card metadata: t2

meta/t2.json

@@ -0,0 +1 @@
+{"assets": "assets/t2", "category_names": ["aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor", "truck", "traffic light", "fire hydrant", "stop sign", "parking meter", "bench", "elephant", "bear", "zebra", "giraffe", "backpack", "umbrella", "handbag", "tie", "suitcase", "microwave", "oven", "toaster", "sink", "refrigerator", "frisbee", "skis", "snowboard", "sports ball", "kite", "baseball bat", "baseball glove", "skateboard", "surfboard", "tennis racket", "banana", "apple", "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut", "cake", "bed", "toilet", "laptop", "mouse", "remote", "keyboard", "cell phone", "book", "clock", "vase", "scissors", "teddy bear", "hair drier", "toothbrush", "wine glass", "cup", "fork", "knife", "spoon", "bowl"], "config": "t2", "default": false, "n_gallery": 4, "notes": "One config per OWOD task (t1-t4): train = that task's training image list, test = the common all-task test list (duplicate list lines collapsed). The 80-class vocabulary is task-ordered (ids 0-19 = t1/VOC ... 60-79 = t4) and identical across configs; per config, boxes are restricted to the classes known at that task. Task 2: boxes limited to the 40 classes known at task 2 (ids 0-39). Note: the strict OWOD training protocol additionally hides the classes of earlier tasks at train time \u2014 reproduce it by dropping train boxes with category id < 20*(K-1). The protocol's 'unknown' class is not materialized.", "orig_ids": null, "splits": {"test": 9119, "train": 45520}}