I am training the model on custom dataset with 7 labels.
I have done the changes required for custom dataset as mentioned in the previous issues.
But when I change from CLASS_IDS_WITH_INSTANCES = [1,2] in KITTI_segtrack.py to CLASS_IDS_WITH_INSTANCES = [1,2,3,4,5,6,7], get error as follows-
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[11] = [11, 1] does not index into param shape [12,1,4] [[node trainnet/tower_gpu_0/frcnn_3/GatherNd (defined at /gpfs/home/niraj.dere/Projects/TrackR-CNN/network/FasterRCNN_utils.py:392) ]]
The training might take about 2 days to complete.
Would you please confirm that is there a need to change the variable CLASS_IDS_WITH_INSTANCES = [1,2].
I am training the model on custom dataset with 7 labels.
I have done the changes required for custom dataset as mentioned in the previous issues.
But when I change from CLASS_IDS_WITH_INSTANCES = [1,2] in KITTI_segtrack.py to CLASS_IDS_WITH_INSTANCES = [1,2,3,4,5,6,7], get error as follows-
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[11] = [11, 1] does not index into param shape [12,1,4] [[node trainnet/tower_gpu_0/frcnn_3/GatherNd (defined at /gpfs/home/niraj.dere/Projects/TrackR-CNN/network/FasterRCNN_utils.py:392) ]]The training might take about 2 days to complete.
Would you please confirm that is there a need to change the variable CLASS_IDS_WITH_INSTANCES = [1,2].