compose([
v2.RandomResizedCrop(size=(224, 224), antialias=True),
v2.RandomHorizontalFlip(p=0.5),
v2.ToDtype(torch.float32, scale=True),
v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])
out = transforms(img)
plot([img, out])
# %%
# Such transformation pipeline is typically passed as the ``transform`` argument
# to the :ref:`Datasets <datasets>`, e.g. ``ImageNet(...,
# transform=transforms)``.
#
# That's pretty much all there is. From there, read through our :ref:`main docs
# <transforms>` to learn more about recommended practices and conventions, or
# explore more :ref:`examples <transforms_gallery>` e.g. how to use augmentation
# transforms like :ref:`CutMix and MixUp
# <sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py>`.
#
# .. note::
#
# If you're already relying on the ``torchvision.transforms`` v1 API,
# we recommend to :ref:`switch to the new v2 transforms<v1_or_v2>`. It's
# very easy: the v2 transforms are fully compatible with the v1 API, so you
# only need to change the import!
#
# Detection, Segmentation, Videos
# -------------------------------
#
# The new Torchvision transforms in the ``torchvision.transforms.v2`` namespace
# support tasks beyond image classification: they can also transform bounding
# boxes, segmentation / detection masks, or videos.
#
# Let's briefly look at a detection example with bounding boxes.
from torchvision import tv_tensors # we'll describe this a bit later, bare with us
boxes = tv_tensors.BoundingBoxes(
[
[15, 10, 370, 510],
[275, 340, 510, 510],
[130, 345, 210, 425]
],
format="XYXY", canvas_size=img.shape[-2:])
transforms = v2.Compose([
v2.RandomResizedCrop(size=(224, 224), antialias=True),
v2.RandomPhotometricDistort(p=1),
v2.RandomHorizontalFlip(p=1),
])
out_img, out_boxes = transforms(img, boxes)
print(type(boxes), type(out_boxes))
plot([(img, boxes), (out_img, out_boxes)])
# %%
#
# The example above focuses on object detection. But if we had masks
# (:class:`torchvision.tv_tensors.Mask`) for object segmentation or semantic
# segmentation, or videos (:class:`torchvision.tv_tensors.Video`), we could have
# passed them to the transforms in exactly the same way.
#
# By now you likely have a few questions: what are these TVTensors, how do we
# use them, and what is the expected input/output of those transforms? We'll
# answer these in the next sections.
# %%
#
# .. _what_are_tv_tensors:
#
# What are TVTensors?
# --------------------
#
# TVTensors are :class:`torch.Tensor` subclasses. The available TVTensors are
# :class:`~torchvision.tv_tensors.Image`,
# :class:`~torchvision.tv_tensors.BoundingBoxes`,
# :class:`~torchvision.tv_tensors.Mask`, and
# :class:`~torchvision.tv_tensors.Video`.
#
# TVTensors look and feel just like regular tensors - they **are** tensors.
# Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()``
# or any ``torch.*`` operator will also work on a TVTensor:
img_dp = tv_tensors.Image(torch.randint(0, 256, (3, 256, 256), dtype=torch.uint8))
print(f"{isinstance(img_dp, torch.Tensor) = }")
print(f"{img_dp.dtype = }, {img_dp.shape = }, {img_dp.sum() = }")
