Closed andreac closed 5 years ago
I think the problem is about the configuration of pythorch e torchnet. I'm using Ubuntu 18.04 Cuda 9.2. I'm tring different setup. Now i've intalled pytorch 0.3 and torchnet with pip install torchnet.
I have the following error:
Traceback (most recent call last): File "learning/main.py", line 387, in <module> main() File "learning/main.py", line 286, in main acc, loss, oacc, avg_iou = train() File "learning/main.py", line 199, in train acc_meter.add(o_cpu, t_cpu) File "/home/spxstrix/DATI/ENEL/retineurali/superpoint/lib/python3.6/site-packages/torchnet/meter/classerrormeter.py", line 38, in add pred = torch.from_numpy(output).topk(maxk, 1, True, True)[1].numpy() RuntimeError: the given numpy array has zero-sized dimensions. Zero-sized dimensions are not supported in PyTorch
Hi, the problem is about the dataset i'm using. i have adataset based on txt but whitout rgb value. The problems depends on it
Hi andreac, I met the same problem. Could you give some suggestions on how to solve it?
I got the following error:
RuntimeError: the given numpy array has zero-sized dimensions. Zero-sized dimensions are not supported in PyTorch
Hi. Can you give more context? Is the error happening line 199 of main.py as well? Does your dataset lacks rgb values as well?
Hi. Can you give more context? Is the error happening line 199 of main.py as well? Does your dataset lacks rgb values as well?
Yes, the error was in line199. I don't have rgb values in my dataset so I just set them as zeros and got the error: RuntimeError: the given numpy array has zero-sized dimensions. Zero-sized dimensions are not supported in PyTorch
Then I remove the rgb values, the code works.
Thank you very much for your reply!
Hi, I'm testing with s3dis dataset. When i run
for FOLD in 0 1 2 3 4; do \ CUDA_VISIBLE_DEVICES=0 python learning/main.py --dataset s3dis --S3DIS_PATH ./dataset --cvfold $FOLD --epochs 350 --lr_steps '[275,320]' \ --test_nth_epoch 50 --model_config 'gru_10_0,f_13' --ptn_nfeat_stn 14 --nworkers 2 --odir "results/dataset/best/cv${FOLD}"; \ done
it start running and after a bit i get this error:
Epoch 0/350 (results/dataset/best/cv4): 0%| | 0/20 [00:00<?, ?it/s]/home/spxstrix/DATI/ENEL/retineurali/superpoint_graph/learning/pointnet.py:131: UserWarning: volatile was removed and now has no effect. Use
with torch.no_grad():instead. out = model.ptn(Variable(clouds, volatile=True), Variable(clouds_global, volatile=True)) /home/spxstrix/miniconda3/lib/python3.7/site-packages/torch/nn/functional.py:1006: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead. warnings.warn("nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.") Warning: source/destination slices have same size but different shape for an index operation. This behavior is deprecated. learning/main.py:196: UserWarning: invalid index of a 0-dim tensor. This will be an error in PyTorch 0.5. Use tensor.item() to convert a 0-dim tensor to a Python number loss_meter.add(loss.data[0]) Exception ignored in: <function _DataLoaderIter.__del__ at 0x7f8896cfa620> Traceback (most recent call last): File "/home/spxstrix/miniconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 399, in __del__ self._shutdown_workers() File "/home/spxstrix/miniconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 378, in _shutdown_workers self.worker_result_queue.get() File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/queues.py", line 354, in get return _ForkingPickler.loads(res) File "/home/spxstrix/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/reductions.py", line 151, in rebuild_storage_fd fd = df.detach() File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/resource_sharer.py", line 57, in detach with _resource_sharer.get_connection(self._id) as conn: File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/resource_sharer.py", line 87, in get_connection c = Client(address, authkey=process.current_process().authkey) File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/connection.py", line 498, in Client answer_challenge(c, authkey) File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/connection.py", line 746, in answer_challenge response = connection.recv_bytes(256) # reject large message File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/connection.py", line 216, in recv_bytes buf = self._recv_bytes(maxlength) File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/connection.py", line 407, in _recv_bytes buf = self._recv(4) File "/home/spxstrix/miniconda3/lib/python3.7/multiprocessing/connection.py", line 379, in _recv chunk = read(handle, remaining) ConnectionResetError: [Errno 104] Connection reset by peer Traceback (most recent call last): File "learning/main.py", line 387, in <module> main() File "learning/main.py", line 286, in main acc, loss, oacc, avg_iou = train() File "learning/main.py", line 199, in train acc_meter.add(o_cpu, t_cpu) File "/home/spxstrix/.local/lib/python3.7/site-packages/torchnet/meter/classerrormeter.py", line 38, in add pred = torch.from_numpy(output).topk(maxk, 1, True, True)[1].numpy() RuntimeError: Dimension out of range (expected to be in range of [-1, 0], but got 1)
Previusly steps ends fine, could you help me?