gchhablani commited on
Commit
2bbf92c
1 Parent(s): ba46d36
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitignore +3 -0
  2. answer_reverse_mapping.json +1 -0
  3. app.py +81 -0
  4. ckpt/ckpt-60k-5999/config.json +0 -0
  5. ckpt/ckpt-60k-5999/flax_model.msgpack +3 -0
  6. ckpt/ckpt-60k-5999/opt_state.msgpack +3 -0
  7. ckpt/ckpt-60k-5999/training_state.json +1 -0
  8. dummy_vqa_multilingual.tsv +101 -0
  9. images/val2014/COCO_val2014_000000000757.jpg +0 -0
  10. images/val2014/COCO_val2014_000000013466.jpg +0 -0
  11. images/val2014/COCO_val2014_000000015517.jpg +0 -0
  12. images/val2014/COCO_val2014_000000017018.jpg +0 -0
  13. images/val2014/COCO_val2014_000000023899.jpg +0 -0
  14. images/val2014/COCO_val2014_000000029465.jpg +0 -0
  15. images/val2014/COCO_val2014_000000032901.jpg +0 -0
  16. images/val2014/COCO_val2014_000000036607.jpg +0 -0
  17. images/val2014/COCO_val2014_000000046269.jpg +0 -0
  18. images/val2014/COCO_val2014_000000050165.jpg +0 -0
  19. images/val2014/COCO_val2014_000000066973.jpg +0 -0
  20. images/val2014/COCO_val2014_000000068411.jpg +0 -0
  21. images/val2014/COCO_val2014_000000074646.jpg +0 -0
  22. images/val2014/COCO_val2014_000000079837.jpg +0 -0
  23. images/val2014/COCO_val2014_000000084533.jpg +0 -0
  24. images/val2014/COCO_val2014_000000090255.jpg +0 -0
  25. images/val2014/COCO_val2014_000000095805.jpg +0 -0
  26. images/val2014/COCO_val2014_000000099177.jpg +0 -0
  27. images/val2014/COCO_val2014_000000109055.jpg +0 -0
  28. images/val2014/COCO_val2014_000000109316.jpg +0 -0
  29. images/val2014/COCO_val2014_000000111546.jpg +0 -0
  30. images/val2014/COCO_val2014_000000117328.jpg +0 -0
  31. images/val2014/COCO_val2014_000000117374.jpg +0 -0
  32. images/val2014/COCO_val2014_000000121031.jpg +0 -0
  33. images/val2014/COCO_val2014_000000132042.jpg +0 -0
  34. images/val2014/COCO_val2014_000000144935.jpg +0 -0
  35. images/val2014/COCO_val2014_000000151393.jpg +0 -0
  36. images/val2014/COCO_val2014_000000153011.jpg +0 -0
  37. images/val2014/COCO_val2014_000000156532.jpg +0 -0
  38. images/val2014/COCO_val2014_000000163348.jpg +0 -0
  39. images/val2014/COCO_val2014_000000163640.jpg +0 -0
  40. images/val2014/COCO_val2014_000000167353.jpg +0 -0
  41. images/val2014/COCO_val2014_000000169143.jpg +0 -0
  42. images/val2014/COCO_val2014_000000171255.jpg +0 -0
  43. images/val2014/COCO_val2014_000000187565.jpg +0 -0
  44. images/val2014/COCO_val2014_000000190585.jpg +0 -0
  45. images/val2014/COCO_val2014_000000190689.jpg +0 -0
  46. images/val2014/COCO_val2014_000000197716.jpg +0 -0
  47. images/val2014/COCO_val2014_000000208971.jpg +0 -0
  48. images/val2014/COCO_val2014_000000209602.jpg +0 -0
  49. images/val2014/COCO_val2014_000000230561.jpg +0 -0
  50. images/val2014/COCO_val2014_000000238537.jpg +0 -0
.gitignore ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ *mvqa_env*
2
+ *.pyc
3
+ *__pycache__*
answer_reverse_mapping.json ADDED
@@ -0,0 +1 @@
 
1
+ {"0": "<unk>", "1": "0", "2": "000", "3": "1", "4": "1 4", "5": "1 foot", "6": "1 hour", "7": "1 in back", "8": "1 in front", "9": "1 in middle", "10": "1 inch", "11": "1 on left", "12": "1 on right", "13": "1 way", "14": "1 world", "15": "1 year", "16": "1.00", "17": "10", "18": "10 feet", "19": "10 inches", "20": "10 years", "21": "100", "22": "100 feet", "23": "100 year party ct", "24": "1000", "25": "101", "26": "106", "27": "10:00", "28": "10:05", "29": "10:08", "30": "10:10", "31": "10:15", "32": "10:20", "33": "10:25", "34": "10:30", "35": "10:35", "36": "10:40", "37": "10:45", "38": "10:50", "39": "10:55", "40": "11", "41": "11:00", "42": "11:05", "43": "11:10", "44": "11:15", "45": "11:20", "46": "11:25", "47": "11:30", "48": "11:35", "49": "11:45", "50": "11:50", "51": "11:55", "52": "12", "53": "12 feet", "54": "120", "55": "12:00", "56": "12:05", "57": "12:10", "58": "12:15", "59": "12:20", "60": "12:25", "61": "12:28", "62": "12:30", "63": "12:35", "64": "12:40", "65": "12:45", "66": "12:50", "67": "12:55", "68": "13", "69": "14", "70": "15", "71": "15 feet", "72": "150", "73": "16", "74": "17", "75": "18", "76": "19", "77": "193", "78": "1950", "79": "1950s", "80": "1980", "81": "1990", "82": "1:00", "83": "1:05", "84": "1:10", "85": "1:15", "86": "1:20", "87": "1:25", "88": "1:30", "89": "1:35", "90": "1:40", "91": "1:45", "92": "1:50", "93": "1:55", "94": "1st", "95": "2", "96": "2 feet", "97": "2 hours", "98": "2 men", "99": "2 people", "100": "2 years", "101": "2.00", "102": "20", "103": "20 feet", "104": "20 ft", "105": "200", "106": "2000", "107": "2007", "108": "2008", "109": "2009", "110": "2010", "111": "2011", "112": "2012", "113": "2013", "114": "2015", "115": "2016", "116": "21", "117": "22", "118": "23", "119": "24", "120": "25", "121": "26", "122": "27", "123": "28", "124": "29", "125": "2:00", "126": "2:05", "127": "2:10", "128": "2:15", "129": "2:20", "130": "2:25", "131": "2:30", "132": "2:35", "133": "2:40", "134": "2:45", "135": "2:50", "136": "2:55", "137": "2nd", "138": "3", "139": "3 feet", "140": "3 inches", "141": "30", "142": "30 mph", "143": "300", "144": "31", "145": "32", "146": "33", "147": "34", "148": "35", "149": "350", "150": "36", "151": "37", "152": "38", "153": "39", "154": "3:00", "155": "3:10", "156": "3:15", "157": "3:20", "158": "3:25", "159": "3:30", "160": "3:45", "161": "3:50", "162": "3:55", "163": "3rd", "164": "4", "165": "4 feet", "166": "4 ft", "167": "4 inches", "168": "4 way", "169": "40", "170": "400", "171": "41", "172": "42", "173": "43", "174": "44", "175": "45", "176": "46", "177": "47", "178": "48", "179": "49", "180": "4:00", "181": "4:05", "182": "4:15", "183": "4:20", "184": "4:30", "185": "4:35", "186": "4:40", "187": "4:45", "188": "4:50", "189": "4:55", "190": "4th of july", "191": "5", "192": "5 feet", "193": "5 ft", "194": "5 star", "195": "5 years", "196": "50", "197": "50 feet", "198": "500", "199": "51", "200": "52", "201": "53", "202": "54", "203": "55", "204": "56", "205": "59", "206": "5:00", "207": "5:05", "208": "5:10", "209": "5:15", "210": "5:18", "211": "5:25", "212": "5:30", "213": "5:40", "214": "5:45", "215": "5:50", "216": "5:55", "217": "6", "218": "6 feet", "219": "6 inches", "220": "60", "221": "600", "222": "61", "223": "64", "224": "65", "225": "66", "226": "68", "227": "6:00", "228": "6:05", "229": "6:20", "230": "6:25", "231": "6:30", "232": "6:35", "233": "6:40", "234": "6:45", "235": "7", "236": "7 eleven", "237": "70", "238": "700", "239": "72", "240": "75", "241": "7:00", "242": "7:05", "243": "7:10", "244": "7:25", "245": "7:35", "246": "7:45", "247": "7:55", "248": "8", "249": "8 feet", "250": "80", "251": "870", "252": "88", "253": "8:00", "254": "8:05", "255": "8:35", "256": "8:50", "257": "8:55", "258": "9", "259": "90", "260": "99", "261": "9:05", "262": "9:12", "263": "9:15", "264": "9:20", "265": "9:25", "266": "9:30", "267": "9:35", "268": "9:45", "269": "9:50", "270": "9:55", "271": "aa", "272": "above", "273": "above door", "274": "above sink", "275": "above stove", "276": "above toilet", "277": "abstract", "278": "accident", "279": "acer", "280": "across street", "281": "adidas", "282": "adult", "283": "adults", "284": "advertisement", "285": "africa", "286": "african", "287": "african american", "288": "after", "289": "afternoon", "290": "against wall", "291": "age", "292": "ahead", "293": "air", "294": "air canada", "295": "air conditioner", "296": "air force", "297": "air france", "298": "airplane", "299": "airplanes", "300": "airport", "301": "alaska", "302": "alcohol", "303": "alive", "304": "all", "305": "all of them", "306": "all way", "307": "alligator", "308": "almonds", "309": "alps", "310": "aluminum", "311": "am", "312": "amazon", "313": "ambulance", "314": "america", "315": "american", "316": "american airlines", "317": "american flag", "318": "amtrak", "319": "ana", "320": "analog", "321": "angel", "322": "angels", "323": "angry", "324": "animal", "325": "animals", "326": "ankle", "327": "anniversary", "328": "antelope", "329": "antenna", "330": "antique", "331": "apartment", "332": "apartments", "333": "apple", "334": "apple and banana", "335": "apples", "336": "apron", "337": "arabic", "338": "arch", "339": "arizona", "340": "arm", "341": "army", "342": "around neck", "343": "arriving", "344": "arrow", "345": "arrows", "346": "art", "347": "ascending", "348": "asia", "349": "asian", "350": "asics", "351": "asleep", "352": "asparagus", "353": "asphalt", "354": "at camera", "355": "at table", "356": "at&t", "357": "athletics", "358": "atv", "359": "audi", "360": "australia", "361": "avocado", "362": "awake", "363": "away", "364": "b", "365": "babies", "366": "baby", "367": "baby's breath", "368": "back", "369": "back left", "370": "background", "371": "backhand", "372": "backpack", "373": "backward", "374": "backwards", "375": "backyard", "376": "bacon", "377": "bad", "378": "badminton", "379": "bag", "380": "bagel", "381": "bagels", "382": "baggage claim", "383": "bags", "384": "baked", "385": "baker", "386": "bakery", "387": "baking", "388": "balance", "389": "balcony", "390": "bald", "391": "ball", "392": "balloon", "393": "balloons", "394": "balls", "395": "bamboo", "396": "banana", "397": "banana bread", "398": "banana peel", "399": "banana split", "400": "bananas", "401": "band", "402": "bandana", "403": "bank", "404": "bank of america", "405": "bar", "406": "barbed wire", "407": "barber shop", "408": "bark", "409": "barn", "410": "barrel", "411": "barrier", "412": "bars", "413": "base", "414": "baseball", "415": "baseball bat", "416": "baseball cap", "417": "baseball field", "418": "baseball game", "419": "baseball glove", "420": "baseball player", "421": "baseball uniform", "422": "basil", "423": "basket", "424": "basketball", "425": "baskets", "426": "bat", "427": "bathing", "428": "bathing