# %%
# These TVTensor classes are at the core of the transforms: in order to
# transform a given input, the transforms first look at the **class** of the
# object, and dispatch to the appropriate implementation accordingly.
#
# You don't need to know much more about TVTensors at this point, but advanced
# users who want to learn more can refer to
# :ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.
#
# What do I pass as input?
# ------------------------
#
# Above, we've seen two examples: one where we passed a single image as input
# i.e. ``out = transforms(img)``, and one where we passed both an image and
# bounding boxes, i.e. ``out_img, out_boxes = transforms(img, boxes)``.
#
# In fact, transforms support **arbitrary input structures**. The input can be a
# single image, a tuple, an arbitrarily nested dictionary... pretty much
# anything. The same structure will be returned as output. Below, we use the
# same detection transforms, but pass a tuple (image, target_dict) as input and
# we're getting the same structure as output:
target = {
"boxes": boxes,
"labels": torch.arange(boxes.shape[0]),
"this_is_ignored": ("arbitrary", {"structure": "!"})
}
# Re-using the transforms and definitions from above.
out_img, out_target = transforms(img, target)
# sphinx_gallery_thumbnail_number = 4
plot([(img, target["boxes"]), (out_img, out_target["boxes"])])
print(f"{out_target['this_is_ignored']}")
# %%
# We passed a tuple so we get a tuple back, and the second element is the
# tranformed target dict. Transforms don't really care about the structure of
# the input; as mentioned above, they only care about the **type** of the
# objects and transforms them accordingly.
#
# *Foreign* objects like strings or ints are simply passed-through. This can be
# useful e.g. if you want to associate a path with every single sample when
# debugging!
#
# .. _passthrough_heuristic:
#
# .. note::
#
# **Disclaimer** This note is slightly advanced and can be safely skipped on
# a first read.
#
# Pure :class:`torch.Tensor` objects are, in general, treated as images (or
# as videos for video-specific transforms). Indeed, you may have noticed
# that in the code above we haven't used the
# :class:`~torchvision.tv_tensors.Image` class at all, and yet our images
# got transformed properly. Transforms follow the following logic to
# determine whether a pure Tensor should be treated as an image (or video),
# or just ignored:
#
# * If there is an :class:`~torchvision.tv_tensors.Image`,
# :class:`~torchvision.tv_tensors.Video`,
# or :class:`PIL.Image.Image` instance in the input, all other pure
# tensors are passed-through.
# * If there is no :class:`~torchvision.tv_tensors.Image` or
# :class:`~torchvision.tv_tensors.Video` instance, only the first pure
# :class:`torch.Tensor` will be transformed as image or video, while all
# others will be passed-through. Here "first" means "first in a depth-wise
# traversal".
#
# This is what happened in the detection example above: the first pure
# tensor was the image so it got transformed properly, and all other pure
# tensor instances like the ``labels`` were passed-through (although labels
# can still be transformed by some transforms like
# :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes`!).
#
# .. _transforms_datasets_intercompatibility:
#
# Transforms and Datasets intercompatibility
# ------------------------------------------
#
# Roughly speaking, the output of the datasets must correspond to the input of
# the transforms. How to do that depends on whether you're using the torchvision
# :ref:`built-in datatsets <datasets>`, or your own custom datasets.
#
# Using built-in datasets
# ^^^^^^^^^^^^^^^^^^^^^^^
#
# If you're just doing image classification, you don't need to do anything. Just
# use ``transform`` argument of the dataset e.g. ``ImageNet(...,
# transform=transforms)`` and you're good to go.
#
# Torchvision also supports datasets for object detection or segmentation like
# :class:`torchvision.datasets.CocoDetection`. Those datasets predate
# the existence of the :mod:`torchvision.transforms.v2` module and of the
# TVTensors, so they don't return TVTensors out of the box.
#
# An easy way to force those datasets to return TVTensors and to make them
# compatible with v2 transforms is to use the
# :func:`torchvision.datasets.wrap_dataset_for_transforms_v2` function:
#
# .. code-block:: python
#
# from torchvision.datasets import CocoDetection, wrap_dataset_for_transforms_v2
#
# dataset = CocoDetection(..., transforms=my_transforms)
# dataset = wrap_dataset_for_transforms_v2(dataset)
# # Now the dataset returns TVTensors!
#
# Using your own datasets
# ^^^^^^^^^^^^^^^^^^^^^^^
#
# If you have a custom dataset, then you'll need to convert your objects into
# the appropriate TVTensor classes. Creating TVTensor instances is very easy,
# refer to :ref:`tv_tensor_creation` for more details.
#
# There are two main places where you can implement that conversion logic:
#
# - At the end of the datasets's ``__getitem__`` method, before returning the
# sample (or by sub-classing the dataset).
# - As the very first step of your transforms pipeline
#
# Either way, the logic will depend on your specific dataset.
......@@ -3,280 +3,329 @@
Illustration of transforms
==========================
This example illustrates the various transforms available in :ref:`the
torchvision.transforms module <transforms>`.
