diff options
Diffstat (limited to 'artifacts/vg_5k/natural_families.json')
| -rw-r--r-- | artifacts/vg_5k/natural_families.json | 570 |
1 files changed, 570 insertions, 0 deletions
diff --git a/artifacts/vg_5k/natural_families.json b/artifacts/vg_5k/natural_families.json new file mode 100644 index 0000000..a823eb2 --- /dev/null +++ b/artifacts/vg_5k/natural_families.json @@ -0,0 +1,570 @@ +{ + "protocol": "Families are recovered from region descriptions alone by mutual exclusivity plus distributional similarity. Reference word sets are read only to label the result.", + "phrases": 80000, + "vocabulary": 271, + "families": [ + { + "size": 17, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "plane", + "horse", + "elephant", + "pizza", + "truck", + "boat", + "motorcycle", + "sheep", + "computer", + "surfboard", + "phone", + "bathroom", + "kite", + "vase" + ] + }, + { + "size": 13, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "bench", + "line", + "toilet", + "eye", + "nose", + "tracks", + "helmet", + "number", + "mirror", + "desk", + "couch", + "logo", + "mountain" + ] + }, + { + "size": 11, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "man", + "person", + "building", + "train", + "cat", + "bear", + "giraffe", + "boy", + "zebra", + "cow", + "tire" + ] + }, + { + "size": 10, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "woman", + "window", + "tree", + "plate", + "girl", + "player", + "windows", + "letter", + "traffic", + "tile" + ] + }, + { + "size": 10, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "car", + "pole", + "trees", + "people", + "bed", + "arm", + "cake", + "cup", + "airplane", + "neck" + ] + }, + { + "size": 8, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "s", + "wall", + "sky", + "table", + "street", + "leaves", + "beach", + "ceiling" + ] + }, + { + "size": 8, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "sitting", + "head", + "door", + "field", + "jacket", + "face", + "wheel", + "roof" + ] + }, + { + "size": 8, + "label": "number", + "purity": 0.125, + "words": [ + "one", + "glasses", + "cap", + "painted", + "reflection", + "stone", + "suit", + "bicycle" + ] + }, + { + "size": 7, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "shirt", + "sign", + "water", + "clouds", + "hand", + "seat", + "kitchen" + ] + }, + { + "size": 7, + "label": "colour", + "purity": 0.14285714285714285, + "words": [ + "light", + "grass", + "snow", + "clock", + "skateboard", + "paper", + "keyboard" + ] + }, + { + "size": 7, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "side", + "ground", + "top", + "tail", + "leg", + "eyes", + "hands" + ] + }, + { + "size": 7, + "label": "size", + "purity": 0.14285714285714285, + "words": [ + "bike", + "food", + "short", + "ski", + "lamp", + "hydrant", + "teddy" + ] + }, + { + "size": 6, + "label": "colour", + "purity": 0.16666666666666666, + "words": [ + "orange", + "down", + "dirt", + "ocean", + "tower", + "house" + ] + }, + { + "size": 6, + "label": "material", + "purity": 0.16666666666666666, + "words": [ + "metal", + "dog", + "hair", + "pants", + "shorts", + "bag" + ] + }, + { + "size": 6, + "label": "size", + "purity": 0.16666666666666666, + "words": [ + "fence", + "left", + "bowl", + "big", + "lines", + "striped" + ] + }, + { + "size": 6, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "chair", + "for", + "hat", + "trunk", + "wave", + "flying" + ] + }, + { + "size": 6, + "label": "material", + "purity": 0.16666666666666666, + "words": [ + "picture", + "wood", + "ear", + "umbrella", + "ball", + "brick" + ] + }, + { + "size": 6, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "round", + "handle", + "shoes", + "court", + "shoe", + "stop" + ] + }, + { + "size": 6, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "piece", + "bottle", + "made", + "pair", + "body", + "group" + ] + }, + { + "size": 6, + "label": "number", + "purity": 0.3333333333333333, + "words": [ + "some", + "flower", + "bright", + "child", + "baby", + "four" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "on", + "in", + "next", + "over", + "around" + ] + }, + { + "size": 5, + "label": "size", + "purity": 0.4, + "words": [ + "small", + "bus", + "tall", + "baseball", + "racket" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "back", + "road", + "background", + "floor", + "sink" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "at", + "bird", + "hanging", + "laying", + "writing" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "by", + "under", + "distance", + "photo", + "room" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "it", + "sidewalk", + "his", + "sand", + "counter" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "board", + "parked", + "from", + "box", + "covered" + ] + }, + { + "size": 5, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "that", + "edge", + "above", + "legs", + "mouth" + ] + }, + { + "size": 5, + "label": "material", + "purity": 0.2, + "words": [ + "plastic", + "post", + "laptop", + "area", + "outside" + ] + }, + { + "size": 5, + "label": "colour", + "purity": 0.2, + "words": [ + "purple", + "out", + "there", + "young", + "old" + ] + }, + { + "size": 4, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "is", + "with", + "are", + "has" + ] + }, + { + "size": 4, + "label": "colour", + "purity": 0.5, + "words": [ + "black", + "green", + "wooden", + "playing" + ] + }, + { + "size": 4, + "label": "colour", + "purity": 0.75, + "words": [ + "red", + "brown", + "standing", + "pink" + ] + }, + { + "size": 4, + "label": "colour", + "purity": 0.5, + "words": [ + "yellow", + "an", + "tan", + "jeans" + ] + }, + { + "size": 4, + "label": "size", + "purity": 0.25, + "words": [ + "large", + "part", + "holding", + "lights" + ] + }, + { + "size": 4, + "label": "colour", + "purity": 0.75, + "words": [ + "gray", + "silver", + "grey", + "riding" + ] + }, + { + "size": 4, + "label": "size", + "purity": 0.25, + "words": [ + "right", + "long", + "fire", + "men" + ] + }, + { + "size": 4, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "walking", + "colored", + "shadow", + "open" + ] + }, + { + "size": 3, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "of", + "wearing", + "looking" + ] + }, + { + "size": 3, + "label": "colour", + "purity": 0.3333333333333333, + "words": [ + "blue", + "glass", + "air" + ] + }, + { + "size": 3, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "to", + "behind", + "near" + ] + }, + { + "size": 3, + "label": "unlabelled", + "purity": 0.0, + "words": [ + "front", + "tennis", + "flowers" + ] + }, + { + "size": 3, + "label": "colour", + "purity": 0.3333333333333333, + "words": [ + "dark", + "up", + "clear" + ] + } + ], + "recovered_labels": { + "unlabelled": 24, + "number": 2, + "colour": 9, + "size": 5, + "material": 3 + } +}
\ No newline at end of file |
