训练时张量的奇怪形状和 ResourceExhaustedError：分配张量时出现 OOM

如何解决训练时张量的奇怪形状和 ResourceExhaustedError：分配张量时出现 OOM

我正在尝试使用这个 github repo 运行对象检测，它利用了简单的 7 层 Single Shot MultiBox Detector。 我使用包在 Google Colab 上运行了这个：keras==2.2.4 & tensorflow-gpu==1.13.1

最终我在训练时遇到了下面这个错误。我想抱怨的另一件事是导致它崩溃的张量的形状有一个形状 [2,1232,1640,48] 哪里...

• 2 是批量大小
• 1232 是宽度的一半（奇怪）
• 1640 是高度的一半（奇怪）
• 不确定 48 在哪里

Epoch 1/5

---------------------------------------------------------------------------

ResourceExhaustedError                    Traceback (most recent call last)

<ipython-input-28-3fbd9e60a593> in <module>()
     19 
     20                               max_queue_size=1,---> 21                               workers=0)

7 frames

/usr/local/lib/python3.6/dist-packages/keras/legacy/interfaces.py in wrapper(*args,**kwargs)
     89                 warnings.warn('Update your `' + object_name + '` call to the ' +
     90                               'Keras 2 API: ' + signature,stacklevel=2)
---> 91             return func(*args,**kwargs)
     92         wrapper._original_function = func
     93         return wrapper

/usr/local/lib/python3.6/dist-packages/keras/engine/training.py in fit_generator(self,generator,steps_per_epoch,epochs,verbose,callbacks,validation_data,validation_steps,class_weight,max_queue_size,workers,use_multiprocessing,shuffle,initial_epoch)
   1416             use_multiprocessing=use_multiprocessing,1417             shuffle=shuffle,-> 1418             initial_epoch=initial_epoch)
   1419 
   1420     @interfaces.legacy_generator_methods_support

/usr/local/lib/python3.6/dist-packages/keras/engine/training_generator.py in fit_generator(model,initial_epoch)
    215                 outs = model.train_on_batch(x,y,216                                             sample_weight=sample_weight,--> 217                                             class_weight=class_weight)
    218 
    219                 outs = to_list(outs)

/usr/local/lib/python3.6/dist-packages/keras/engine/training.py in train_on_batch(self,x,sample_weight,class_weight)
   1215             ins = x + y + sample_weights
   1216         self._make_train_function()
-> 1217         outputs = self.train_function(ins)
   1218         return unpack_singleton(outputs)
   1219 

/usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py in __call__(self,inputs)
   2713                 return self._legacy_call(inputs)
   2714 
-> 2715             return self._call(inputs)
   2716         else:
   2717             if py_any(is_tensor(x) for x in inputs):

/usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py in _call(self,inputs)
   2673             fetched = self._callable_fn(*array_vals,run_metadata=self.run_metadata)
   2674         else:
-> 2675             fetched = self._callable_fn(*array_vals)
   2676         return fetched[:len(self.outputs)]
   2677 

/usr/local/lib/python3.6/dist-packages/tensorflow/python/client/session.py in __call__(self,*args,**kwargs)
   1437           ret = tf_session.TF_SessionRunCallable(
   1438               self._session._session,self._handle,args,status,-> 1439               run_metadata_ptr)
   1440         if run_metadata:
   1441           proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)

/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/errors_impl.py in __exit__(self,type_arg,value_arg,traceback_arg)
    526             None,None,527             compat.as_text(c_api.TF_Message(self.status.status)),--> 528             c_api.TF_GetCode(self.status.status))
    529     # Delete the underlying status object from memory otherwise it stays alive
    530     # as there is a reference to status from this from the traceback due to

ResourceExhaustedError: OOM when allocating tensor with shape[2,48] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc
     [[{{node training/Adam/gradients/zeros_22-0-1-TransposeNCHWToNHWC-LayoutOptimizer}}]]
Hint: If you want to see a list of allocated tensors when OOM happens,add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

     [[{{node loss/add_14}}]]
Hint: If you want to see a list of allocated tensors when OOM happens,add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

请说明发生了什么以及如何解决这个问题。如果这有助于找到错误，我还可以分享有关模型结构的更多相关细节。

解决方法

如果你有形状的数据

(2,1232,1640,3)

在它通过带有 "SAME" 填充的 42 个过滤器的卷积层后，它将具有形状

(2,42)

并且在您的 GPU 上没有该张量的位置。

我查看了 repo，有一堆带有 48 个过滤器的层

conv2 = Conv2D(48,(3,3),strides=(1,1),padding="same",kernel_initializer='he_normal',kernel_regularizer=l2(l2_reg),name='conv2')(pool1)
conv2 = BatchNormalization(axis=3,momentum=0.99,name='bn2')(conv2)
conv2 = ELU(name='elu2')(conv2)
pool2 = MaxPooling2D(pool_size=(2,2),name='pool2')(conv2)