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_illustrations.py>` to download the full example code.
This example illustrates some of the various transforms available in :ref:`the
torchvision.transforms.v2 module <transforms>`.
"""
# %%
# sphinx_gallery_thumbnail_path = "../../gallery/assets/transforms_thumbnail.png"
from PIL import Image
from pathlib import Path
import matplotlib.pyplot as plt
import numpy as np
import torch
import torchvision.transforms as T
from torchvision.transforms import v2
plt.rcParams["savefig.bbox"] = 'tight'
orig_img = Image.open(Path('assets') / 'astronaut.jpg')
# if you change the seed, make sure that the randomly-applied transforms
# properly show that the image can be both transformed and *not* transformed!
torch.manual_seed(0)
def plot(imgs, with_orig=True, row_title=None, **imshow_kwargs):
if not isinstance(imgs[0], list):
# Make a 2d grid even if there's just 1 row
imgs = [imgs]
num_rows = len(imgs)
num_cols = len(imgs[0]) + with_orig
fig, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)
for row_idx, row in enumerate(imgs):
row = [orig_img] + row if with_orig else row
for col_idx, img in enumerate(row):
ax = axs[row_idx, col_idx]
ax.imshow(np.asarray(img), **imshow_kwargs)
ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
if with_orig:
axs[0, 0].set(title='Original image')
axs[0, 0].title.set_size(8)
if row_title is not None:
for row_idx in range(num_rows):
axs[row_idx, 0].set(ylabel=row_title[row_idx])
plt.tight_layout()
####################################
# If you're trying to run that on collab, you can download the assets and the
# helpers from https://github.com/pytorch/vision/tree/main/gallery/
from helpers import plot
orig_img = Image.open(Path('../assets') / 'astronaut.jpg')
# %%
# Geometric Transforms
# --------------------
# Geometric image transformation refers to the process of altering the geometric properties of an image,
# such as its shape, size, orientation, or position.
# It involves applying mathematical operations to the image pixels or coordinates to achieve the desired transformation.
#
# Pad
# ---
# ~~~
# The :class:`~torchvision.transforms.Pad` transform
# (see also :func:`~torchvision.transforms.functional.pad`)
# fills image borders with some pixel values.
padded_imgs = [T.Pad(padding=padding)(orig_img) for padding in (3, 10, 30, 50)]
plot(padded_imgs)
# pads all image borders with some pixel values.
padded_imgs = [v2.Pad(padding=padding)(orig_img) for padding in (3, 10, 30, 50)]
plot([orig_img] + padded_imgs)
####################################
# %%
# Resize
# ------
# ~~~~~~
# The :class:`~torchvision.transforms.Resize` transform
# (see also :func:`~torchvision.transforms.functional.resize`)
# resizes an image.
resized_imgs = [T.Resize(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
plot(resized_imgs)
resized_imgs = [v2.Resize(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
plot([orig_img] + resized_imgs)
####################################
# %%
# CenterCrop
# ----------
# ~~~~~~~~~~
# The :class:`~torchvision.transforms.CenterCrop` transform
# (see also :func:`~torchvision.transforms.functional.center_crop`)
# crops the given image at the center.
center_crops = [T.CenterCrop(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
plot(center_crops)
center_crops = [v2.CenterCrop(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
plot([orig_img] + center_crops)
####################################
# %%
# FiveCrop
# --------
# ~~~~~~~~
# The :class:`~torchvision.transforms.FiveCrop` transform
# (see also :func:`~torchvision.transforms.functional.five_crop`)
# crops the given image into four corners and the central crop.
(top_left, top_right, bottom_left, bottom_right, center) = T.FiveCrop(size=(100, 100))(orig_img)
plot([top_left, top_right, bottom_left, bottom_right, center])
####################################
# Grayscale
# ---------
# The :class:`~torchvision.transforms.Grayscale` transform
# (see also :func:`~torchvision.transforms.functional.to_grayscale`)
# converts an image to grayscale
gray_img = T.Grayscale()(orig_img)
plot([gray_img], cmap='gray')
