Pytorch调试过程的BUG
1.损失函数的loss向量对不上
// An highlighted block
var foo = 'bar';Traceback (most recent call last):
File "train.py", line 182, in <module>
val_percent=args.val / 100)
File "train.py", line 81, in train_net
loss = criterion(masks_pred, true_masks)
File "/public/home/lidd/.conda/envs/lgg/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in __call__
result = self.forward(*input, **kwargs)
File "/public/home/lidd/.conda/envs/lgg/lib/python3.6/site-packages/torch/nn/modules/loss.py", line 862, in forward
ignore_index=self.ignore_index, reduction=self.reduction)
File "/public/home/lidd/.conda/envs/lgg/lib/python3.6/site-packages/torch/nn/functional.py", line 1550, in cross_entropy
return nll_loss(log_softmax(input, 1), target, weight, None, ignore_index, None, reduction)
File "/public/home/lidd/.conda/envs/lgg/lib/python3.6/site-packages/torch/nn/functional.py", line 1409, in nll_loss
return torch._C._nn.nll_loss2d(input, target, weight, _Reduction.get_enum(reduction), ignore_index)
RuntimeError: invalid argument 3: only batches of spatial targets supported (3D tensors) but got targets of dimension: 4 at /opt/conda/conda-bld/pytorch_1533672544752/work/aten/src/THNN/generic/SpatialClassNLLCriterion.c:59
这里的意思是计算损失函数的时候,target支持为3维向量,而这里得到的是四维向量。
输入向量:[batch-size,7,H,W]
输出向量:[batch-size,1,H,W]
那么到底哪里是应该抛弃的维度呢?
Pytorch官方文档 CrossEntropyLoss
红框内为输入向量维度,蓝框内为目标向量维度。
所以我们应该把输入向量,输出向量的维度修改为。
输入向量:[batch-size,7,H,W]
输出向量:[batch-size,H,W]
2.测试过程中向量维度不符合要求
// An highlighted block
Traceback (most recent call last):
File "train.py", line 182, in <module>
val_percent=args.val / 100)
File "train.py", line 99, in train_net
val_score = eval_net(net, val_loader, device)
File "/public/home/lidd/lgg/Pytorch-UNet-master/eval.py", line 25, in eval_net
tot += F.cross_entropy(mask_pred, true_masks).item()
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/nn/functional.py", line 2009, in cross_entropy
return nll_loss(log_softmax(input, 1), target, weight, None, ignore_index, None, reduction)
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/nn/functional.py", line 1840, in nll_loss
ret = torch._C._nn.nll_loss2d(input, target, weight, _Reduction.get_enum(reduction), ignore_index)
RuntimeError: invalid argument 3: only batches of spatial targets supported (3D tensors) but got targets of dimension: 4 at /tmp/pip-req-build-ocx5vxk7/aten/src/THNN/generic/SpatialClassNLLCriterion.c:61
修改测试过程中的向量维度。
3.语法错误
// An highlighted block
(lgg) [lidd@login03 Pytorch-UNet-master]$ python train.py -b=8
INFO: Using device cpu
INFO: Network:
1 input channels
7 output channels (classes)
Bilinear upscaling
INFO: Creating dataset with 868 examples
INFO: Starting training:
Epochs: 5
Batch size: 8
Learning rate: 0.001
Training size: 782
Validation size: 86
Checkpoints: True
Device: cpu
Images scaling: 1
Epoch 1/5: 0%| | 0/782 [00:00<?, ?img/s]
Illegal instruction (core dumped)
报错非法指令。
解决方案:重新安装环境后解决。
4.Batch-size=n训练过程中loss突然变成0
// loss 突然变成0
python train.py -b=8
INFO: Using device cpu
INFO: Network:
1 input channels
7 output channels (classes)
Bilinear upscaling
INFO: Creating dataset with 868 examples
INFO: Starting training:
Epochs: 5
Batch size: 8
Learning rate: 0.001
Training size: 782
Validation size: 86
Checkpoints: True
Device: cpu
Images scaling: 1
Epoch 1/5: 10%|██████████████▏ | 80/782 [01:33<13:21, 1.14s/img, loss (batch)=0.886I
NFO: Validation cross entropy: 1.86862473487854
Epoch 1/5: 20%|███████████████████████████▊ | 160/782 [03:34<11:51, 1.14s/img, loss (batch)=2.35e-7I
NFO: Validation cross entropy: 5.887489884504049e-10