suit", "429": "bathroom", "430": "bathtub", "431": "batman", "432": "bats", "433": "batter", "434": "batting", "435": "beach", "436": "beads", "437": "beagle", "438": "beanie", "439": "beans", "440": "bear", "441": "beard", "442": "bears", "443": "bed", "444": "bedroom", "445": "beef", "446": "beer", "447": "beets", "448": "before", "449": "behind", "450": "behind bench", "451": "behind bus", "452": "behind clouds", "453": "behind fence", "454": "behind woman", "455": "beige", "456": "beijing", "457": "bell", "458": "below", "459": "belt", "460": "bench", "461": "benches", "462": "bending", "463": "berries", "464": "best buy", "465": "bib", "466": "bible", "467": "bicycle", "468": "bicycles", "469": "bidet", "470": "big", "471": "big ben", "472": "bike", "473": "bike rack", "474": "biker", "475": "bikers", "476": "bikes", "477": "biking", "478": "bikini", "479": "billabong", "480": "bin", "481": "biplane", "482": "bird", "483": "bird feeder", "484": "birds", "485": "birthday", "486": "birthday cake", "487": "birthday party", "488": "black", "489": "black and blue", "490": "black and brown", "491": "black and gray", "492": "black and orange", "493": "black and pink", "494": "black and red", "495": "black and silver", "496": "black and white", "497": "black and yellow", "498": "black white", "499": "blackberry", "500": "blanket", "501": "blankets", "502": "bleachers", "503": "blender", "504": "blending", "505": "blinders", "506": "blinds", "507": "blonde", "508": "blood", "509": "blt", "510": "blue", "511": "blue and black", "512": "blue and gray", "513": "blue and green", "514": "blue and orange", "515": "blue and pink", "516": "blue and red", "517": "blue and white", "518": "blue and yellow", "519": "blue jay", "520": "blue team", "521": "blueberries", "522": "blueberry", "523": "blurry", "524": "bmw", "525": "bnsf", "526": "board", "527": "boarding", "528": "boardwalk", "529": "boat", "530": "boating", "531": "boats", "532": "bob", "533": "bone", "534": "boogie board", "535": "book", "536": "books", "537": "bookshelf", "538": "boot", "539": "boots", "540": "bored", "541": "boredom", "542": "boston", "543": "both", "544": "bottle", "545": "bottles", "546": "bottom", "547": "bottom left", "548": "bottom right", "549": "boundaries", "550": "bow", "551": "bow tie", "552": "bowl", "553": "bowling", "554": "bowls", "555": "bowtie", "556": "box", "557": "boxer", "558": "boxes", "559": "boxing", "560": "boy", "561": "boys", "562": "brace", "563": "bracelet", "564": "braid", "565": "branch", "566": "branches", "567": "brand", "568": "brass", "569": "braves", "570": "brazil", "571": "bread", "572": "breakfast", "573": "brewers", "574": "brick", "575": "bricks", "576": "bride", "577": "bridge", "578": "bridle", "579": "briefcase", "580": "bright", "581": "britain", "582": "british", "583": "british airways", "584": "broadway", "585": "broccoli", "586": "broccoli and carrots", "587": "broke", "588": "broken", "589": "bronze", "590": "broom", "591": "brown", "592": "brown and black", "593": "brown and white", "594": "brush", "595": "brushing", "596": "brushing hair", "597": "brushing her teeth", "598": "brushing his teeth", "599": "brushing teeth", "600": "bucket", "601": "bud light", "602": "budweiser", "603": "buffalo", "604": "building", "605": "buildings", "606": "bull", "607": "bulldog", "608": "bun", "609": "bundt", "610": "bunk", "611": "bunny", "612": "bunt", "613": "buoy", "614": "buoys", "615": "burger", "616": "burgers", "617": "burrito", "618": "burton", "619": "bus", "620": "bus driver", "621": "bus station", "622": "bus stop", "623": "buses", "624": "bush", "625": "bushes", "626": "business", "627": "busy", "628": "butt", "629": "butter", "630": "butterfly", "631": "button", "632": "button up", "633": "buttons", "634": "by window", "635": "c", "636": "cabbage", "637": "cabinet", "638": "cabinets", "639": "cactus", "640": "cadillac", "641": "cafe", "642": "cage", "643": "cake", "644": "cakes", "645": "calendar", "646": "calico", "647": "california", "648": "calm", "649": "camel", "650": "camera", "651": "cameraman", "652": "cameras", "653": "camo", "654": "camouflage", "655": "camper", "656": "camping", "657": "can", "658": "can't see", "659": "can't see it", "660": "can't tell", "661": "canada", "662": "candle", "663": "candles", "664": "candy", "665": "cane", "666": "cannot tell", "667": "canoe", "668": "canon", "669": "canopy", "670": "cantaloupe", "671": "cap", "672": "captivity", "673": "car", "674": "caramel", "675": "cardboard", "676": "cardinal", "677": "cardinals", "678": "cargo", "679": "carnation", "680": "carnations", "681": "carpet", "682": "carriage", "683": "carrot", "684": "carrot cake", "685": "carrots", "686": "cars", "687": "cart", "688": "cartoon", "689": "case", "690": "casserole", "691": "cast iron", "692": "castle", "693": "casual", "694": "cat", "695": "cat and dog", "696": "cat food", "697": "catch", "698": "catch ball", "699": "catch frisbee", "700": "catcher", "701": "catching", "702": "catching frisbee", "703": "catholic", "704": "cats", "705": "caucasian", "706": "cauliflower", "707": "caution", "708": "cd", "709": "cds", "710": "ceiling", "711": "celery", "712": "cell", "713": "cell phone", "714": "cell phones", "715": "cement", "716": "center", "717": "ceramic", "718": "cereal", "719": "cessna", "720": "chain", "721": "chain link", "722": "chains", "723": "chair", "724": "chairs", "725": "chalk", "726": "champagne", "727": "chandelier", "728": "charging", "729": "chase", "730": "checkerboard", "731": "checkered", "732": "checkers", "733": "cheddar", "734": "cheese", "735": "cheesecake", "736": "chef", "737": "cherries", "738": "cherry", "739": "chest", "740": "chevrolet", "741": "chevron", "742": "chevy", "743": "chicago", "744": "chicken", "745": "chihuahua", "746": "child", "747": "children", "748": "chili", "749": "chimney", "750": "china", "751": "china airlines", "752": "chinese", "753": "chips", "754": "chiquita", "755": "chocolate", "756": "choppy", "757": "chopsticks", "758": "christian", "759": "christmas", "760": "christmas tree", "761": "chrome", "762": "church", "763": "cigarette", "764": "cigarettes", "765": "cilantro", "766": "cinnamon", "767": "circle", "768": "circles", "769": "circus", "770": "cirrus", "771": "citizen", "772": "city", "773": "city bus", "774": "clams", "775": "classic", "776": "classroom", "777": "clay", "778": "clean", "779": "cleaner", "780": "cleaning", "781": "clear", "782": "cleats", "783": "climbing", "784": "clip", "785": "clock", "786": "clock tower", "787": "clocks", "788": "close", "789": "close up", "790": "closed", "791": "closet", "792": "cloth", "793": "clothes", "794": "clothing", "795": "cloud", "796": "clouds", "797": "cloudy", "798": "club", "799": "cluttered", "800": "clydesdale", "801": "cnn", "802": "coach", "803": "coal", "804": "coaster", "805": "coat", "806": "coats", "807": "cobblestone", "808": "coca cola", "809": "cocker spaniel", "810": "coconut", "811": "coffee", "812": "coffee cup", "813": "coffee maker", "814": "coffee pot", "815": "coffee table", "816": "coins", "817": "coke", "818": "cold", "819": "coleslaw", "820": "colgate", "821": "collage", "822": "collar", "823": "collie", "824": "color", "825": "colorado", "826": "colored", "827": "comcast", "828": "comfort", "829": "comforter", "830": "coming", "831": "commercial", "832": "commuter", "833": "compaq", "834": "competition", "835": "computer", "836": "computers", "837": "concentration", "838": "concert", "839": "concrete", "840": "condiments", "841": "conductor", "842": "cone", "843": "cones", "844": "conference", "845": "conference room", "846": "confused", "847": "congratulations", "848": "construction", "849": "container", "850": "continental", "851": "control", "852": "controller", "853": "controllers", "854": "converse", "855": "cook", "856": "cooked", "857": "cookie", "858": "cookies", "859": "cooking", "860": "cool", "861": "cooler", "862": "copper", "863": "copyright", "864": "cord", "865": "corgi", "866": "corn", "867": "corner", "868": "corona", "869": "cosmo", "870": "costume", "871": "cotton", "872": "couch", "873": "counter", "874": "country", "875": "countryside", "876": "couple", "877": "court", "878": "cover", "879": "cow", "880": "cowboy", "881": "cows", "882": "crafts", "883": "crane", "884": "cranes", "885": "crates", "886": "cream", "887": "crest", "888": "crib", "889": "crocs", "890": "croissant", "891": "cross", "892": "cross country", "893": "crossing", "894": "crosstown", "895": "crosswalk", "896": "crow", "897": "crown", "898": "crows", "899": "cruise ship", "900": "csx", "901": "cubs", "902": "cucumber", "903": "cucumbers", "904": "cuddling", "905": "cumulus", "906": "cup", "907": "cupcake", "908": "cupcakes", "909": "cups", "910": "curb", "911": "curious", "912": "curly", "913": "current", "914": "curtain", "915": "curtains", "916": "curved", "917": "cushion", "918": "cut", "919": "cute", "920": "cutting", "921": "cutting board", "922": "cutting cake", "923": "cutting hair", "924": "cycling", "925": "cylinder", "926": "d", "927": "dachshund", "928": "dad", "929": "daffodil", "930": "daffodils", "931": "dairy", "932": "dairy queen", "933": "daisies", "934": "daisy", "935": "dalmatian", "936": "dancing", "937": "dandelions", "938": "dark", "939": "dawn", "940": "day", "941": "day time", "942": "daytime", "943": "db", "944": "dc", "945": "dead", "946": "dead end", "947": "deck", "948": "decoration", "949": "decorative", "950": "deep", "951": "deer", "952": "defense", "953": "deli", "954": "delivery", "955": "dell", "956": "delta", "957": "denim", "958": "descending", "959": "desert", "960": "design", "961": "desk", "962": "desktop", "963": "dessert", "964": "desserts", "965": "detroit", "966": "diamond", "967": "diamonds", "968": "diesel", "969": "diet