AttributeError: Can't pickle local object 'ADE20KDataset.__init__.<locals>.<lambda>'

[JiahangWu]

Hi, I downloaded all the available Model Zoo, and ran the evaluation.py. However, there are some bugs in my project. Please, give me some suggestions and solutions. Thanks.

The error is following:

Traceback (most recent call last):
  File "evaluation.py", line 279, in <module>
    sample = next(iter(val_loader))
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\site-packages\torch\utils\data\dataloader.py", line 279, in __iter__
    return _MultiProcessingDataLoaderIter(self)
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\site-packages\torch\utils\data\dataloader.py", line 719, in __init__
    w.start()
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\multiprocessing\process.py", line 112, in start
    self._popen = self._Popen(self)
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\multiprocessing\context.py", line 223, in _Popen
    return _default_context.get_context().Process._Popen(process_obj)
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\multiprocessing\context.py", line 322, in _Popen
    return Popen(process_obj)
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\multiprocessing\popen_spawn_win32.py", line 89, in __init__
    reduction.dump(process_obj, to_child)
  File "D:\Software\Code\Anaconda3\envs\SA-Scene-Recognition\lib\multiprocessing\reduction.py", line 60, in dump
    ForkingPickler(file, protocol).dump(obj)
AttributeError: Can't pickle local object 'ADE20KDataset.__init__.<locals>.<lambda>'

[DGideas]

@Lollipop9z If you're use Windows, please consider switching to Ubuntu.

Some strange error with pickle often occur when your OS environment is inconsistent with the author's recommand operation system.

[JiahangWu]

But I still have other error on the Ubuntu with the GPU. This error told me that there is no memory of the GPU, but it still have. Please, help me. ^&^

Traceback (most recent call last):
  File "evaluation.py", line 300, in <module>
    val_top1, val_top2, val_top5, val_loss, val_ClassTPDic = evaluationDataLoader(val_loader, model, set='Validation')
  File "evaluation.py", line 76, in evaluationDataLoader
    outputSceneLabel, feature_conv, outputSceneLabelRGB, outputSceneLabelSEM = model(RGB_image, semanticTensor)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/Semantic-Aware-Scene-Recognition/SASceneNet.py", line 165, in forward
    e1 = self.encoder1(x)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/container.py", line 92, in forward
    input = module(input)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torchvision/models/resnet.py", line 41, in forward
    out = self.conv1(x)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 320, in forward
    self.padding, self.dilation, self.groups)
RuntimeError: CUDA out of memory. Tried to allocate 338.50 MiB (GPU 0; 5.79 GiB total capacity; 3.59 GiB already allocated; 340.94 MiB free; 307.46 MiB cached)

[JiahangWu]

But I still have other error on the Ubuntu with the GPU. This error told me that there is no memory of the GPU, but it still have. Please, help me. ^&^

Traceback (most recent call last):
  File "evaluation.py", line 300, in <module>
    val_top1, val_top2, val_top5, val_loss, val_ClassTPDic = evaluationDataLoader(val_loader, model, set='Validation')
  File "evaluation.py", line 76, in evaluationDataLoader
    outputSceneLabel, feature_conv, outputSceneLabelRGB, outputSceneLabelSEM = model(RGB_image, semanticTensor)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/Semantic-Aware-Scene-Recognition/SASceneNet.py", line 165, in forward
    e1 = self.encoder1(x)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/container.py", line 92, in forward
    input = module(input)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torchvision/models/resnet.py", line 41, in forward
    out = self.conv1(x)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/lollipop/anaconda3/envs/SA-Scene-Recognition/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 320, in forward
    self.padding, self.dilation, self.groups)
RuntimeError: CUDA out of memory. Tried to allocate 338.50 MiB (GPU 0; 5.79 GiB total capacity; 3.59 GiB already allocated; 340.94 MiB free; 307.46 MiB cached)

The solution is that modify the batch_size being smaller, such as 2, 4, 6, in the config file. For example:

/Config/config_MITIndoor.yaml

VALIDATION:
    PRINT_FREQ: 10
    BATCH_SIZE:
        TRAIN: 100
        TEST: 2