问题描述
我正在尝试实现与三重损失有关的自定义损失功能。三元组损失提供了一个自定义距离度量，可返回嵌入之间的成对距离。我定义了一个自定义函数，该函数可以在正向传播中正常工作。但是在反向传播时会引发一些错误。以下是错误。
InvalidArgumentError: slice index 16 of dimension 1 out of bounds.
[[{{node TripletSemiHardLoss/PartitionedCall/while_1/body/_226/while_1/strided_slice}}]] [Op:__inference_train_function_31232]
16是我输入的批次大小。我没有在自定义代码中使用任何while循环。但是，有一个for循环。
我尝试了以下方法。
- 我使用tf.size（input）检索批次大小。在前进道具上工作正常。
- 我尝试了while循环和for循环。在向前传播时，两者都工作正常。两者都产生相同的结果。但是在反向传播中，两者都抛出相同的错误。
这是总错误堆栈：
---------------------------------------------------------------------------
InvalidArgumentError Traceback (most recent call last)
<ipython-input-22-70c4ddc79f73> in <module>
11 epochs=25,12 callbacks=[checkpoint],---> 13 verbose=1)
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/util/deprecation.py in new_func(*args,**kwargs)
322 'in a future version' if date is None else ('after %s' % date),323 instructions)
--> 324 return func(*args,**kwargs)
325 return tf_decorator.make_decorator(
326 func,new_func,'deprecated',~/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py in fit_generator(self,generator,steps_per_epoch,epochs,verbose,callbacks,validation_data,validation_steps,validation_freq,class_weight,max_queue_size,workers,use_multiprocessing,shuffle,initial_epoch)
1827 use_multiprocessing=use_multiprocessing,1828 shuffle=shuffle,-> 1829 initial_epoch=initial_epoch)
1830
1831 @deprecation.deprecated(
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py in _method_wrapper(self,*args,**kwargs)
106 def _method_wrapper(self,**kwargs):
107 if not self._in_multi_worker_mode(): # pylint: disable=protected-access
--> 108 return method(self,**kwargs)
109
110 # Running inside `run_distribute_coordinator` already.
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py in fit(self,x,y,batch_size,validation_split,sample_weight,initial_epoch,validation_batch_size,use_multiprocessing)
1096 batch_size=batch_size):
1097 callbacks.on_train_batch_begin(step)
-> 1098 tmp_logs = train_function(iterator)
1099 if data_handler.should_sync:
1100 context.async_wait()
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py in __call__(self,**kwds)
778 else:
779 compiler = "nonXla"
--> 780 result = self._call(*args,**kwds)
781
782 new_tracing_count = self._get_tracing_count()
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py in _call(self,**kwds)
838 # Lifting succeeded,so variables are initialized and we can run the
839 # stateless function.
--> 840 return self._stateless_fn(*args,**kwds)
841 else:
842 canon_args,canon_kwds = \
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py in __call__(self,**kwargs)
2827 with self._lock:
2828 graph_function,args,kwargs = self._maybe_define_function(args,kwargs)
-> 2829 return graph_function._filtered_call(args,kwargs) # pylint: disable=protected-access
2830
2831 @property
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py in _filtered_call(self,kwargs,cancellation_manager)
1846 resource_variable_ops.BaseResourceVariable))],1847 captured_inputs=self.captured_inputs,-> 1848 cancellation_manager=cancellation_manager)
1849
1850 def _call_flat(self,captured_inputs,cancellation_manager=None):
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py in _call_flat(self,cancellation_manager)
1922 # No tape is watching; skip to running the function.
1923 return self._build_call_outputs(self._inference_function.call(
-> 1924 ctx,cancellation_manager=cancellation_manager))
1925 forward_backward = self._select_forward_and_backward_functions(
1926 args,~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py in call(self,ctx,cancellation_manager)
548 inputs=args,549 attrs=attrs,--> 550 ctx=ctx)
551 else:
552 outputs = execute.execute_with_cancellation(
~/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/execute.py in quick_execute(op_name,num_outputs,inputs,attrs,name)
58 ctx.ensure_initialized()
59 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle,device_name,op_name,---> 60 inputs,num_outputs)
61 except core._NotOkStatusException as e:
62 if name is not None:
InvalidArgumentError: slice index 16 of dimension 1 out of bounds.
[[{{node TripletSemiHardLoss/PartitionedCall/while_1/body/_226/while_1/strided_slice}}]] [Op:__inference_train_function_31232]
Function call stack:
train_function
解决方法
实际上是因为numpy样式数组切片。使用tf.slice解决了该问题。