####################################
# Random transforms
# -----------------
# The following transforms are random, which means that the same transfomer
# instance will produce different result each time it transforms a given image.
#
# ColorJitter
# ~~~~~~~~~~~
# The :class:`~torchvision.transforms.ColorJitter` transform
# randomly changes the brightness, saturation, and other properties of an image.
jitter = T.ColorJitter(brightness=.5, hue=.3)
jitted_imgs = [jitter(orig_img) for _ in range(4)]
plot(jitted_imgs)
####################################
# GaussianBlur
# ~~~~~~~~~~~~
# The :class:`~torchvision.transforms.GaussianBlur` transform
# (see also :func:`~torchvision.transforms.functional.gaussian_blur`)
# performs gaussian blur transform on an image.
blurrer = T.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5))
blurred_imgs = [blurrer(orig_img) for _ in range(4)]
plot(blurred_imgs)
(top_left, top_right, bottom_left, bottom_right, center) = v2.FiveCrop(size=(100, 100))(orig_img)
plot([orig_img] + [top_left, top_right, bottom_left, bottom_right, center])
####################################
# %%
# RandomPerspective
# ~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomPerspective` transform
# (see also :func:`~torchvision.transforms.functional.perspective`)
# performs random perspective transform on an image.
perspective_transformer = T.RandomPerspective(distortion_scale=0.6, p=1.0)
perspective_transformer = v2.RandomPerspective(distortion_scale=0.6, p=1.0)
perspective_imgs = [perspective_transformer(orig_img) for _ in range(4)]
plot(perspective_imgs)
plot([orig_img] + perspective_imgs)
####################################
# %%
# RandomRotation
# ~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomRotation` transform
# (see also :func:`~torchvision.transforms.functional.rotate`)
# rotates an image with random angle.
rotater = T.RandomRotation(degrees=(0, 180))
rotater = v2.RandomRotation(degrees=(0, 180))
rotated_imgs = [rotater(orig_img) for _ in range(4)]
plot(rotated_imgs)
plot([orig_img] + rotated_imgs)
####################################
# %%
# RandomAffine
# ~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomAffine` transform
# (see also :func:`~torchvision.transforms.functional.affine`)
# performs random affine transform on an image.
affine_transfomer = T.RandomAffine(degrees=(30, 70), translate=(0.1, 0.3), scale=(0.5, 0.75))
affine_transfomer = v2.RandomAffine(degrees=(30, 70), translate=(0.1, 0.3), scale=(0.5, 0.75))
affine_imgs = [affine_transfomer(orig_img) for _ in range(4)]
plot(affine_imgs)
####################################
plot([orig_img] + affine_imgs)
# %%
# ElasticTransform
# ~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.ElasticTransform` transform
# (see also :func:`~torchvision.transforms.functional.elastic_transform`)
# Randomly transforms the morphology of objects in images and produces a
# see-through-water-like effect.
elastic_transformer = v2.ElasticTransform(alpha=250.0)
transformed_imgs = [elastic_transformer(orig_img) for _ in range(2)]
plot([orig_img] + transformed_imgs)
# %%
# RandomCrop
# ~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomCrop` transform
# (see also :func:`~torchvision.transforms.functional.crop`)
# crops an image at a random location.
cropper = T.RandomCrop(size=(128, 128))
cropper = v2.RandomCrop(size=(128, 128))
crops = [cropper(orig_img) for _ in range(4)]
plot(crops)
plot([orig_img] + crops)
####################################
# %%
# RandomResizedCrop
# ~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomResizedCrop` transform
# (see also :func:`~torchvision.transforms.functional.resized_crop`)
# crops an image at a random location, and then resizes the crop to a given
# size.
resize_cropper = T.RandomResizedCrop(size=(32, 32))
resize_cropper = v2.RandomResizedCrop(size=(32, 32))
resized_crops = [resize_cropper(orig_img) for _ in range(4)]
plot(resized_crops)
plot([orig_img] + resized_crops)