coke", "970": "different teams", "971": "digital", "972": "dim", "973": "dining", "974": "dining room", "975": "dinner", "976": "dinosaur", "977": "dip", "978": "direction", "979": "directions", "980": "dirt", "981": "dirt bike", "982": "dirty", "983": "dishes", "984": "dishwasher", "985": "disney", "986": "display", "987": "distance", "988": "do not enter", "989": "dock", "990": "dodge", "991": "dodgers", "992": "dog", "993": "dog and cat", "994": "dog bed", "995": "dog food", "996": "dog show", "997": "dogs", "998": "dole", "999": "doll", "1000": "dome", "1001": "domestic", "1002": "don't know", "1003": "don't walk", "1004": "donkey", "1005": "donut", "1006": "donut shop", "1007": "donuts", "1008": "door", "1009": "doorway", "1010": "dots", "1011": "double", "1012": "double decker", "1013": "doubles", "1014": "dough", "1015": "doughnut", "1016": "doughnuts", "1017": "down", "1018": "down street", "1019": "downhill", "1020": "downtown", "1021": "dr pepper", "1022": "dragon", "1023": "drain", "1024": "drawer", "1025": "drawing", "1026": "dreadlocks", "1027": "dress", "1028": "dresser", "1029": "drink", "1030": "drinking", "1031": "drinking water", "1032": "drinks", "1033": "drive", "1034": "driver", "1035": "driveway", "1036": "driving", "1037": "drums", "1038": "dry", "1039": "drying", "1040": "drywall", "1041": "ducati", "1042": "duck", "1043": "ducks", "1044": "dugout", "1045": "dump", "1046": "dump truck", "1047": "dunkin donuts", "1048": "dusk", "1049": "e", "1050": "each other", "1051": "eagle", "1052": "ear", "1053": "earbuds", "1054": "earring", "1055": "earrings", "1056": "ears", "1057": "east", "1058": "easter", "1059": "easton", "1060": "easy", "1061": "easyjet", "1062": "eat", "1063": "eaten", "1064": "eating", "1065": "egg", "1066": "egg salad", "1067": "eggs", "1068": "eiffel tower", "1069": "electric", "1070": "electricity", "1071": "electronics", "1072": "elephant", "1073": "elephants", "1074": "elm", "1075": "elmo", "1076": "email", "1077": "emergency", "1078": "emirates", "1079": "empty", "1080": "enclosure", "1081": "end", "1082": "engine", "1083": "england", "1084": "english", "1085": "entering", "1086": "equestrian", "1087": "europe", "1088": "evening", "1089": "evergreen", "1090": "exhaust", "1091": "exit", "1092": "eyes", "1093": "f", "1094": "fabric", "1095": "face", "1096": "facebook", "1097": "factory", "1098": "fair", "1099": "fake", "1100": "fall", "1101": "falling", "1102": "family", "1103": "fan", "1104": "fancy", "1105": "fans", "1106": "fanta", "1107": "far", "1108": "far right", "1109": "farm", "1110": "farmer", "1111": "farmers", "1112": "farmers market", "1113": "fashion", "1114": "fast", "1115": "fast food", "1116": "father", "1117": "faucet", "1118": "feathers", "1119": "fedex", "1120": "fedora", "1121": "feeder", "1122": "feeding", "1123": "feeding giraffe", "1124": "feet", "1125": "fell", "1126": "female", "1127": "fence", "1128": "fern", "1129": "ferris wheel", "1130": "ferry", "1131": "festival", "1132": "feta", "1133": "few", "1134": "field", "1135": "fighter", "1136": "fighting", "1137": "finch", "1138": "finger", "1139": "fire", "1140": "fire extinguisher", "1141": "fire hydrant", "1142": "fire truck", "1143": "firefighter", "1144": "fireman", "1145": "fireplace", "1146": "fires", "1147": "first", "1148": "first base", "1149": "fish", "1150": "fisheye", "1151": "fishing", "1152": "fishing boat", "1153": "flag", "1154": "flags", "1155": "flamingo", "1156": "flashlight", "1157": "flat", "1158": "flat screen", "1159": "flats", "1160": "flickr", "1161": "flip", "1162": "flip flops", "1163": "flip phone", "1164": "floating", "1165": "flood", "1166": "floor", "1167": "floral", "1168": "florida", "1169": "flour", "1170": "flower", "1171": "flowers", "1172": "fluffy", "1173": "fluorescent", "1174": "fly", "1175": "fly kite", "1176": "flying", "1177": "flying kite", "1178": "flying kites", "1179": "foam", "1180": "focus", "1181": "fog", "1182": "foggy", "1183": "foil", "1184": "food", "1185": "food processor", "1186": "food truck", "1187": "foot", "1188": "football", "1189": "footprints", "1190": "for balance", "1191": "for fun", "1192": "for photo", "1193": "for sale", "1194": "ford", "1195": "foreground", "1196": "forehand", "1197": "forest", "1198": "fork", "1199": "fork and knife", "1200": "fork and spoon", "1201": "forks", "1202": "formal", "1203": "formica", "1204": "forward", "1205": "fountain", "1206": "fox", "1207": "frame", "1208": "france", "1209": "free", "1210": "freezer", "1211": "freight", "1212": "freightliner", "1213": "french", "1214": "french fries", "1215": "fresh", "1216": "fridge", "1217": "fried", "1218": "friend", "1219": "friends", "1220": "fries", "1221": "frisbee", "1222": "frisbees", "1223": "frog", "1224": "front", "1225": "frosted", "1226": "frosting", "1227": "fruit", "1228": "fruit salad", "1229": "fruits", "1230": "full", "1231": "fun", "1232": "fur", "1233": "furniture", "1234": "futon", "1235": "g", "1236": "game", "1237": "game controller", "1238": "gaming", "1239": "garage", "1240": "garbage", "1241": "garden", "1242": "garlic", "1243": "gas", "1244": "gas station", "1245": "gate", "1246": "gatorade", "1247": "gazebo", "1248": "ge", "1249": "geese", "1250": "genetics", "1251": "german", "1252": "german shepherd", "1253": "germany", "1254": "ghost", "1255": "giants", "1256": "ginger", "1257": "giraffe", "1258": "giraffe and zebra", "1259": "giraffes", "1260": "girl", "1261": "girl on right", "1262": "girls", "1263": "give way", "1264": "glass", "1265": "glasses", "1266": "glaze", "1267": "glazed", "1268": "globe", "1269": "glove", "1270": "gloves", "1271": "gmc", "1272": "go", "1273": "goal", "1274": "goalie", "1275": "goat", "1276": "goatee", "1277": "goats", "1278": "goggles", "1279": "going", "1280": "gold", "1281": "golden gate", "1282": "golden retriever", "1283": "golf", "1284": "gone", "1285": "good", "1286": "google", "1287": "goose", "1288": "gothic", "1289": "graduation", "1290": "graffiti", "1291": "grandfather", "1292": "granite", "1293": "grape", "1294": "grapefruit", "1295": "grapes", "1296": "grass", "1297": "grassy", "1298": "gravel", "1299": "gravy", "1300": "gray", "1301": "gray and black", "1302": "gray and red", "1303": "gray and white", "1304": "grazing", "1305": "green", "1306": "green and black", "1307": "green and blue", "1308": "green and brown", "1309": "green and orange", "1310": "green and red", "1311": "green and white", "1312": "green and yellow", "1313": "green beans", "1314": "greyhound", "1315": "grill", "1316": "grilled", "1317": "grilled cheese", "1318": "grind", "1319": "grinding", "1320": "grizzly", "1321": "grocery", "1322": "grocery store", "1323": "ground", "1324": "guitar", "1325": "guitar hero", "1326": "gun", "1327": "gym", "1328": "h", "1329": "hair", "1330": "hair dryer", "1331": "haircut", "1332": "half", "1333": "half full", "1334": "halloween", "1335": "hallway", "1336": "ham", "1337": "ham and cheese", "1338": "hamburger", "1339": "hammer time", "1340": "hammock", "1341": "hand", "1342": "handicap", "1343": "handle", "1344": "handlebars", "1345": "hands", "1346": "hanger", "1347": "hanging", "1348": "happiness", "1349": "happy", "1350": "happy birthday", "1351": "harbor", "1352": "hard", "1353": "hardwood", "1354": "harley", "1355": "harley davidson", "1356": "harness", "1357": "harry potter", "1358": "hat", "1359": "hats", "1360": "hauling", "1361": "hawaii", "1362": "hawaiian", "1363": "hawk", "1364": "hay", "1365": "hazy", "1366": "he isn't", "1367": "he's not", "1368": "head", "1369": "headband", "1370": "headphones", "1371": "healthy", "1372": "heart", "1373": "hearts", "1374": "heat", "1375": "heater", "1376": "heavy", "1377": "heels", "1378": "heineken", "1379": "heinz", "1380": "helicopter", "1381": "hello kitty", "1382": "helmet", "1383": "helmets", "1384": "herd", "1385": "herding", "1386": "herself", "1387": "hexagon", "1388": "hiding", "1389": "high", "1390": "high chair", "1391": "high heels", "1392": "highway", "1393": "hiking", "1394": "hill", "1395": "hills", "1396": "hilly", "1397": "himself", "1398": "hispanic", "1399": "hit", "1400": "hit ball", "1401": "hitting", "1402": "hitting ball", "1403": "hockey", "1404": "holding", "1405": "holding baby", "1406": "holding it", "1407": "holding phone", "1408": "holding umbrella", "1409": "hollywood", "1410": "home", "1411": "home plate", "1412": "homemade", "1413": "honda", "1414": "honey", "1415": "hood", "1416": "hoodie", "1417": "horizontal", "1418": "horizontally", "1419": "horns", "1420": "horse", "1421": "horse racing", "1422": "horseback riding", "1423": "horses", "1424": "hose", "1425": "hospital", "1426": "hot", "1427": "hot dog", "1428": "hot dogs", "1429": "hot sauce", "1430": "hotel", "1431": "hotel room", "1432": "house", "1433": "houses", "1434": "hp", "1435": "hsbc", "1436": "htc", "1437": "huge", "1438": "hugging", "1439": "human", "1440": "humans", "1441": "hummingbird", "1442": "hundreds", "1443": "hungry", "1444": "husky", "1445": "hydrant", "1446": "i", "1447": "i don't know", "1448": "ibm", "1449": "ice", "1450": "ice cream", "1451": "icing", "1452": "identification", "1453": "illinois", "1454": "in", "1455": "in air", "1456": "in back", "1457": "in background", "1458": "in basket", "1459": "in bowl", "1460": "in box", "1461": "in cabbage town", "1462": "in car", "1463": "in corner", "1464": "in cup", "1465": "in field", "1466": "in front", "1467": "in grass", "1468": "in hand", "1469": "in her hand", "1470": "in his hand", "1471": "in middle", "1472": "in motion", "1473": "in sink", "1474": "in sky", "1475": "in snow", "1476": "in stands", "1477": "in street", "1478": "in suitcase", "1479": "in vase", "1480": "in