文件“ train.py”,第50行,在 model.fit(x = [H_input,H,SNR],y = H,batch_size = 256,epochs = 50000,verbose = 2,validation_split = 0.1,callbacks = [reduce_lr,checkpoint]) _method_wrapper中的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py”,第66行 返回方法(自身,* args,** kwargs) 文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py”,第848行,适合 tmp_logs = train_function(迭代器) 调用中的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py”,第580行 结果= self._call(* args,** kwds) _call的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py”,第644行 返回self._stateless_fn(* args,** kwds) 调用中的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py”,第2420行 return graph_function._filtered_call(args,kwargs)#pylint:disable =受保护的访问 _filtered_call中的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py”,行1665 self.captured_inputs) _call_flat中的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py”,行1746 ctx,args,cancelation_manager = cancellation_manager)) 调用中的文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py”,第598行 ctx = ctx) 在quick_execute中的第60行,文件“ /usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/execute.py” 输入,属性,num_outputs) tensorflow.python.framework.errors_impl.InvalidArgumentError:In [0]和In [1]必须具有兼容的批处理尺寸:[9,1,64]与[64,9,1] [[节点模型/ lambda_1 / MatMul(在/ content / drive /我的驱动器/BF-design-with-DL-master/BF-design-with-DL-master/utils.py:28中定义)]] [操作: __inference_train_function_1697]
函数调用堆栈: train_function
答案 0 :(得分:0)
似乎是您在must have compatible batch dimensions: [9,1,64] vs. [64,9,1]