Tensorflow Error when Training Bigan Model

[pdefusco]

Hello, I am using this AMP in CML and the deployment of all assets was successful. I am running the train_all() method in train.py and when it gets to the bigan model I get the following error:

ValueError: Graph disconnected: cannot obtain value for tensor KerasTensor(type_spec=TensorSpec(shape=(None, 18), dtype=tf.float32, name='inputdata'), name='inputdata', description="created by layer 'inputdata'") at layer "dropout_2". The following previous layers were accessed without issue: [] ValueError Traceback (most recent call last) in engine ----> 1 train_all()

in train_all() 4 train_vae() 5 train_svm() ----> 6 train_bigan() 7 train_seq2seq()

in train_bigan() 10 bigan.train(in_train, in_test) 11 bigan.save_model() ---> 12 inlier_scores = bigan.compute_anomaly_score(in_test) 13 outlier_scores = bigan.compute_anomaly_score(out_test) 14 print(inlier_scores)

/home/cdsw/models/bigan.py in compute_anomaly_score(self, input_data, weight_parameter, lnorm_degree) 227 # This model is used to compute feature loss 228 feat_model = Model(inputs=self.discriminator.get_input_at( --> 229 0), outputs=self.discriminator.get_layer("discriminator_features").output) 230 231 # Compute z values from the input data using our encoder

/home/cdsw/.local/lib/python3.6/site-packages/tensorflow/python/training/tracking/base.py in _method_wrapper(self, *args, **kwargs) 515 self._self_setattr_tracking = False # pylint: disable=protected-access 516 try: --> 517 result = method(self, *args, **kwargs) 518 finally: 519 self._self_setattr_tracking = previous_value # pylint: disable=protected-access

/home/cdsw/.local/lib/python3.6/site-packages/tensorflow/python/keras/engine/functional.py in init(self, inputs, outputs, name, trainable, **kwargs) 118 generic_utils.validate_kwargs(kwargs, {}) 119 super(Functional, self).init(name=name, trainable=trainable) --> 120 self._init_graph_network(inputs, outputs) 121 122 @trackable.no_automatic_dependency_tracking

/home/cdsw/.local/lib/python3.6/site-packages/tensorflow/python/training/tracking/base.py in _method_wrapper(self, *args, **kwargs) 515 self._self_setattr_tracking = False # pylint: disable=protected-access 516 try: --> 517 result = method(self, *args, **kwargs) 518 finally: 519 self._self_setattr_tracking = previous_value # pylint: disable=protected-access

/home/cdsw/.local/lib/python3.6/site-packages/tensorflow/python/keras/engine/functional.py in _init_graph_network(self, inputs, outputs) 202 # Keep track of the network's nodes and layers. 203 nodes, nodes_by_depth, layers, _ = _map_graph_network( --> 204 self.inputs, self.outputs) 205 self._network_nodes = nodes 206 self._nodes_by_depth = nodes_by_depth

/home/cdsw/.local/lib/python3.6/site-packages/tensorflow/python/keras/engine/functional.py in _map_graph_network(inputs, outputs) 988 'The following previous layers ' 989 'were accessed without issue: ' + --> 990 str(layers_with_complete_input)) 991 for x in nest.flatten(node.outputs): 992 computable_tensors.add(id(x))

ValueError: Graph disconnected: cannot obtain value for tensor KerasTensor(type_spec=TensorSpec(shape=(None, 18), dtype=tf.float32, name='inputdata'), name='inputdata', description="created by layer 'inputdata'") at layer "dropout_2". The following previous layers were accessed without issue: []

[daniraviQSA]

Yes, I had the same issue. Has someone managed to solve this?

[Landino98]

I have the same problem, train of bigan is faulty. Why?