water", "1481": "index", "1482": "india", "1483": "indian", "1484": "indians", "1485": "indoor", "1486": "indoors", "1487": "information", "1488": "inside", "1489": "intersection", "1490": "iphone", "1491": "ipod", "1492": "ireland", "1493": "iris", "1494": "iron", "1495": "island", "1496": "it isn't", "1497": "it's not", "1498": "it's raining", "1499": "italian", "1500": "italy", "1501": "ivory", "1502": "ivy", "1503": "j", "1504": "jacket", "1505": "jackets", "1506": "jal", "1507": "japan", "1508": "japanese", "1509": "jar", "1510": "jeans", "1511": "jeep", "1512": "jelly", "1513": "jesus", "1514": "jet", "1515": "jet ski", "1516": "jetblue", "1517": "jets", "1518": "jockey", "1519": "john", "1520": "jones", "1521": "joshua", "1522": "jp morgan", "1523": "juice", "1524": "jump", "1525": "jumped", "1526": "jumping", "1527": "jungle", "1528": "junk", "1529": "k", "1530": "kangaroo", "1531": "kawasaki", "1532": "kayak", "1533": "kayaking", "1534": "kenmore", "1535": "ketchup", "1536": "ketchup and mustard", "1537": "kettle", "1538": "keyboard", "1539": "keys", "1540": "khaki", "1541": "kia", "1542": "kicking", "1543": "kickstand", "1544": "kid", "1545": "kids", "1546": "king", "1547": "kissing", "1548": "kitchen", "1549": "kitchenaid", "1550": "kite", "1551": "kite flying", "1552": "kite string", "1553": "kites", "1554": "kitesurfing", "1555": "kiting", "1556": "kitten", "1557": "kiwi", "1558": "klm", "1559": "knee pads", "1560": "kneeling", "1561": "knife", "1562": "knife and fork", "1563": "knives", "1564": "kodak", "1565": "korean air", "1566": "krispy kreme", "1567": "l", "1568": "la", "1569": "lab", "1570": "labrador", "1571": "lace", "1572": "lacoste", "1573": "ladder", "1574": "lady", "1575": "ladybug", "1576": "lake", "1577": "lamb", "1578": "lamp", "1579": "lamps", "1580": "land", "1581": "landing", "1582": "landscape", "1583": "lanes", "1584": "lanyard", "1585": "lap", "1586": "laptop", "1587": "laptops", "1588": "large", "1589": "laughing", "1590": "laundry", "1591": "laying", "1592": "laying down", "1593": "lays", "1594": "leaf", "1595": "leaning", "1596": "learning", "1597": "leash", "1598": "leather", "1599": "leaves", "1600": "leaving", "1601": "left", "1602": "left 1", "1603": "left and right", "1604": "left side", "1605": "leg", "1606": "lego", "1607": "legos", "1608": "legs", "1609": "lemon", "1610": "lemonade", "1611": "lemons", "1612": "leopard", "1613": "letters", "1614": "lettuce", "1615": "lexus", "1616": "lg", "1617": "library", "1618": "license plate", "1619": "licking", "1620": "lid", "1621": "life", "1622": "life jacket", "1623": "life vest", "1624": "lifeguard", "1625": "lift", "1626": "light", "1627": "lighter", "1628": "lighthouse", "1629": "lighting", "1630": "lights", "1631": "lilac", "1632": "lilies", "1633": "lily", "1634": "lime", "1635": "limes", "1636": "lines", "1637": "linoleum", "1638": "lion", "1639": "liquor", "1640": "listening", "1641": "listening to music", "1642": "little", "1643": "little girl", "1644": "living", "1645": "living room", "1646": "lizard", "1647": "loading", "1648": "lobster", "1649": "log", "1650": "logitech", "1651": "logo", "1652": "logs", "1653": "london", "1654": "long", "1655": "long sleeve", "1656": "long time", "1657": "looking", "1658": "looking at camera", "1659": "looking at phone", "1660": "looking out window", "1661": "los angeles", "1662": "lot", "1663": "lotion", "1664": "lots", "1665": "love", "1666": "low", "1667": "lufthansa", "1668": "luggage", "1669": "lunch", "1670": "lying down", "1671": "m", "1672": "mac", "1673": "macaroni", "1674": "machine", "1675": "mack", "1676": "magazine", "1677": "magazines", "1678": "magnet", "1679": "magnets", "1680": "mailbox", "1681": "main", "1682": "main st", "1683": "main street", "1684": "makeup", "1685": "male", "1686": "males", "1687": "mall", "1688": "man", "1689": "man in middle", "1690": "man made", "1691": "man on left", "1692": "man on right", "1693": "man's", "1694": "mane", "1695": "mango", "1696": "mantle", "1697": "many", "1698": "map", "1699": "maple", "1700": "maple leaf", "1701": "marble", "1702": "marina", "1703": "mariners", "1704": "mario", "1705": "marker", "1706": "market", "1707": "maroon", "1708": "married", "1709": "marshmallows", "1710": "mask", "1711": "mat", "1712": "mattress", "1713": "mayo", "1714": "mayonnaise", "1715": "mcdonald's", "1716": "me", "1717": "meat", "1718": "meatballs", "1719": "medium", "1720": "meeting", "1721": "men", "1722": "men's", "1723": "menu", "1724": "meow", "1725": "mercedes", "1726": "mercedes benz", "1727": "messy", "1728": "metal", "1729": "meter", "1730": "metro", "1731": "mets", "1732": "mexican", "1733": "mexico", "1734": "miami", "1735": "michigan", "1736": "mickey mouse", "1737": "microphone", "1738": "microsoft", "1739": "microwave", "1740": "middle", "1741": "middle 1", "1742": "military", "1743": "milk", "1744": "millions", "1745": "minnie mouse", "1746": "mint", "1747": "mirror", "1748": "mirrors", "1749": "mississippi", "1750": "mitsubishi", "1751": "mitt", "1752": "mixer", "1753": "model", "1754": "modern", "1755": "mohawk", "1756": "mom", "1757": "monday", "1758": "money", "1759": "monitor", "1760": "monkey", "1761": "monster", "1762": "moon", "1763": "moped", "1764": "more", "1765": "morning", "1766": "mosaic", "1767": "moss", "1768": "motel", "1769": "mother", "1770": "mother and child", "1771": "motion", "1772": "motocross", "1773": "motor", "1774": "motorbike", "1775": "motorcycle", "1776": "motorcycles", "1777": "motorola", "1778": "mound", "1779": "mountain", "1780": "mountain dew", "1781": "mountainous", "1782": "mountains", "1783": "mouse", "1784": "mouse pad", "1785": "mouth", "1786": "mouthwash", "1787": "movement", "1788": "movie", "1789": "moving", "1790": "mozzarella", "1791": "mt airy", "1792": "mud", "1793": "muffin", "1794": "muffins", "1795": "mug", "1796": "multi", "1797": "multi colored", "1798": "multicolored", "1799": "multiple", "1800": "mural", "1801": "museum", "1802": "mushroom", "1803": "mushrooms", "1804": "music", "1805": "mustache", "1806": "mustard", "1807": "mutt", "1808": "n", "1809": "name", "1810": "name tag", "1811": "napkin", "1812": "napkins", "1813": "nasa", "1814": "nathan's", "1815": "national express", "1816": "natural", "1817": "nature", "1818": "navy", "1819": "neck", "1820": "necklace", "1821": "neither", "1822": "neon", "1823": "nest", "1824": "net", "1825": "never", "1826": "new", "1827": "new orleans", "1828": "new york", "1829": "news", "1830": "newspaper", "1831": "next to toilet", "1832": "night", "1833": "night time", "1834": "nightstand", "1835": "nighttime", "1836": "nike", "1837": "nikon", "1838": "nintendo", "1839": "nissan", "1840": "no", "1841": "no 1", "1842": "no cat", "1843": "no clock", "1844": "no dog", "1845": "no flag", "1846": "no grass", "1847": "no hat", "1848": "no left turn", "1849": "no light", "1850": "no man", "1851": "no number", "1852": "no parking", "1853": "no plate", "1854": "no shirt", "1855": "no sign", "1856": "no smoking", "1857": "no train", "1858": "no water", "1859": "nobody", "1860": "nokia", "1861": "noodles", "1862": "noon", "1863": "normal", "1864": "north", "1865": "north america", "1866": "north face", "1867": "nose", "1868": "not", "1869": "not at all", "1870": "not here", "1871": "not high", "1872": "not in service", "1873": "not likely", "1874": "not long", "1875": "not possible", "1876": "not sure", "1877": "not there", "1878": "not very", "1879": "notebook", "1880": "notes", "1881": "nothing", "1882": "now", "1883": "nowhere", "1884": "numbers", "1885": "nursing", "1886": "nuts", "1887": "ny", "1888": "o", "1889": "oak", "1890": "oar", "1891": "oars", "1892": "obama", "1893": "ocean", "1894": "octagon", "1895": "octopus", "1896": "off", "1897": "office", "1898": "oil", "1899": "old", "1900": "older", "1901": "olives", "1902": "ollie", "1903": "olympics", "1904": "omelet", "1905": "on", "1906": "on beach", "1907": "on bed", "1908": "on bench", "1909": "on bike", "1910": "on boat", "1911": "on building", "1912": "on bus", "1913": "on car", "1914": "on chair", "1915": "on couch", "1916": "on counter", "1917": "on desk", "1918": "on dresser", "1919": "on elephant", "1920": "on floor", "1921": "on fridge", "1922": "on grass", "1923": "on ground", "1924": "on his face", "1925": "on his head", "1926": "on horse", "1927": "on laptop", "1928": "on left", "1929": "on man", "1930": "on motorcycle", "1931": "on napkin", "1932": "on phone", "1933": "on pizza", "1934": "on plane", "1935": "on plate", "1936": "on pole", "1937": "on rack", "1938": "on right", "1939": "on road", "1940": "on rock", "1941": "on runway", "1942": "on shelf", "1943": "on shore", "1944": "on sidewalk", "1945": "on sign", "1946": "on sink", "1947": "on skateboard", "1948": "on stove", "1949": "on street", "1950": "on suitcase", "1951": "on table", "1952": "on toilet", "1953": "on top", "1954": "on tower", "1955": "on track", "1956": "on tracks", "1957": "on train", "1958": "on tray", "1959": "on tree", "1960": "on wall", "1961": "on water", "1962": "on woman", "1963": "onion", "1964": "onion rings", "1965": "onions", "1966": "only", "1967": "opaque", "1968": "open", "1969": "opponent", "1970": "orange", "1971": "orange and black", "1972": "orange and blue", "1973": "orange and white", "1974": "orange and yellow", "1975": "orange juice", "1976": "oranges", "1977": "orchid", "1978": "oregon", "1979": "organic", "1980": "oriental", "1981": "orioles", "1982": "ostrich", "1983": "ottoman", "1984": "out", "1985": "out of focus", "1986": "outdoor", "1987": "outdoors", "1988": "outfield", "1989": "outside", "1990": "oval", "1991": "oven", "1992": "over", "1993": "over easy", "1994": "overalls", "1995": "overcast", "1996": "owl", "1997": "owner", "1998": "p", "1999": "pacific", "2000": "pacifier", "2001": "packing", "2002": "paddle", "2003": "paddle