Unable to serialize

[n20100117]

Unable to serialize [2.0896919 2.1128857 2.1081853] to JSON. Unrecognized type <class 'tensorflow.python.framework.ops.EagerTensor'>.

What should I do to fix it?

[gauravreddy08]

Tensorflow 2.10 and 2.11 have some bugs with EfficientNet series. Downgrade your tensorflow version to 2.9.

[gauravreddy08]

try 2.9.1

[n20100117]

OK

[n20100117]

Still the same result

---

TypeError Traceback (most recent call last)
in
4 # Create base model
5 input_shape = (224, 224, 3)
----> 6 base_model = tf.keras.applications.EfficientNetB1(include_top=False)
7
8 # Input and Data Augmentation

4 frames
/usr/local/lib/python3.9/dist-packages/keras/applications/efficientnet.py in EfficientNetB1(include_top, weights, input_tensor, input_shape, pooling, classes, classifier_activation, **kwargs)
572 classifier_activation='softmax',
573 **kwargs):
--> 574 return EfficientNet(
575 1.0,
576 1.1,

/usr/local/lib/python3.9/dist-packages/keras/applications/efficientnet.py in EfficientNet(width_coefficient, depth_coefficient, default_size, dropout_rate, drop_connect_rate, depth_divisor, activation, blocks_args, model_name, include_top, weights, input_tensor, input_shape, pooling, classes, classifier_activation)
332 # original implementation.
333 # See tensorflow/tensorflow#49930 for more details
--> 334 x = x / tf.math.sqrt(IMAGENET_STDDEV_RGB)
335
336 x = layers.ZeroPadding2D(

/usr/local/lib/python3.9/dist-packages/tensorflow/python/util/traceback_utils.py in error_handler(*args, **kwargs)
151 except Exception as e:
152 filtered_tb = _process_traceback_frames(e.traceback)
--> 153 raise e.with_traceback(filtered_tb) from None
154 finally:
155 del filtered_tb

/usr/local/lib/python3.9/dist-packages/keras/layers/core/tf_op_layer.py in handle(self, op, args, kwargs)
105 isinstance(x, keras_tensor.KerasTensor)
106 for x in tf.nest.flatten([args, kwargs])):
--> 107 return TFOpLambda(op)(*args, **kwargs)
108 else:
109 return self.NOT_SUPPORTED

/usr/local/lib/python3.9/dist-packages/keras/utils/traceback_utils.py in error_handler(*args, **kwargs)
65 except Exception as e: # pylint: disable=broad-except
66 filtered_tb = _process_traceback_frames(e.traceback)
---> 67 raise e.with_traceback(filtered_tb) from None
68 finally:
69 del filtered_tb

TypeError: Exception encountered when calling layer "tf.math.truediv" (type TFOpLambda).

x and y must have the same dtype, got tf.float16 != tf.float32.

Call arguments received by layer "tf.math.truediv" (type TFOpLambda):
• x=tf.Tensor(shape=(None, None, None, 3), dtype=float16)
• y=tf.Tensor(shape=(3,), dtype=float32)
• name=None

[n20100117]

I have solve the problem.
I use TensorFlow 2.7.0

[inikresek]

interesting