Inference error in line 336, in resume model.load_state_dict(checkpoint['state_dict'] about superpoint_graph

[loicland]

Hi,

SPG is still only compatible with pyorch 3.1. We are working on it and will upgrade soon. In the meantime, I advise you to create a virtual env with 3.1.

Loic

[chowkamlee81]

Kindly help
Downgraded to version pytorch3.1 rather than 4.0. But got an error below.
0%| | 0/14 [00:00<?, ?it/s]Traceback (most recent call last):
File "learning/main.py", line 387, in
main()
File "learning/main.py", line 311, in main
acc_test, oacc_test, avg_iou_test, per_class_iou_test, predictions_test, avg_acc_test, confusion_matrix = eval_final()
File "learning/main.py", line 254, in eval_final
outputs = model.ecc(embeddings)
File "/home/administrator/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 357, in call
result = self.forward(*input, **kwargs)
File "/home/administrator/Documents.orig/FileSystem/Server_Code/sematic3d/superpoint_graph-release/learning/graphnet.py", line 97, in forward
input = module(input)
File "/home/administrator/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 357, in call
result = self.forward(*input, **kwargs)
File "/home/administrator/Documents.orig/FileSystem/Server_Code/sematic3d/superpoint_graph-release/learning/modules.py", line 54, in forward
input = ecc.GraphConvFunction(nc, nc, idxn, idxe, degs, degs_gpu, self._edge_mem_limit)(hx, weights)
File "/home/administrator/Documents.orig/FileSystem/Server_Code/sematic3d/superpoint_graph-release/learning/ecc/GraphConvModule.py", line 67, in forward
cuda_kernels.conv_aggregate_fw(output.narrow(0,startd,numd), products.view(-1,self._out_channels), self._degs_gpu.narrow(0,startd,numd))
File "/home/administrator/Documents.orig/FileSystem/Server_Code/sematic3d/superpoint_graph-release/learning/ecc/cuda_kernels.py", line 120, in conv_aggregate_fw
function, stream = get_kernel_func('conv_aggregate_fw_kernel_v2', conv_aggregate_fw_kernel_v2(), get_dtype(src))
File "/home/administrator/Documents.orig/FileSystem/Server_Code/sematic3d/superpoint_graph-release/learning/ecc/cuda_kernels.py", line 33, in get_kernel_func
prog = Program(ksrc.encode('utf-8'), (kname+dtype+'.cu').encode('utf-8'))
File "/home/administrator/anaconda3/lib/python3.6/site-packages/pynvrtc/compiler.py", line 52, in init
include_names)
File "/home/administrator/anaconda3/lib/python3.6/site-packages/pynvrtc/interface.py", line 200, in nvrtcCreateProgram
c_char_p(encode_str(src)), c_char_p(encode_str(name)),
File "/home/administrator/anaconda3/lib/python3.6/site-packages/pynvrtc/interface.py", line 54, in encode_str
return s.encode("utf-8")
AttributeError: 'bytes' object has no attribute 'encode'
Exception ignored in: <bound method Program.del of <pynvrtc.compiler.Program object at 0x7f828338c4a8>>
Traceback (most recent call last):
File "/home/administrator/anaconda3/lib/python3.6/site-packages/pynvrtc/compiler.py", line 56, in del
self._interface.nvrtcDestroyProgram(self._program)
AttributeError: 'Program' object has no attribute '_program'
^Z
[6]+ Stopped CUDA_VISIBLE_DEVICES=0,1 python learning/main.py --dataset sema3d --SEMA3D_PATH /home/administrator/Documents.orig/FileSystem/Server_Code/sematic3d --db_test_name testfull --db_train_name trainval --epochs -1 --lr_steps '[350, 400, 450]' --test_nth_epoch 100 --model_config 'gru_10,f_8' --ptn_nfeat_stn 11 --nworkers 2 --odir "/home/administrator/Documents.orig/FileSystem/results/sema3d/trainval_best" --resume RESUME

[chowkamlee81]

solved by using
replacing line 3̶5̶ 33 of /learning/ecc/cuda_kernels.py by

prog = Program(ksrc, (kname+dtype+'.cu'))

Closing the issue

[mys007]

This was caused by recently merged NVIDIA/pynvrtc#2 and fixed here today - 44bcbe8