boarding", "2004": "paddling", "2005": "paint", "2006": "painted", "2007": "painting", "2008": "paisley", "2009": "pajamas", "2010": "palm", "2011": "palm tree", "2012": "palm trees", "2013": "pan", "2014": "pancake", "2015": "pancakes", "2016": "panda", "2017": "pans", "2018": "pants", "2019": "paper", "2020": "paper towels", "2021": "papers", "2022": "parachute", "2023": "parade", "2024": "parakeet", "2025": "parasailing", "2026": "pare", "2027": "paris", "2028": "park", "2029": "parked", "2030": "parking", "2031": "parking garage", "2032": "parking lot", "2033": "parking meter", "2034": "parking meters", "2035": "parmesan", "2036": "parmesan cheese", "2037": "parrot", "2038": "parrots", "2039": "parsley", "2040": "partly cloudy", "2041": "party", "2042": "passenger", "2043": "passengers", "2044": "pasta", "2045": "pastries", "2046": "pastry", "2047": "pasture", "2048": "patio", "2049": "patterned", "2050": "paved", "2051": "pavement", "2052": "paw", "2053": "pc", "2054": "peace", "2055": "peach", "2056": "peaches", "2057": "peacock", "2058": "peanut butter", "2059": "peanuts", "2060": "pear", "2061": "pearl", "2062": "peas", "2063": "pedestal", "2064": "pedestrian", "2065": "pedestrian crossing", "2066": "pedestrians", "2067": "pee", "2068": "peeing", "2069": "pelican", "2070": "pelicans", "2071": "pen", "2072": "pencil", "2073": "penguin", "2074": "penne", "2075": "pens", "2076": "people", "2077": "pepper", "2078": "pepperoni", "2079": "peppers", "2080": "pepsi", "2081": "persian", "2082": "person", "2083": "petting", "2084": "petting horse", "2085": "philadelphia", "2086": "phillies", "2087": "phone", "2088": "phones", "2089": "photo", "2090": "photograph", "2091": "photographer", "2092": "photography", "2093": "photoshop", "2094": "piano", "2095": "pickle", "2096": "pickles", "2097": "pickup", "2098": "picnic", "2099": "picnic table", "2100": "picture", "2101": "pictures", "2102": "pie", "2103": "pier", "2104": "pig", "2105": "pigeon", "2106": "pigeons", "2107": "pigtails", "2108": "pillow", "2109": "pillows", "2110": "pilot", "2111": "pine", "2112": "pineapple", "2113": "ping pong", "2114": "pink", "2115": "pink and black", "2116": "pink and blue", "2117": "pink and white", "2118": "pink and yellow", "2119": "pipe", "2120": "pipes", "2121": "pirate", "2122": "pirates", "2123": "pitbull", "2124": "pitch", "2125": "pitcher", "2126": "pitching", "2127": "pizza", "2128": "pizza box", "2129": "pizza cutter", "2130": "pizza hut", "2131": "placemat", "2132": "plaid", "2133": "plain", "2134": "plane", "2135": "planes", "2136": "plant", "2137": "planter", "2138": "plants", "2139": "plaster", "2140": "plastic", "2141": "plastic wrap", "2142": "plate", "2143": "plates", "2144": "platform", "2145": "play", "2146": "play tennis", "2147": "player", "2148": "players", "2149": "playing", "2150": "playing baseball", "2151": "playing frisbee", "2152": "playing game", "2153": "playing soccer", "2154": "playing tennis", "2155": "playing video game", "2156": "playing video games", "2157": "playing wii", "2158": "playstation", "2159": "plow", "2160": "plunger", "2161": "pm", "2162": "pocket", "2163": "pockets", "2164": "pointing", "2165": "polar", "2166": "polar bear", "2167": "polar bears", "2168": "pole", "2169": "poles", "2170": "police", "2171": "police officer", "2172": "polka dot", "2173": "polka dots", "2174": "polo", "2175": "pomeranian", "2176": "pond", "2177": "pony", "2178": "ponytail", "2179": "poodle", "2180": "pool", "2181": "poop", "2182": "pooping", "2183": "poor", "2184": "porcelain", "2185": "porch", "2186": "pork", "2187": "posing", "2188": "post", "2189": "poster", "2190": "posts", "2191": "pot", "2192": "potato", "2193": "potato salad", "2194": "potatoes", "2195": "pots", "2196": "pottery", "2197": "powdered", "2198": "powdered sugar", "2199": "power", "2200": "power lines", "2201": "practice", "2202": "prince", "2203": "print", "2204": "printer", "2205": "privacy", "2206": "private", "2207": "produce", "2208": "professional", "2209": "prom", "2210": "propeller", "2211": "protection", "2212": "protest", "2213": "public", "2214": "public market center", "2215": "pug", "2216": "pull", "2217": "puma", "2218": "pumpkin", "2219": "puppy", "2220": "purple", "2221": "purple and white", "2222": "purse", "2223": "qantas", "2224": "qatar", "2225": "queen", "2226": "quilt", "2227": "r", "2228": "rabbit", "2229": "race", "2230": "racing", "2231": "rack", "2232": "racket", "2233": "rackets", "2234": "racquet", "2235": "radiator", "2236": "radio", "2237": "radish", "2238": "raft", "2239": "rail", "2240": "railing", "2241": "railroad crossing", "2242": "rain", "2243": "rainbow", "2244": "raining", "2245": "rainy", "2246": "ram", "2247": "ramp", "2248": "ranch", "2249": "raspberries", "2250": "raspberry", "2251": "raw", "2252": "rays", "2253": "reading", "2254": "real", "2255": "rear", "2256": "recently", "2257": "recessed", "2258": "recliner", "2259": "rectangle", "2260": "rectangles", "2261": "red", "2262": "red and black", "2263": "red and blue", "2264": "red and gray", "2265": "red and green", "2266": "red and silver", "2267": "red and white", "2268": "red and yellow", "2269": "red bull", "2270": "red light", "2271": "red sox", "2272": "red velvet", "2273": "red white and blue", "2274": "red white blue", "2275": "reds", "2276": "referee", "2277": "reflection", "2278": "refrigerator", "2279": "refrigerators", "2280": "regular", "2281": "reins", "2282": "relaxing", "2283": "relish", "2284": "remodeling", "2285": "remote", "2286": "remote control", "2287": "remotes", "2288": "residential", "2289": "restaurant", "2290": "resting", "2291": "ribbon", "2292": "rice", "2293": "ride", "2294": "riding", "2295": "riding bike", "2296": "riding bikes", "2297": "riding elephant", "2298": "riding horse", "2299": "riding horses", "2300": "riding motorcycle", "2301": "right", "2302": "right 1", "2303": "right hand", "2304": "right side", "2305": "ring", "2306": "ring finger", "2307": "ripe", "2308": "river", "2309": "road", "2310": "roast beef", "2311": "robe", "2312": "robin", "2313": "robot", "2314": "rock", "2315": "rocks", "2316": "rocky", "2317": "rodeo", "2318": "rolex", "2319": "roll", "2320": "roman", "2321": "roman numerals", "2322": "roof", "2323": "room", "2324": "rooster", "2325": "rope", "2326": "rose", "2327": "roses", "2328": "rottweiler", "2329": "rough", "2330": "round", "2331": "roundabout", "2332": "rowing", "2333": "rubber", "2334": "rug", "2335": "rugby", "2336": "run", "2337": "running", "2338": "runway", "2339": "rural", "2340": "russia", "2341": "russian", "2342": "rust", "2343": "rv", "2344": "rye", "2345": "s", "2346": "sad", "2347": "saddle", "2348": "safari", "2349": "safe", "2350": "safety", "2351": "sail", "2352": "sailboat", "2353": "sailboats", "2354": "sailing", "2355": "salad", "2356": "salmon", "2357": "salon", "2358": "salt", "2359": "salt and pepper", "2360": "samsung", "2361": "san diego", "2362": "san francisco", "2363": "sand", "2364": "sandals", "2365": "sandwich", "2366": "sandwiches", "2367": "santa", "2368": "santa hat", "2369": "sas", "2370": "sauce", "2371": "sauerkraut", "2372": "sausage", "2373": "savannah", "2374": "savory", "2375": "scale", "2376": "scania", "2377": "scarf", "2378": "scenery", "2379": "schnauzer", "2380": "school", "2381": "school bus", "2382": "scissors", "2383": "scooter", "2384": "scrambled", "2385": "scratching", "2386": "screen", "2387": "seafood", "2388": "seagull", "2389": "seagulls", "2390": "seat", "2391": "seattle", "2392": "seaweed", "2393": "second", "2394": "security", "2395": "sedan", "2396": "seeds", "2397": "selfie", "2398": "selling", "2399": "semi", "2400": "sepia", "2401": "serious", "2402": "serve", "2403": "serving", "2404": "sesame", "2405": "sesame seeds", "2406": "setting", "2407": "several", "2408": "sewing", "2409": "shade", "2410": "shadow", "2411": "shadows", "2412": "shaking hands", "2413": "shallow", "2414": "shampoo", "2415": "shape", "2416": "shark", "2417": "shaved", "2418": "shearing", "2419": "shed", "2420": "sheep", "2421": "sheepdog", "2422": "sheet", "2423": "sheets", "2424": "shelf", "2425": "shell", "2426": "shells", "2427": "shelter", "2428": "shelves", "2429": "shepherd", "2430": "shih tzu", "2431": "shingles", "2432": "ship", "2433": "shirt", "2434": "shirt and tie", "2435": "shirts", "2436": "shoe", "2437": "shoes", "2438": "shop", "2439": "shopping", "2440": "shopping cart", "2441": "shore", "2442": "short", "2443": "shorter", "2444": "shorts", "2445": "shoulder", "2446": "show", "2447": "shower", "2448": "shower curtain", "2449": "shower head", "2450": "shrimp", "2451": "shut", "2452": "siamese", "2453": "siblings", "2454": "side", "2455": "side of road", "2456": "sidecar", "2457": "sidewalk", "2458": "sideways", "2459": "sign", "2460": "signs", "2461": "silk", "2462": "silver", "2463": "silver and black", "2464": "silver and red", "2465": "silverware", "2466": "singapore", "2467": "singing", "2468": "single", "2469": "single engine", "2470": "singles", "2471": "sink", "2472": "sitting", "2473": "size", "2474": "skate", "2475": "skate park", "2476": "skateboard", "2477": "skateboarder", "2478": "skateboarding", "2479": "skateboards", "2480": "skatepark", "2481": "skating", "2482": "skeleton", "2483": "ski", "2484": "ski boots", "2485": "ski lift", "2486": "ski pole", "2487": "ski poles", "2488": "ski resort", "2489": "ski slope", "2490": "skier", "2491": "skiers", "2492": "skiing", "2493": "skirt", "2494": "skis", "2495": "skull", "2496": "skull and crossbones", "2497": "sky", "2498": "skyscraper", "2499": "skyscrapers", "2500": "slacks", "2501": "sled", "2502": "sleep", "2503": "sleeping", "2504": "sleeve", "2505": "sliced", "2506": "slide", "2507": "sliding", "2508": "slippers", "2509": "slope", "2510": "slow", "2511": "slow down", "2512": "small", "2513": "smaller", "2514": "smartphone", "2515": "smile", "2516": "smiley