diffConvNet
diffConvNet copied to clipboard
TypeError: sequence item 0: expected str instance, int found
(diffconv)root@ailab-PowerEdge-T640:/home/ailab/diffconvNet#pvthon3 data_prep.py --dataset=toronto3d
Traceback (most recent call last):
File "data_prep.py", line 104, in
how to solve the question?
Hi, I didn't encounter this error before. But it seems that it failed to load open3d. In 'match_torch_ver = '.'.join( o3d torch version.release[:2] + ('*',))', o3d_torch_version.release[:2] is an integer, which should be a string. You may want to change the type of it by str()? This is an open3D error thus please also check if open3d was installed correctly:)
can you tell me the required open3d version
I am using open3d 0.15.2
Is this the "processing 63.65% points in region 2" when running? It is so slow.
Yes, I am afraid it is slow.
I also encountered problems raised by others during the Scene segmentation,many values remain unchanged,Does it change to change the default super parameter,i may try it
Train 152, loss: 1.704556, train acc: 0.579693, train avg acc: 0.111111, train iou: 0.072462, mean iou: 0.716269 Test 152, loss: 1.720625, test acc: 0.545929, test avg acc: 0.111111, test iou: 0.068241, mean iou: 0.693581 Train 153, loss: 1.699433, train acc: 0.584002, train avg acc: 0.111111, train iou: 0.073000, mean iou: 0.716085 Test 153, loss: 1.721535, test acc: 0.545929, test avg acc: 0.111111, test iou: 0.068241, mean iou: 0.693581 Train 154, loss: 1.701942, train acc: 0.581193, train avg acc: 0.111111, train iou: 0.072649, mean iou: 0.716602 Test 154, loss: 1.719878, test acc: 0.545929, test avg acc: 0.111111, test iou: 0.068241, mean iou: 0.693581 Train 155, loss: 1.699353, train acc: 0.582737, train avg acc: 0.111111, train iou: 0.072842, mean iou: 0.716361 Test 155, loss: 1.722167, test acc: 0.545929, test avg acc: 0.111111, test iou: 0.068241, mean iou: 0.693581
Hi, I am sorry for the problem. As you may notice from the other post I failed to reproduce the error when I ran the experiments from scratch and I have no idea about this.
Hi, if you are still interested please check issue #7