# %%
# Photometric Transforms
# ----------------------
# Photometric image transformation refers to the process of modifying the photometric properties of an image,
# such as its brightness, contrast, color, or tone.
# These transformations are applied to change the visual appearance of an image
# while preserving its geometric structure.
#
# Except :class:`~torchvision.transforms.Grayscale`, the following transforms are random,
# which means that the same transform
# instance will produce different result each time it transforms a given image.
#
# Grayscale
# ~~~~~~~~~
# The :class:`~torchvision.transforms.Grayscale` transform
# (see also :func:`~torchvision.transforms.functional.to_grayscale`)
# converts an image to grayscale
gray_img = v2.Grayscale()(orig_img)
plot([orig_img, gray_img], cmap='gray')
####################################
# %%
# ColorJitter
# ~~~~~~~~~~~
# The :class:`~torchvision.transforms.ColorJitter` transform
# randomly changes the brightness, contrast, saturation, hue, and other properties of an image.
jitter = v2.ColorJitter(brightness=.5, hue=.3)
jittered_imgs = [jitter(orig_img) for _ in range(4)]
plot([orig_img] + jittered_imgs)
# %%
# GaussianBlur
# ~~~~~~~~~~~~
# The :class:`~torchvision.transforms.GaussianBlur` transform
# (see also :func:`~torchvision.transforms.functional.gaussian_blur`)
# performs gaussian blur transform on an image.
blurrer = v2.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5.))
blurred_imgs = [blurrer(orig_img) for _ in range(4)]
plot([orig_img] + blurred_imgs)
# %%
# RandomInvert
# ~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomInvert` transform
# (see also :func:`~torchvision.transforms.functional.invert`)
# randomly inverts the colors of the given image.
inverter = T.RandomInvert()
inverter = v2.RandomInvert()
invertered_imgs = [inverter(orig_img) for _ in range(4)]
plot(invertered_imgs)
plot([orig_img] + invertered_imgs)
####################################
# %%
# RandomPosterize
# ~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomPosterize` transform
# (see also :func:`~torchvision.transforms.functional.posterize`)
# randomly posterizes the image by reducing the number of bits
# of each color channel.
posterizer = T.RandomPosterize(bits=2)
posterizer = v2.RandomPosterize(bits=2)
posterized_imgs = [posterizer(orig_img) for _ in range(4)]
plot(posterized_imgs)
plot([orig_img] + posterized_imgs)
####################################
# %%
# RandomSolarize
# ~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomSolarize` transform
# (see also :func:`~torchvision.transforms.functional.solarize`)
# randomly solarizes the image by inverting all pixel values above
# the threshold.
solarizer = T.RandomSolarize(threshold=192.0)
solarizer = v2.RandomSolarize(threshold=192.0)
solarized_imgs = [solarizer(orig_img) for _ in range(4)]
plot(solarized_imgs)
plot([orig_img] + solarized_imgs)
####################################
# %%
# RandomAdjustSharpness
# ~~~~~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomAdjustSharpness` transform
# (see also :func:`~torchvision.transforms.functional.adjust_sharpness`)
# randomly adjusts the sharpness of the given image.
sharpness_adjuster = T.RandomAdjustSharpness(sharpness_factor=2)
sharpness_adjuster = v2.RandomAdjustSharpness(sharpness_factor=2)
sharpened_imgs = [sharpness_adjuster(orig_img) for _ in range(4)]
plot(sharpened_imgs)
plot([orig_img] + sharpened_imgs)
####################################
# %%
# RandomAutocontrast
# ~~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomAutocontrast` transform
# (see also :func:`~torchvision.transforms.functional.autocontrast`)
# randomly applies autocontrast to the given image.
autocontraster = T.RandomAutocontrast()
autocontraster = v2.RandomAutocontrast()
autocontrasted_imgs = [autocontraster(orig_img) for _ in range(4)]
plot(autocontrasted_imgs)
plot([orig_img] + autocontrasted_imgs)
####################################
# %%
# RandomEqualize
# ~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomEqualize` transform
# (see also :func:`~torchvision.transforms.functional.equalize`)
# randomly equalizes the histogram of the given image.
equalizer = T.RandomEqualize()
equalizer = v2.RandomEqualize()
equalized_imgs = [equalizer(orig_img) for _ in range(4)]
plot(equalized_imgs)
plot([orig_img] + equalized_imgs)
####################################
# %%
# JPEG
# ~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.v2.JPEG` transform
# (see also :func:`~torchvision.transforms.v2.functional.jpeg`)
# applies JPEG compression to the given image with random
# degree of compression.
jpeg = v2.JPEG((5, 50))
jpeg_imgs = [jpeg(orig_img) for _ in range(4)]
plot([orig_img] + jpeg_imgs)