Epoch 1/5: 31%|███████████████████████████████████████████▌ | 240/782 [05:41<11:29, 1.27s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 41%|██████████████████████████████████████████████████████████ | 320/782 [07:49<09:16, 1.20s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 51%|████████████████████████████████████████████████████████████████████████▋ | 400/782 [09:55<07:31, 1.18s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 61%|███████████████████████████████████████████████████████████████████████████████████████▏ | 480/782 [12:02<05:58, 1.19s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 72%|█████████████████████████████████████████████████████████████████████████████████████████████████████▋ | 560/782 [14:04<04:16, 1.15s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 82%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ | 640/782 [16:11<02:49, 1.20s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 92%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▋ | 720/782 [18:21<01:18, 1.26s/img, loss (batch)=0I
NFO: Validation cross entropy: 0.0
Epoch 1/5: 94%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▋ | 736/782 [19:17<01:12, 1.57s/img, loss (batch)=0]
Traceback (most recent call last):
File "train.py", line 182, in <module>
val_percent=args.val / 100)
File "train.py", line 66, in train_net
for batch in train_loader:
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 819, in __next__
return self._process_data(data)
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 846, in _process_data
data.reraise()
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/_utils.py", line 385, in reraise
raise self.exc_type(msg)
RuntimeError: Caught RuntimeError in DataLoader worker process 4.
Original Traceback (most recent call last):
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 178, in _worker_loop
data = fetcher.fetch(index)
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 47, in fetch
return self.collate_fn(data)
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/_utils/collate.py", line 74, in default_collate
return {key: default_collate([d[key] for d in batch]) for key in elem}
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/_utils/collate.py", line 74, in <dictcomp>
return {key: default_collate([d[key] for d in batch]) for key in elem}
File "/public/home/lidd/.conda/envs/lgg2/lib/python3.6/site-packages/torch/utils/data/_utils/collate.py", line 55, in default_collate
return torch.stack(batch, 0, out=out)
RuntimeError: Expected object of scalar type Double but got scalar type Byte for sequence element 4 in sequence argument at position #1 'tensors'
分析问题:
训练的loss和测试的loss初始没有问题。
问题关键出在训练中,出现了loss突降的原因。
5.Batch-size=1训练过程中loss跳动
// An highlighted block
(lgg2) [lidd@login03 Pytorch-UNet-master]$ python train.py -b=1
INFO: Using device cpu
INFO: Network:
1 input channels
7 output channels (classes)
Bilinear upscaling
INFO: Creating dataset with 868 examples
INFO: Starting training:
Epochs: 5
Batch size: 1
Learning rate: 0.001
Training size: 782
Validation size: 86
Checkpoints: True
Device: cpu
Images scaling: 1
Epoch 1/5: 11%|█████████████▉ | 86/782 [02:12<12:41, 1.09s/img, loss (batch)=0.0636] I
NFO: Validation cross entropy: 0.05946924779997315
Epoch 1/5: 22%|███████████████████████████▍ | 172/782 [04:52<10:44, 1.06s/img, loss (batch)=1.15e-8] I
NFO: Validation cross entropy: 3.730315458357298
Epoch 1/5: 33%|█████████████████████████████████████████▏ | 258/782 [08:16<25:50, 2.96s/img, loss (batch)=5.31e-9] I
NFO: Validation cross entropy: 2.2019995843036284
Epoch 1/5: 44%|██████████████████████████████████████████████████████▉ | 344/782 [13:43<21:47, 2.98s/img, loss (batch)=7.86e-9] I
NFO: Validation cross entropy: 1.200446563662029
Epoch 1/5: 54%|████████████████████████████████████████████████████████████████████▏ | 423/782 [18:43<10:34, 1.77s/img, loss (batch)=6.7e-9]
应该随机梯度下降不稳定的原因。