if you’re training the model with

torch.nn.DataParallel

encapsulates the model

then please set multiple GPU

CUDA = '0,1'
os.environ['CUDA_VISIBLE_DEVICES'] = CUDA

Read More:

• Attributeerror: ‘module’ object has no attribute ‘handlers’ — Python sub module import problem
• PyTorch – AttributeError: ‘bool‘ object has no attribute ‘sum‘
• Attributeerror: ‘bytes’ object has no attribute’ encode ‘
• Problem: attributeerror: ‘tensor’ object has no attribute ‘creator’
• AttributeError: ‘NoneType‘ object has no attribute ‘shape‘
• Keras Model AttributeError：’str‘ object has no attribute ’call‘
• AttributeError: ‘Tensor‘ object has no attribute ‘_numpy‘
• AttributeError: module ‘torch’ has no attribute’_six’ [The problem is solved after restart]
• python gensim AttributeError: ‘Doc2Vec‘ object has no attribute ‘dv‘
• AttributeError: ‘numpy.ndarray‘ object has no attribute ‘softmax‘
• AttributeError: module ‘sacrebleu‘ has no attribute ‘compute_bleu‘
• Attributeerror: object has no attribute
• Attributeerror: ‘STR’ object has no attribute ‘decode’ solution: the pro test is successful.
• AttributeError: ‘dict‘ object has no attribute ‘iteritems‘
• AttributeError: ‘NoneType‘ object has no attribute ‘append‘
• AttributeError: ‘PipelinedRDD‘ object has no attribute ‘toDF‘
• Attributeerror: ‘dataframe’ object has no attribute ‘IX’ error
• module ‘os’ has no attribute ‘mknod’
• AttributeError: ‘WebDriver‘ object has no attribute ‘w3c‘
• AttributeError: type object ‘Image‘ has no attribute ‘open‘