face", "2517": "smiling", "2518": "smoke", "2519": "smoking", "2520": "smooth", "2521": "smoothie", "2522": "snake", "2523": "sneakers", "2524": "sniffing", "2525": "snow", "2526": "snowboard", "2527": "snowboarder", "2528": "snowboarding", "2529": "snowboards", "2530": "snowflakes", "2531": "snowing", "2532": "snowsuit", "2533": "snowy", "2534": "soap", "2535": "soccer", "2536": "soccer ball", "2537": "soccer field", "2538": "socks", "2539": "soda", "2540": "sofa", "2541": "soft", "2542": "softball", "2543": "soldier", "2544": "soldiers", "2545": "solid", "2546": "someone", "2547": "sony", "2548": "sony ericsson", "2549": "soon", "2550": "soup", "2551": "south", "2552": "southwest", "2553": "space", "2554": "space needle", "2555": "space shuttle", "2556": "spaghetti", "2557": "spanish", "2558": "sparrow", "2559": "spatula", "2560": "speaker", "2561": "speakers", "2562": "spectators", "2563": "speed limit", "2564": "spices", "2565": "spider", "2566": "spiderman", "2567": "spinach", "2568": "spiral", "2569": "spoon", "2570": "spoons", "2571": "sports", "2572": "spots", "2573": "spotted", "2574": "spray paint", "2575": "spring", "2576": "sprinkles", "2577": "sprint", "2578": "sprite", "2579": "square", "2580": "squares", "2581": "squash", "2582": "squatting", "2583": "squirrel", "2584": "st patrick's day", "2585": "stability", "2586": "stadium", "2587": "stagecoach", "2588": "stained glass", "2589": "stainless steel", "2590": "stairs", "2591": "stand", "2592": "standing", "2593": "standing still", "2594": "stands", "2595": "star", "2596": "star alliance", "2597": "star wars", "2598": "starbucks", "2599": "staring", "2600": "stars", "2601": "state farm", "2602": "station", "2603": "statue", "2604": "statues", "2605": "steak", "2606": "steam", "2607": "steamed", "2608": "steel", "2609": "steeple", "2610": "steering wheel", "2611": "steps", "2612": "stew", "2613": "stick", "2614": "sticker", "2615": "stickers", "2616": "sticks", "2617": "still", "2618": "stir fry", "2619": "stomach", "2620": "stone", "2621": "stones", "2622": "stool", "2623": "stop", "2624": "stop light", "2625": "stop sign", "2626": "stopped", "2627": "stopping", "2628": "storage", "2629": "store", "2630": "stork", "2631": "storm", "2632": "stove", "2633": "straight", "2634": "straight ahead", "2635": "strap", "2636": "straw", "2637": "strawberries", "2638": "strawberry", "2639": "street", "2640": "street light", "2641": "street name", "2642": "street sign", "2643": "stretching", "2644": "strike", "2645": "string", "2646": "stripe", "2647": "striped", "2648": "stripes", "2649": "stroller", "2650": "stucco", "2651": "student", "2652": "students", "2653": "stuffed", "2654": "stuffed animal", "2655": "stuffed animals", "2656": "style", "2657": "styrofoam", "2658": "sub", "2659": "subway", "2660": "sugar", "2661": "suit", "2662": "suitcase", "2663": "suitcases", "2664": "suits", "2665": "summer", "2666": "sun", "2667": "sun hat", "2668": "sunbathing", "2669": "sunflower", "2670": "sunflowers", "2671": "sunglasses", "2672": "sunlight", "2673": "sunny", "2674": "sunrise", "2675": "sunset", "2676": "supreme", "2677": "surf", "2678": "surfboard", "2679": "surfboards", "2680": "surfer", "2681": "surfers", "2682": "surfing", "2683": "surprise", "2684": "surprised", "2685": "sushi", "2686": "suspenders", "2687": "suv", "2688": "suzuki", "2689": "swan", "2690": "swans", "2691": "sweat", "2692": "sweatband", "2693": "sweater", "2694": "sweatshirt", "2695": "sweet", "2696": "sweet potato", "2697": "swim", "2698": "swim trunks", "2699": "swimming", "2700": "swimsuit", "2701": "swing", "2702": "swinging", "2703": "swinging bat", "2704": "swirls", "2705": "swiss", "2706": "switzerland", "2707": "sydney", "2708": "syrup", "2709": "t", "2710": "t shirt", "2711": "t shirt and jeans", "2712": "tabby", "2713": "table", "2714": "tablecloth", "2715": "tables", "2716": "tablet", "2717": "tag", "2718": "tags", "2719": "tail", "2720": "take off", "2721": "taking off", "2722": "taking photo", "2723": "taking picture", "2724": "taking pictures", "2725": "taking selfie", "2726": "talking", "2727": "talking on phone", "2728": "tall", "2729": "taller", "2730": "tam", "2731": "tan", "2732": "tank", "2733": "tank top", "2734": "tape", "2735": "target", "2736": "tarmac", "2737": "tarp", "2738": "tater tots", "2739": "tattoo", "2740": "tattoos", "2741": "taxi", "2742": "tea", "2743": "teacher", "2744": "teal", "2745": "team", "2746": "teddy", "2747": "teddy bear", "2748": "teddy bears", "2749": "teeth", "2750": "telephone", "2751": "television", "2752": "tell time", "2753": "telling time", "2754": "tennis", "2755": "tennis ball", "2756": "tennis court", "2757": "tennis player", "2758": "tennis racket", "2759": "tennis rackets", "2760": "tennis racquet", "2761": "tennis shoes", "2762": "tent", "2763": "tents", "2764": "terrier", "2765": "texas", "2766": "texting", "2767": "thai", "2768": "thailand", "2769": "thanksgiving", "2770": "theater", "2771": "they aren't", "2772": "thick", "2773": "thin", "2774": "thomas", "2775": "thoroughbred", "2776": "thousands", "2777": "throw", "2778": "throw ball", "2779": "throw frisbee", "2780": "throwing", "2781": "throwing frisbee", "2782": "thumb", "2783": "thumbs up", "2784": "tiara", "2785": "tie", "2786": "tie dye", "2787": "ties", "2788": "tiger", "2789": "tigers", "2790": "tile", "2791": "tiled", "2792": "tiles", "2793": "tim hortons", "2794": "time", "2795": "tinkerbell", "2796": "tire", "2797": "tired", "2798": "tires", "2799": "tissue", "2800": "tissues", "2801": "to catch ball", "2802": "to catch frisbee", "2803": "to dry", "2804": "to eat", "2805": "to get to other side", "2806": "to hit ball", "2807": "to left", "2808": "to right", "2809": "to see", "2810": "toast", "2811": "toasted", "2812": "toaster", "2813": "toaster oven", "2814": "toilet", "2815": "toilet brush", "2816": "toilet paper", "2817": "toiletries", "2818": "toilets", "2819": "tokyo", "2820": "tomato", "2821": "tomatoes", "2822": "tongs", "2823": "tongue", "2824": "tools", "2825": "toothbrush", "2826": "toothbrushes", "2827": "toothpaste", "2828": "toothpick", "2829": "toothpicks", "2830": "top", "2831": "top hat", "2832": "top left", "2833": "top right", "2834": "toronto", "2835": "toshiba", "2836": "tour", "2837": "tourist", "2838": "tow", "2839": "tow truck", "2840": "toward", "2841": "towards", "2842": "towel", "2843": "towels", "2844": "tower", "2845": "towing", "2846": "town", "2847": "toy", "2848": "toyota", "2849": "toys", "2850": "track", "2851": "tracks", "2852": "tractor", "2853": "traffic", "2854": "traffic light", "2855": "traffic lights", "2856": "trailer", "2857": "train", "2858": "train car", "2859": "train station", "2860": "train tracks", "2861": "trains", "2862": "transport", "2863": "transportation", "2864": "trash", "2865": "trash can", "2866": "travel", "2867": "traveling", "2868": "tray", "2869": "tree", "2870": "tree branch", "2871": "trees", "2872": "triangle", "2873": "triangles", "2874": "trick", "2875": "tripod", "2876": "triumph", "2877": "trolley", "2878": "tropical", "2879": "tropicana", "2880": "truck", "2881": "trucks", "2882": "trunk", "2883": "trunks", "2884": "tub", "2885": "tube", "2886": "tugboat", "2887": "tulip", "2888": "tulips", "2889": "tuna", "2890": "tunnel", "2891": "turkey", "2892": "turn", "2893": "turn right", "2894": "turning", "2895": "turtle", "2896": "tusks", "2897": "tuxedo", "2898": "tv", "2899": "tv stand", "2900": "twin", "2901": "twins", "2902": "tying tie", "2903": "typing", "2904": "uk", "2905": "umbrella", "2906": "umbrellas", "2907": "umpire", "2908": "unclear", "2909": "under", "2910": "under armour", "2911": "under sink", "2912": "under table", "2913": "under tree", "2914": "uniform", "2915": "uniforms", "2916": "union station", "2917": "united", "2918": "united states", "2919": "unknown", "2920": "unsure", "2921": "up", "2922": "uphill", "2923": "upright", "2924": "ups", "2925": "upside down", "2926": "urban", "2927": "urinal", "2928": "urinals", "2929": "us", "2930": "us air force", "2931": "us airways", "2932": "us airways express", "2933": "us open", "2934": "usa", "2935": "used", "2936": "using computer", "2937": "using laptop", "2938": "utensils", "2939": "v", "2940": "vacation", "2941": "vaio", "2942": "valentine's day", "2943": "van", "2944": "vanilla", "2945": "vans", "2946": "vase", "2947": "vases", "2948": "vegetable", "2949": "vegetables", "2950": "vegetarian", "2951": "veggie", "2952": "veggies", "2953": "vehicles", "2954": "venice", "2955": "vent", "2956": "verizon", "2957": "vertical", "2958": "very", "2959": "very big", "2960": "very deep", "2961": "very fast", "2962": "very high", "2963": "very long", "2964": "very old", "2965": "very tall", "2966": "vest", "2967": "vests", "2968": "victoria", "2969": "victorian", "2970": "video", "2971": "video game", "2972": "vines", "2973": "virgin", "2974": "virgin atlantic", "2975": "visibility", "2976": "visilab", "2977": "visor", "2978": "volkswagen", "2979": "volleyball", "2980": "volvo", "2981": "w", "2982": "waffle", "2983": "wagon", "2984": "waiting", "2985": "wakeboard", "2986": "walgreens", "2987": "walk", "2988": "walking", "2989": "wall", "2990": "wall st", "2991": "wallet", "2992": "wallpaper", "2993": "war", "2994": "warm", "2995": "warmth", "2996": "warning", "2997": "washing", "2998": "washington", "2999": "washington dc", "3000": "washington monument", "3001": "watch", "3002": "watch tv", "3003": "watching", "3004": "watching tv", "3005": "water", "3006": "water bottle", "3007": "water ski", "3008": "water skiing", "3009": "water skis", "3010": "watermark", "3011": "watermelon", "3012": "wave", "3013": "waves", "3014": "waving", "3015": "wavy", "3016": "wax", "3017": "wax paper", "3018": "weather