# %%
# Augmentation Transforms
# -----------------------
# The following transforms are combinations of multiple transforms,
# either geometric or photometric, or both.
#
# AutoAugment
# ~~~~~~~~~~~
# The :class:`~torchvision.transforms.AutoAugment` transform
# automatically augments data based on a given auto-augmentation policy.
# See :class:`~torchvision.transforms.AutoAugmentPolicy` for the available policies.
policies = [T.AutoAugmentPolicy.CIFAR10, T.AutoAugmentPolicy.IMAGENET, T.AutoAugmentPolicy.SVHN]
augmenters = [T.AutoAugment(policy) for policy in policies]
policies = [v2.AutoAugmentPolicy.CIFAR10, v2.AutoAugmentPolicy.IMAGENET, v2.AutoAugmentPolicy.SVHN]
augmenters = [v2.AutoAugment(policy) for policy in policies]
imgs = [
[augmenter(orig_img) for _ in range(4)]
for augmenter in augmenters
]
row_title = [str(policy).split('.')[-1] for policy in policies]
plot(imgs, row_title=row_title)
plot([[orig_img] + row for row in imgs], row_title=row_title)
####################################
# Randomly-applied transforms
# %%
# RandAugment
# ~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandAugment` is an alternate version of AutoAugment.
augmenter = v2.RandAugment()
imgs = [augmenter(orig_img) for _ in range(4)]
plot([orig_img] + imgs)
# %%
# TrivialAugmentWide
# ~~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.TrivialAugmentWide` is an alternate implementation of AutoAugment.
# However, instead of transforming an image multiple times, it transforms an image only once
# using a random transform from a given list with a random strength number.
augmenter = v2.TrivialAugmentWide()
imgs = [augmenter(orig_img) for _ in range(4)]
plot([orig_img] + imgs)
# %%
# AugMix
# ~~~~~~
# The :class:`~torchvision.transforms.AugMix` transform interpolates between augmented versions of an image.
augmenter = v2.AugMix()
imgs = [augmenter(orig_img) for _ in range(4)]
plot([orig_img] + imgs)
# %%
# Randomly-applied Transforms
# ---------------------------
#
# Some transforms are randomly-applied given a probability ``p``. That is, the
# transformed image may actually be the same as the original one, even when
# called with the same transformer instance!
# The following transforms are randomly-applied given a probability ``p``. That is, given ``p = 0.5``,
# there is a 50% chance to return the original image, and a 50% chance to return the transformed image,
# even when called with the same transform instance!
#
# RandomHorizontalFlip
# ~~~~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomHorizontalFlip` transform
# (see also :func:`~torchvision.transforms.functional.hflip`)
# performs horizontal flip of an image, with a given probability.
hflipper = T.RandomHorizontalFlip(p=0.5)
hflipper = v2.RandomHorizontalFlip(p=0.5)
transformed_imgs = [hflipper(orig_img) for _ in range(4)]
plot(transformed_imgs)
plot([orig_img] + transformed_imgs)
####################################
# %%
# RandomVerticalFlip
# ~~~~~~~~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomVerticalFlip` transform
# (see also :func:`~torchvision.transforms.functional.vflip`)
# performs vertical flip of an image, with a given probability.
vflipper = T.RandomVerticalFlip(p=0.5)
vflipper = v2.RandomVerticalFlip(p=0.5)
transformed_imgs = [vflipper(orig_img) for _ in range(4)]
plot(transformed_imgs)
plot([orig_img] + transformed_imgs)
####################################
# %%
# RandomApply
# ~~~~~~~~~~~
# The :class:`~torchvision.transforms.RandomApply` transform
# randomly applies a list of transforms, with a given probability.
applier = T.RandomApply(transforms=[T.RandomCrop(size=(64, 64))], p=0.5)
applier = v2.RandomApply(transforms=[v2.RandomCrop(size=(64, 64))], p=0.5)
transformed_imgs = [applier(orig_img) for _ in range(4)]
plot(transformed_imgs)
plot([orig_img] + transformed_imgs)
"""
=============
TVTensors FAQ
=============
.. note::
Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_tv_tensors.ipynb>`_
or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_tv_tensors.py>` to download the full example code.
