leVirve/lsun-room

from trainer.賣扣老師 import build_resnet101_FCN

pjspillai opened this issue · 7 comments

pjspillai commented

Getting Error:

Can't find trainer.賣扣老師

leVirve commented

Duplicated with #4.

Some files have been modified, and we will update the code base as we're permitted. But, you now can try to replace them with

# Old
from trainer.賣扣老師 import build_resnet101_FCN
model = build_resnet101_FCN(pretrained=False, nb_classes=37, stage_2=True, joint_class=joint_class)

# Use this instead
from trainer.model import ResPlanarSeg
model = ResPlanarSeg(num_classes=5, pretrained=False)
pjspillai commented

I did that, now I am getting a different error:

File "/home/ppillai/anaconda3/envs/torch/lib/python3.6/site-packages/torch/serialization.py", line 172, in raise_err_msg
raise type(e)(msg)
AttributeError: 'NoneType' object has no attribute 'seek'. You can only torch.load from a file that is seekable. Please pre-load the data into a buffer like io.BytesIO and try to load from it instead.

leVirve commented

Seems that you run the script without argument weight_path, and you need to train the network first.

SpencerKD commented

I want to train the network, but unfortunately，I cant find the LSUN DataSet anywhere. So can you upload the pretrained model or the LSUN dataset for the semantic segmentation，many thanks

  • 👍1
huangshizhong commented

I want to train the network, but unfortunately，I cant find the LSUN DataSet anywhere. So can you upload the pretrained model or the LSUN dataset for the semantic segmentation，many thanks

do you solve this problem ,i have the same error,i would be very grateful for you to tell me how to solve this problem

leVirve commented

Unfortunately, the challenge website is down 😢 , am trying to find other resources for that.

leVirve commented

Hi guys, please refer to the updated readme for the dataset https://github.com/leVirve/lsun-room#dataset