vane", "3019": "website", "3020": "wedding", "3021": "weeds", "3022": "welcome", "3023": "west", "3024": "western", "3025": "westin", "3026": "westjet", "3027": "wet", "3028": "wetsuit", "3029": "wetsuits", "3030": "whale", "3031": "wheat", "3032": "wheel", "3033": "wheelchair", "3034": "wheelie", "3035": "wheels", "3036": "whipped cream", "3037": "whirlpool", "3038": "white", "3039": "white and black", "3040": "white and blue", "3041": "white and brown", "3042": "white and gray", "3043": "white and green", "3044": "white and orange", "3045": "white and pink", "3046": "white and red", "3047": "white and yellow", "3048": "white house", "3049": "whole", "3050": "wicker", "3051": "wide", "3052": "wii", "3053": "wii controller", "3054": "wii controllers", "3055": "wii remote", "3056": "wii remotes", "3057": "wiimote", "3058": "wild", "3059": "wildebeest", "3060": "willow", "3061": "wilson", "3062": "wind", "3063": "windmill", "3064": "window", "3065": "window sill", "3066": "windows", "3067": "windowsill", "3068": "windsor", "3069": "windsurfing", "3070": "windy", "3071": "wine", "3072": "wine bottle", "3073": "wine glass", "3074": "wine glasses", "3075": "wine tasting", "3076": "wing", "3077": "wings", "3078": "winnie pooh", "3079": "winter", "3080": "wire", "3081": "wireless", "3082": "wires", "3083": "wisconsin", "3084": "woman", "3085": "woman's", "3086": "women", "3087": "women's", "3088": "wood", "3089": "wooden", "3090": "woodpecker", "3091": "woods", "3092": "wool", "3093": "words", "3094": "work", "3095": "working", "3096": "worms", "3097": "wreath", "3098": "wrist", "3099": "wristband", "3100": "writing", "3101": "x", "3102": "xbox", "3103": "y", "3104": "yacht", "3105": "yamaha", "3106": "yankees", "3107": "yard", "3108": "yarn", "3109": "years", "3110": "yellow", "3111": "yellow and black", "3112": "yellow and blue", "3113": "yellow and green", "3114": "yellow and orange", "3115": "yellow and red", "3116": "yellow and white", "3117": "yes", "3118": "yield", "3119": "yogurt", "3120": "young", "3121": "younger", "3122": "zebra", "3123": "zebra and giraffe", "3124": "zebras", "3125": "zig zag", "3126": "zipper", "3127": "zoo", "3128": "zucchini"}
app.py ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from io import BytesIO
2
+ import streamlit as st
3
+ import pandas as pd
4
+ import json
5
+ import os
6
+ import numpy as np
7
+ from model.flax_clip_vision_bert.modeling_clip_vision_bert import FlaxCLIPVisionBertForSequenceClassification
8
+ from utils import get_transformed_image, get_text_attributes, get_top_5_predictions, plotly_express_horizontal_bar_plot, translate_labels
9
+ import matplotlib.pyplot as plt
10
+ from mtranslate import translate
11
+ from PIL import Image
12
+
13
+
14
+ @st.cache
15
+ def load_model(ckpt):
16
+ return FlaxCLIPVisionBertForSequenceClassification.from_pretrained(ckpt)
17
+
18
+ def softmax(logits):
19
+ return np.exp(logits)/np.sum(np.exp(logits), axis=0)
20
+
21
+ checkpoints = ['./ckpt/ckpt-60k-5999'] # TODO: Maybe add more checkpoints?
22
+ dummy_data = pd.read_csv('dummy_vqa_multilingual.tsv', sep='\t')
23
+ with open('answer_reverse_mapping.json') as f:
24
+ answer_reverse_mapping = json.load(f)
25
+
26
+ # Init Session State
27
+ if 'image_file' not in st.session_state:
28
+ st.session_state.image_file = dummy_data.loc[0,'image_file']
29
+ st.session_state.question = dummy_data.loc[0,'question']
30
+ st.session_state.answer_label = dummy_data.loc[0,'answer_label']
31
+ st.session_state.question_lang_id = dummy_data.loc[0, 'lang_id']
32
+ st.session_state.answer_lang_id = dummy_data.loc[0, 'lang_id']
33
+
34
+ image_path = os.path.join('images',st.session_state.image_file)
35
+ image = plt.imread(image_path)
36
+ st.session_state.image = image
37
+
38
+ col1, col2 = st.beta_columns([5,5])
39
+ if col1.button('Get a Random Example'):
40
+ sample = dummy_data.sample(1).reset_index()
41
+ st.session_state.image_file = sample.loc[0,'image_file']
42
+ st.session_state.question = sample.loc[0,'question']
43
+ st.session_state.answer_label = sample.loc[0,'answer_label']
44
+ st.session_state.question_lang_id = sample.loc[0, 'lang_id']
45
+ st.session_state.answer_lang_id = sample.loc[0, 'lang_id']
46
+
47
+ image_path = os.path.join('images',st.session_state.image_file)
48
+ image = plt.imread(image_path)
49
+ st.session_state.image = image
50
+
51
+
52
+ uploaded_file = col2.file_uploader('Upload your image', type=['png','jpg','jpeg'])
53
+ if uploaded_file is not None:
54
+ st.session_state.image_file = os.path.join('images/val2014',uploaded_file.name)
55
+ st.session_state.image = np.array(Image.open(uploaded_file))
56
+
57
+
58
+ transformed_image = get_transformed_image(st.session_state.image)
59
+
60
+ # Display Image
61
+ st.image(st.session_state.image, use_column_width='always')
62
+
63
+ # Display Question
64
+ question = st.text_input(label="Question", value=st.session_state.question)
65
+ st.markdown(f"""**English Translation**: {question if st.session_state.question_lang_id == "en" else translate(question, 'en')}""")
66
+ question_inputs = get_text_attributes(question)
67
+
68
+ # Select Language
69
+ options = ['en', 'de', 'es', 'fr']
70
+ st.session_state.answer_lang_id = st.selectbox('Answer Language', index=options.index(st.session_state.answer_lang_id), options=options)
71
+ # Display Top-5 Predictions
72
+ with st.spinner('Loading model...'):
73
+ model = load_model(checkpoints[0])
74
+ with st.spinner('Predicting...'):
75
+ predictions = model(pixel_values = transformed_image, **question_inputs)
76
+ logits = np.array(predictions[0][0])
77
+ logits = softmax(logits)
78
+ labels, values = get_top_5_predictions(logits, answer_reverse_mapping)
79
+ translated_labels = translate_labels(labels, st.session_state.answer_lang_id)
80
+ fig = plotly_express_horizontal_bar_plot(values, translated_labels)
81
+ st.plotly_chart(fig)
ckpt/ckpt-60k-5999/config.json ADDED
The diff for this file is too large to render. See raw diff
ckpt/ckpt-60k-5999/flax_model.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f49d5f923ece9e5934a390d8c98abce35a11f207a8a750c6b9450a98c516c7ef
3
+ size 1032830156
ckpt/ckpt-60k-5999/opt_state.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd75d12313ab374bef6be1d5d33f11989b062f84d907a17aeae5e7cc9cfe1f81
3
+ size 2065660552
ckpt/ckpt-60k-5999/training_state.json ADDED
@@ -0,0 +1 @@
 
1
+ {"step": 6000}
dummy_vqa_multilingual.tsv ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ image_file question answer_label question_type lang_id
2
+ val2014/COCO_val2014_000000579060.jpg How many people are wearing hats? 3 how many people are en
3
+ val2014/COCO_val2014_000000574256.jpg Qu'est-ce que c'est? 1363 what kind of fr
4
+ val2014/COCO_val2014_000000084533.jpg - ¿El motociclista lleva casco, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh 1840 is the es
5
+ val2014/COCO_val2014_000000308441.jpg - ¿Qué clase de juego está viendo la multitud? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 2754 what kind of es
6
+ val2014/COCO_val2014_000000512416.jpg - ¿Cuantas velas hay en el cupcake? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 138 how many es
7
+ val2014/COCO_val2014_000000246717.jpg - ¿Puedes girar a la izquierda? - ¡A la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, a la izquierda, 3117 can you es
8
+ val2014/COCO_val2014_000000156532.jpg Y a-t-il un bateau sur le lac? 3117 is there a fr
9
+ val2014/COCO_val2014_000000351589.jpg - Haben die Fenster Bildschirme? 3117 do de
10
+ val2014/COCO_val2014_000000542024.jpg - ¿Qué equipo deportivo está a punto de batear? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 573 what es
11
+ val2014/COCO_val2014_000000467721.jpg Dans quel pays se trouve cette rue? 1208 what fr
12
+ val2014/COCO_val2014_000000412019.jpg Is the plane ascending or descending? 958 is the en
13
+ val2014/COCO_val2014_000000435321.jpg Is the woman wearing comfortable shoes? 1840 is the woman en
14
+ val2014/COCO_val2014_000000261758.jpg - ¿De qué está hecha la pared? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 3088 what is the es
15
+ val2014/COCO_val2014_000000363645.jpg - ¿A cuánta gente le están tomando una foto? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 138 how many people are es
16
+ val2014/COCO_val2014_000000242610.jpg Pourriez-vous vous lever de ce lit? 1840 could fr
17
+ val2014/COCO_val2014_000000382729.jpg Ist der Ball auf dem Weg? 3117 is the de
18
+ val2014/COCO_val2014_000000558834.jpg Quel sport utilise ce ballon? 2535 what fr
19
+ val2014/COCO_val2014_000000473299.jpg Would it be hot or cold where the people are in this picture? 818 none of the above en
20
+ val2014/COCO_val2014_000000095805.jpg Do they like each other? 3117 do en
21
+ val2014/COCO_val2014_000000000757.jpg - ¿Cuántos elefantes hay en esta foto? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 138 how many es
22
+ val2014/COCO_val2014_000000546596.jpg - Sind das drei Teddybären? 3117 are these de
23
+ val2014/COCO_val2014_000000370929.jpg Qu'est-ce qu'il y a sur les rails? 2857 what is on the fr
24
+ val2014/COCO_val2014_000000277440.jpg Y a-t-il un livre ouvert dans l'image? 3117 is there fr
25
+ val2014/COCO_val2014_000000066973.jpg De quelle couleur sont les pots de fleurs? 591 what color are the fr
26
+ val2014/COCO_val2014_000000441500.jpg Quelle année est indiquée sur le devant de sa chemise? 1 what fr
27
+ val2014/COCO_val2014_000000090255.jpg Is the woman drinking tea? 1840 is the woman en
28
+ val2014/COCO_val2014_000000209602.jpg - ¿Se lo tomaron en invierno? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 3117 was es
29
+ val2014/COCO_val2014_000000017018.jpg What color is this person's coat? 510 what color is en
30
+ val2014/COCO_val2014_000000354978.jpg De quelle couleur est la robe de la fille? 2261 what color is the fr
31