TVTensors are Tensor subclasses introduced together with
``torchvision.transforms.v2``. This example showcases what these TVTensors are
and how they behave.
.. warning::
**Intended Audience** Unless you're writing your own transforms or your own TVTensors, you
probably do not need to read this guide. This is a fairly low-level topic
that most users will not need to worry about: you do not need to understand
the internals of TVTensors to efficiently rely on
``torchvision.transforms.v2``. It may however be useful for advanced users
trying to implement their own datasets, transforms, or work directly with
the TVTensors.
"""
# %%
import PIL.Image
import torch
from torchvision import tv_tensors
# %%
# What are TVTensors?
# -------------------
#
# TVTensors are zero-copy tensor subclasses:
tensor = torch.rand(3, 256, 256)
image = tv_tensors.Image(tensor)
assert isinstance(image, torch.Tensor)
assert image.data_ptr() == tensor.data_ptr()
# %%
# Under the hood, they are needed in :mod:`torchvision.transforms.v2` to correctly dispatch to the appropriate function
# for the input data.
#
# :mod:`torchvision.tv_tensors` supports four types of TVTensors:
#
# * :class:`~torchvision.tv_tensors.Image`
# * :class:`~torchvision.tv_tensors.Video`
# * :class:`~torchvision.tv_tensors.BoundingBoxes`
# * :class:`~torchvision.tv_tensors.Mask`
#
# What can I do with a TVTensor?
# ------------------------------
#
# TVTensors look and feel just like regular tensors - they **are** tensors.
# Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()`` or
# any ``torch.*`` operator will also work on TVTensors. See
# :ref:`tv_tensor_unwrapping_behaviour` for a few gotchas.
# %%
# .. _tv_tensor_creation:
#
# How do I construct a TVTensor?
# ------------------------------
#
# Using the constructor
# ^^^^^^^^^^^^^^^^^^^^^
#
# Each TVTensor class takes any tensor-like data that can be turned into a :class:`~torch.Tensor`
image = tv_tensors.Image([[[[0, 1], [1, 0]]]])
print(image)
# %%
# Similar to other PyTorch creations ops, the constructor also takes the ``dtype``, ``device``, and ``requires_grad``
# parameters.
float_image = tv_tensors.Image([[[0, 1], [1, 0]]], dtype=torch.float32, requires_grad=True)
print(float_image)
# %%
# In addition, :class:`~torchvision.tv_tensors.Image` and :class:`~torchvision.tv_tensors.Mask` can also take a
# :class:`PIL.Image.Image` directly:
image = tv_tensors.Image(PIL.Image.open("../assets/astronaut.jpg"))
print(image.shape, image.dtype)
# %%
# Some TVTensors require additional metadata to be passed in ordered to be constructed. For example,
# :class:`~torchvision.tv_tensors.BoundingBoxes` requires the coordinate format as well as the size of the
# corresponding image (``canvas_size``) alongside the actual values. These
# metadata are required to properly transform the bounding boxes.
bboxes = tv_tensors.BoundingBoxes(
[[17, 16, 344, 495], [0, 10, 0, 10]],
format=tv_tensors.BoundingBoxFormat.XYXY,
canvas_size=image.shape[-2:]
)
print(bboxes)
# %%
# Using ``tv_tensors.wrap()``
# ^^^^^^^^^^^^^^^^^^^^^^^^^^^
#
# You can also use the :func:`~torchvision.tv_tensors.wrap` function to wrap a tensor object
# into a TVTensor. This is useful when you already have an object of the
# desired type, which typically happens when writing transforms: you just want
# to wrap the output like the input.
new_bboxes = torch.tensor([0, 20, 30, 40])
new_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)
assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
assert new_bboxes.canvas_size == bboxes.canvas_size
# %%
# The metadata of ``new_bboxes`` is the same as ``bboxes``, but you could pass
# it as a parameter to override it.
#
# .. _tv_tensor_unwrapping_behaviour:
#
# I had a TVTensor but now I have a Tensor. Help!
# -----------------------------------------------
#
# By default, operations on :class:`~torchvision.tv_tensors.TVTensor` objects
# will return a pure Tensor:
assert isinstance(bboxes, tv_tensors.BoundingBoxes)
# Shift bboxes by 3 pixels in both H and W
new_bboxes = bboxes + 3
assert isinstance(new_bboxes, torch.Tensor)
assert not isinstance(new_bboxes, tv_tensors.BoundingBoxes)
# %%
# .. note::
#
# This behavior only affects native ``torch`` operations. If you are using
# the built-in ``torchvision`` transforms or functionals, you will always get
# as output the same type that you passed as input (pure ``Tensor`` or
# ``TVTensor``).
# %%
# But I want a TVTensor back!
# ^^^^^^^^^^^^^^^^^^^^^^^^^^^
#
# You can re-wrap a pure tensor into a TVTensor by just calling the TVTensor
# constructor, or by using the :func:`~torchvision.tv_tensors.wrap` function
# (see more details above in :ref:`tv_tensor_creation`):
new_bboxes = bboxes + 3
new_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)
assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
# %%
# Alternatively, you can use the :func:`~torchvision.tv_tensors.set_return_type`
# as a global config setting for the whole program, or as a context manager
# (read its docs to learn more about caveats):
with tv_tensors.set_return_type("TVTensor"):
new_bboxes = bboxes + 3
assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
# %%
# Why is this happening?
# ^^^^^^^^^^^^^^^^^^^^^^
#
# **For performance reasons**. :class:`~torchvision.tv_tensors.TVTensor`
# classes are Tensor subclasses, so any operation involving a
# :class:`~torchvision.tv_tensors.TVTensor` object will go through the
# `__torch_function__
# <https://pytorch.org/docs/stable/notes/extending.html#extending-torch>`_
# protocol. This induces a small overhead, which we want to avoid when possible.
# This doesn't matter for built-in ``torchvision`` transforms because we can
# avoid the overhead there, but it could be a problem in your model's
# ``forward``.
#
# **The alternative isn't much better anyway.** For every operation where
# preserving the :class:`~torchvision.tv_tensors.TVTensor` type makes
# sense, there are just as many operations where returning a pure Tensor is
# preferable: for example, is ``img.sum()`` still an :class:`~torchvision.tv_tensors.Image`?
# If we were to preserve :class:`~torchvision.tv_tensors.TVTensor` types all
# the way, even model's logits or the output of the loss function would end up
# being of type :class:`~torchvision.tv_tensors.Image`, and surely that's not
# desirable.
#
# .. note::
#
# This behaviour is something we're actively seeking feedback on. If you find this surprising or if you
# have any suggestions on how to better support your use-cases, please reach out to us via this issue:
# https://github.com/pytorch/vision/issues/7319
#
# Exceptions
# ^^^^^^^^^^
#
# There are a few exceptions to this "unwrapping" rule:
# :meth:`~torch.Tensor.clone`, :meth:`~torch.Tensor.to`,
# :meth:`torch.Tensor.detach`, and :meth:`~torch.Tensor.requires_grad_` retain
# the TVTensor type.
#
# Inplace operations on TVTensors like ``obj.add_()`` will preserve the type of
# ``obj``. However, the **returned** value of inplace operations will be a pure
# tensor:
image = tv_tensors.Image([[[0, 1], [1, 0]]])
new_image = image.add_(1).mul_(2)
# image got transformed in-place and is still a TVTensor Image, but new_image
# is a Tensor. They share the same underlying data and they're equal, just
# different classes.
assert isinstance(image, tv_tensors.Image)
print(image)
assert isinstance(new_image, torch.Tensor) and not isinstance(new_image, tv_tensors.Image)
assert (new_image == image).all()
assert new_image.data_ptr() == image.data_ptr()
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment