Calls the model on new inputs. In this case `call` just reapplies all ops in the graph to the new inputs (e.g. build a new computational graph from the provided inputs). Arguments: inputs: A tensor or list of tensors. training: Boolean or boolean scalar tensor, indi
(self, inputs, training=None, mask=None)
| 671 | self.built = True |
| 672 | |
| 673 | def call(self, inputs, training=None, mask=None): |
| 674 | """Calls the model on new inputs. |
| 675 | |
| 676 | In this case `call` just reapplies |
| 677 | all ops in the graph to the new inputs |
| 678 | (e.g. build a new computational graph from the provided inputs). |
| 679 | |
| 680 | Arguments: |
| 681 | inputs: A tensor or list of tensors. |
| 682 | training: Boolean or boolean scalar tensor, indicating whether to run |
| 683 | the `Network` in training mode or inference mode. |
| 684 | mask: A mask or list of masks. A mask can be |
| 685 | either a tensor or None (no mask). |
| 686 | |
| 687 | Returns: |
| 688 | A tensor if there is a single output, or |
| 689 | a list of tensors if there are more than one outputs. |
| 690 | """ |
| 691 | if not self._is_graph_network: |
| 692 | raise NotImplementedError('When subclassing the `Model` class, you should' |
| 693 | ' implement a `call` method.') |
| 694 | |
| 695 | return self._run_internal_graph(inputs, training=training, mask=mask) |
| 696 | |
| 697 | def compute_output_shape(self, input_shape): |
| 698 | if not self._is_graph_network: |
no test coverage detected