+ val2014/COCO_val2014_000000548843.jpg - ¿El niño está eligiendo un snack saludable? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 1840 is the es
32
+ val2014/COCO_val2014_000000099177.jpg What color socks is the girl on the ground wearing ?L? 0 what color en
33
+ val2014/COCO_val2014_000000109316.jpg Trägt diese Dame einen Regenschirm? 3117 is this de
34
+ val2014/COCO_val2014_000000547047.jpg What is the point in this sign? 0 what is the en
35
+ val2014/COCO_val2014_000000167353.jpg Does this clock have a second hand? 1840 does this en
36
+ val2014/COCO_val2014_000000015517.jpg What type of business is in the photo? 619 what type of en
37
+ val2014/COCO_val2014_000000036607.jpg Schreibt die Person SMS? 1840 is the person de
38
+ val2014/COCO_val2014_000000313386.jpg Welcher Buchstabe wird fast ganz links vom Rahmen abgeschnitten? 635 what de
39
+ val2014/COCO_val2014_000000050165.jpg What purpose does the train car serve now? 2862 what en
40
+ val2014/COCO_val2014_000000144935.jpg Où cette photo est-elle prise? 435 none of the above fr
41
+ val2014/COCO_val2014_000000163348.jpg - Welche Farbe hat der Laptop? 3038 what color is the de
42
+ val2014/COCO_val2014_000000381576.jpg Combien de bancs? 95 how many fr
43
+ val2014/COCO_val2014_000000013466.jpg Le ciel est-il beau? 3117 is the fr
44
+ val2014/COCO_val2014_000000481480.jpg What are the men riding? 1423 what are the en
45
+ val2014/COCO_val2014_000000068411.jpg De quelle couleur est l'ours supérieur? 510 what color is the fr
46
+ val2014/COCO_val2014_000000314907.jpg - ¿Cuántos carriles hay en este camino? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 95 how many es
47
+ val2014/COCO_val2014_000000230561.jpg Was this taken inside or outside? 1488 was en
48
+ val2014/COCO_val2014_000000032901.jpg Combien y a-t-il? 164 how many people are fr
49
+ val2014/COCO_val2014_000000117328.jpg - Welche Farbe haben die Schuhe? 3038 what color are the de
50
+ val2014/COCO_val2014_000000308441.jpg What kind of game is the crowd watching? 2754 what kind of en
51
+ val2014/COCO_val2014_000000109055.jpg - ¿El gato está ronroneando? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 1840 is the es
52
+ val2014/COCO_val2014_000000410231.jpg Est-ce que l'œil de la girafe est fermé? 1840 is the fr
53
+ val2014/COCO_val2014_000000023899.jpg - ¿A cuántas personas se les muestra? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, ¡no, no, no, no, no, no, no, no, no, no, 138 how many people are es
54
+ val2014/COCO_val2014_000000197716.jpg Wie viele Busse gibt es auf der Straße? 3 how many de
55
+ val2014/COCO_val2014_000000526359.jpg What is the name of some stores would you be able to buy some of that furniture in? 0 what is the name en
56
+ val2014/COCO_val2014_000000079837.jpg Sieht die Sonne mit diesen Vögeln aus? 1840 is the de
57
+ val2014/COCO_val2014_000000433499.jpg Ist das ein Pendler-Zug? 1840 is this a de
58
+ val2014/COCO_val2014_000000171255.jpg Quelle est la caractéristique du terrain en arrière-plan? 435 what fr
59
+ val2014/COCO_val2014_000000187565.jpg What shape is the vase? 2330 what en
60
+ val2014/COCO_val2014_000000046269.jpg What color is the car? 488 what color is the en
61
+ val2014/COCO_val2014_000000153011.jpg Was sind die beiden Buchstaben auf dem grünen Teil des Trucks? 0 what are the de
62
+ val2014/COCO_val2014_000000117374.jpg De quelle couleur est l'oiseau? 488 what color is the fr
63
+ val2014/COCO_val2014_000000519744.jpg Sieht der Mann auf dem Bild glücklich aus? 3117 does the de
64
+ val2014/COCO_val2014_000000567290.jpg Is the store busy? 1840 is the en
65
+ val2014/COCO_val2014_000000563050.jpg - ¿Dónde está el agua? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 0 where is the es
66
+ val2014/COCO_val2014_000000518348.jpg Est-ce que ces animaux sont contenus? 1840 are these fr
67
+ val2014/COCO_val2014_000000303543.jpg Are the men wearing bodysuits? 1840 are the en
68
+ val2014/COCO_val2014_000000379402.jpg Elle est dans un restaurant? 1840 is fr
69
+ val2014/COCO_val2014_000000541351.jpg What color is the ground? 1305 what color is the en
70
+ val2014/COCO_val2014_000000260094.jpg L'avion est-il à l'intérieur ou à l'extérieur? 1989 is the fr
71
+ val2014/COCO_val2014_000000460927.jpg How many bears do you see? 3 how many en
72
+ val2014/COCO_val2014_000000190585.jpg Are the utensils in the photo? 3117 are the en
73
+ val2014/COCO_val2014_000000490320.jpg - ¿Estos teléfonos son tapas rotativas? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 3117 are these es
74
+ val2014/COCO_val2014_000000121031.jpg - ¿A qué profundidad está el agua? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 166 how es
75
+ val2014/COCO_val2014_000000074646.jpg Gibt es auf diesem Bild einen oder mehrere Drachen? 3 is there de
76
+ val2014/COCO_val2014_000000566414.jpg - Welche Farbe hat das Gebäude? 2261 what color is the de
77
+ val2014/COCO_val2014_000000564851.jpg What color is the toilet? 3038 what color is the en
78
+ val2014/COCO_val2014_000000577223.jpg - ¿Para qué cadena de hamburguesas conduce este camión? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 0 what es
79
+ val2014/COCO_val2014_000000375765.jpg - ¿Y las jirafas, qué hacen? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 1064 what are the es
80
+ val2014/COCO_val2014_000000321437.jpg Is the clock ornate? 3117 is the en
81
+ val2014/COCO_val2014_000000298994.jpg Are there any fruits among the vegetables? 1840 are there any en
82
+ val2014/COCO_val2014_000000029465.jpg What is the ground covered in? 2525 what is the en
83
+ val2014/COCO_val2014_000000190689.jpg Quelle partie du snowboard est touchée? 1224 what fr
84
+ val2014/COCO_val2014_000000151393.jpg Are any of the signs the same? 1840 are en
85
+ val2014/COCO_val2014_000000486046.jpg How many zebras are there? 3 how many en
86
+ val2014/COCO_val2014_000000163640.jpg Is this person happy? 3117 is this person en
87
+ val2014/COCO_val2014_000000560757.jpg - Wie kalt ist das Wasser? 0 how de
88
+ val2014/COCO_val2014_000000111546.jpg Ist das Gebäude orniert? 3117 is the de
89
+ val2014/COCO_val2014_000000132042.jpg - ¿De qué color es su camisa? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 1300 what color is es
90
+ val2014/COCO_val2014_000000488743.jpg Worauf sitzt dieses orangefarbene Ding? 1345 what is this de
91
+ val2014/COCO_val2014_000000300000.jpg Was guckt der Hund denn so? 1141 what is the de
92
+ val2014/COCO_val2014_000000295574.jpg - ¿A qué habitación perteneces, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, eh, 429 which es
93
+ val2014/COCO_val2014_000000416331.jpg - ¿De qué color es el tren en el camino correcto? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no 1300 what color is the es
94
+ val2014/COCO_val2014_000000238537.jpg Ist das Brett aus Holz gefertigt? 3117 is the de
95
+ val2014/COCO_val2014_000000208971.jpg La plante est-elle grande? 3117 is the fr
96
+ val2014/COCO_val2014_000000169143.jpg Is the player running? 3117 is the en
97
+ val2014/COCO_val2014_000000570171.jpg - ¿Qué otros animales hay en esta foto? - ¡No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, 3059 what es
98
+ val2014/COCO_val2014_000000552074.jpg Was befindet sich in der rechten Ecke des Fußbodens? 2814 what is in the de
99
+ val2014/COCO_val2014_000000568107.jpg - ¿La moto está parada, o va hacia delante, sin moverse, o qué, sin moverse, o qué, sin moverse, no va hacia delante, o qué, sin moverse, ¡no, no, no, no, no, no, no, no, no, no, no, no 0 is the es
100
+ val2014/COCO_val2014_000000499374.jpg Regnet es denn? 1840 is it de
101
+ val2014/COCO_val2014_000000270381.jpg Where is the bench? 1134 where is the en
images/val2014/COCO_val2014_000000000757.jpg ADDED
images/val2014/COCO_val2014_000000013466.jpg ADDED
images/val2014/COCO_val2014_000000015517.jpg ADDED
images/val2014/COCO_val2014_000000017018.jpg ADDED
images/val2014/COCO_val2014_000000023899.jpg ADDED
images/val2014/COCO_val2014_000000029465.jpg ADDED
images/val2014/COCO_val2014_000000032901.jpg ADDED
images/val2014/COCO_val2014_000000036607.jpg ADDED
images/val2014/COCO_val2014_000000046269.jpg ADDED
images/val2014/COCO_val2014_000000050165.jpg ADDED
images/val2014/COCO_val2014_000000066973.jpg ADDED
images/val2014/COCO_val2014_000000068411.jpg ADDED
images/val2014/COCO_val2014_000000074646.jpg ADDED
images/val2014/COCO_val2014_000000079837.jpg ADDED
images/val2014/COCO_val2014_000000084533.jpg ADDED
images/val2014/COCO_val2014_000000090255.jpg ADDED
images/val2014/COCO_val2014_000000095805.jpg ADDED
images/val2014/COCO_val2014_000000099177.jpg ADDED
images/val2014/COCO_val2014_000000109055.jpg ADDED
images/val2014/COCO_val2014_000000109316.jpg ADDED
images/val2014/COCO_val2014_000000111546.jpg ADDED
images/val2014/COCO_val2014_000000117328.jpg ADDED
images/val2014/COCO_val2014_000000117374.jpg ADDED
images/val2014/COCO_val2014_000000121031.jpg ADDED
images/val2014/COCO_val2014_000000132042.jpg ADDED
images/val2014/COCO_val2014_000000144935.jpg ADDED
images/val2014/COCO_val2014_000000151393.jpg ADDED
images/val2014/COCO_val2014_000000153011.jpg ADDED
images/val2014/COCO_val2014_000000156532.jpg ADDED
images/val2014/COCO_val2014_000000163348.jpg ADDED
images/val2014/COCO_val2014_000000163640.jpg ADDED
images/val2014/COCO_val2014_000000167353.jpg ADDED
images/val2014/COCO_val2014_000000169143.jpg ADDED
images/val2014/COCO_val2014_000000171255.jpg ADDED
images/val2014/COCO_val2014_000000187565.jpg ADDED
images/val2014/COCO_val2014_000000190585.jpg ADDED
images/val2014/COCO_val2014_000000190689.jpg ADDED
images/val2014/COCO_val2014_000000197716.jpg ADDED
images/val2014/COCO_val2014_000000208971.jpg ADDED
images/val2014/COCO_val2014_000000209602.jpg ADDED
images/val2014/COCO_val2014_000000230561.jpg ADDED
images/val2014/COCO_val2014_000000238537.jpg ADDED