aRI0U / RandLA-Net-pytorch

PyTorch implementation of RandLA-Net

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

about result

Taousst opened this issue · comments

=== EPOCH 50/50 ===
Training loss: 2.6826021 Validation loss: 2.5952491
Accuracy | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | OA
Training: | 0.519 | 0.139 | 0.128 | 0.360 | 0.269 | 0.003 | 0.003 | 0.142 | 0.234 | 0.124 | 0.017 | 0.002 | 0.017 | 0.006 | 0.273
Validation: | 0.767 | 0.027 | 0.043 | 0.116 | 0.139 | 0.000 | 0.000 | 0.152 | 0.199 | 0.075 | 0.000 | 0.000 | 0.010 | 0.000 | 0.197
IoU | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | mIoU
Training: | 0.369 | 0.058 | 0.044 | 0.259 | 0.170 | 0.000 | 0.000 | 0.032 | 0.055 | 0.026 | 0.002 | 0.000 | 0.002 | 0.001 | 0.073
Validation: | 0.281 | 0.021 | 0.019 | 0.088 | 0.081 | 0.000 | 0.000 | 0.028 | 0.039 | 0.025 | 0.000 | 0.000 | 0.003 | 0.000 | 0.054

Why was my training so bad? What went wrong

haoxiongdi,niyetaicanle,huangemoxingpaoquba...