search

ApolloScapeAuto / dataset-api

The ApolloScape Open Dataset for Autonomous Driving and its Application.
http://apolloscape.auto/
Apache License 2.0
569 stars 140 forks source link

eval.py in 3d object detection doesn't work #108

Open Tartisan opened 2 years ago

Tartisan commented 2 years ago

I used part of train dataset to evalute my detection results, but it seems doesn't work.

ValueError
cannot reshape array of size 35 into shape (6,5)
  File "/home/tartisan/work/github/dataset-api/3d_detection_tracking/eval.py", line 334, in assignPair
    subPair = pair[mask].reshape(subShape)
  File "/home/tartisan/work/github/dataset-api/3d_detection_tracking/eval.py", line 373, in seekBijection
    framePairs[:gtNum, dtInds] = np.multiply(assignPair(score, pair), ~invalidInds)
  File "/home/tartisan/work/github/dataset-api/3d_detection_tracking/eval.py", line 456, in evaluate
    typeFilterFlag, gtCubeTypes[objType], dtCubeTypes[objType])
  File "/home/tartisan/work/github/dataset-api/3d_detection_tracking/eval.py", line 488, in main
    evaluate(args.modeType, args.dataType, dtPath, evalList, args.objTypes, args.minScores, objIdFlag, args.minIoUs, args.typeFilterFlag, gts, args.apSampleNum)
  File "/home/tartisan/work/github/dataset-api/3d_detection_tracking/eval.py", line 493, in <module>
    main()