# QNN KerasLayer with Model **URL:** \ **Category:** PennyLane Help\ **Created:** [June 2, 2020, 10:28am UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413 "2020-06-02T10:28:43Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![cnada](https://avatars.discourse-cdn.com/v4/letter/c/f05b48/32.png) [@cnada](https://discuss.pennylane.ai/u/cnada)\ **Post date:** [June 2, 2020, 10:28am UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/1 "2020-06-02T10:28:43Z")
Hello, Let’s say I already have a keras Model (call it classical\_nn) into which I want to add a qnn.KerasLayer at the end. Classically in tensorflow, one would add layers to a keras Model as : classical\_nn = Model(…) pre\_out = classical\_nn.layers[-1].output pre\_out = Dense(16,activation=‘tanh’) (pre\_out) … And let’s say I want to add a qnn.KerasLayer set up as: qlayer = qml.qnn.KerasLayer(quantum\_classifier, weight\_shapes, output\_dim=1) How one would connect qlayer and pre\_out in this case to obtain a keras.Model? Thanks in advance for your support 🙂 ---
**Author:** ![Tom\_Bromley](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/tom_bromley/32/129_2.png) [@Tom\_Bromley](https://discuss.pennylane.ai/u/Tom_Bromley)\ **Post date:** [June 2, 2020, 12:11pm UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/2 "2020-06-02T12:11:22Z")
Hey @cnada, PennyLane’s `KerasLayer` is a subclass of the [Layer](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Layer) class in Keras. This means that you can treat it just like any other layer in Keras. For example, suppose we turn our QNode into a Keras Layer: ```python qlayer = qml.qnn.KerasLayer(qnode, weight_shapes, output_dim=2) ``` We can use the `Sequential` model to create a hybrid: ```python clayer = tf.keras.layers.Dense(2) clayer2 = tf.keras.layers.Dense(2) model = tf.keras.models.Sequential([clayer, qlayer, clayer2]) ``` We can also use Keras’ [functional API](https://keras.io/guides/functional_api/) approach: ```python inputs = tf.keras.Input(shape=(2,)) x = tf.keras.layers.Dense(2, activation="tanh")(inputs) x = qlayer(x) outputs = tf.keras.layers.Dense(2)(x) model = tf.keras.Model(inputs=inputs, outputs=outputs) ``` You can also add `qlayer` to an existing model using: - `model.add(qlayer)` is using a `Sequential` model - If using the functional API approach: ```python outputs2 = qlayer(outputs) model = tf.keras.Model(inputs=inputs, outputs=outputs2) ``` Hope this helps! ---
**Author:** ![cnada](https://avatars.discourse-cdn.com/v4/letter/c/f05b48/32.png) [@cnada](https://discuss.pennylane.ai/u/cnada)\ **Post date:** [June 2, 2020, 1:08pm UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/3 "2020-06-02T13:08:25Z")
Thanks @Tom_Bromley. And if I may ask, when doing x = qlayer(x), if I get the error: AttributeError: ‘tuple’ object has no attribute ‘layer’ Do you know what it means? I have no clue here. ---
**Author:** ![Tom\_Bromley](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/tom_bromley/32/129_2.png) [@Tom\_Bromley](https://discuss.pennylane.ai/u/Tom_Bromley)\ **Post date:** [June 2, 2020, 1:26pm UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/4 "2020-06-02T13:26:06Z")
No problem! Would you be able to share the code you’re running so I can take a closer look? ---
**Author:** ![cnada](https://avatars.discourse-cdn.com/v4/letter/c/f05b48/32.png) [@cnada](https://discuss.pennylane.ai/u/cnada)\ **Post date:** [June 2, 2020, 1:44pm UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/5 "2020-06-02T13:44:30Z")
@Tom_Bromley Here is an example code replicating the error.[error\_pennylane\_classical\_net\_quantunlayer.py](https://discuss.pennylane.ai/uploads/short-url/jgDdmSRSI6xoXNYgISSMfzXD1EI.py) (3.9 KB) ---
**Author:** ![Tom\_Bromley](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/tom_bromley/32/129_2.png) [@Tom\_Bromley](https://discuss.pennylane.ai/u/Tom_Bromley)\ **Post date:** [June 2, 2020, 3:47pm UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/6 "2020-06-02T15:47:50Z")
Thanks for sharing! I managed to get an output from the hybrid network by changing the script from line 100 to: ```python import pennylane as qml n_qubits = 5 dev = qml.device('default.qubit', wires=n_qubits) @qml.qnode(dev) def quantum_classifier(inputs, weights): qml.templates.AngleEmbedding(inputs, wires=range(n_qubits)) qml.templates.StronglyEntanglingLayers(weights, wires=range(n_qubits)) return qml.expval(qml.PauliZ(0)) weight_shapes = {"weights": (2, n_qubits, 3)} qlayer = qml.qnn.KerasLayer(quantum_classifier, weight_shapes, output_dim=1) flat = tf.keras.layers.Flatten() downsize_layer = tf.keras.layers.Dense(5) x = downsize_layer(flat(model.output)) outputs = qlayer(x) hybrid_model = Model(inputs=model.input, outputs=outputs) import numpy as np in_data = np.random.random((1, 128, 128, 1)) hybrid_model(in_data) ``` Note that we need to flatten the output of the classical network and shrink it down to a 5-dimensional input for the quantum layer. This could probably be done with a few more fully connected intermediate layers. However, it’s also important to bear in mind that the quantum layer will add a simulation overhead to your network, due to the overhead of simulating a quantum system. Combined with the size of your classical network and resultant number of parameters (2,161,361), it will take a long time to even perform one optimization step. I’d recommend starting out with a smaller scale model as more of a prototype. ---
**Author:** ![cnada](https://avatars.discourse-cdn.com/v4/letter/c/f05b48/32.png) [@cnada](https://discuss.pennylane.ai/u/cnada)\ **Post date:** [June 2, 2020, 4:23pm UTC](https://discuss.pennylane.ai/t/qnn-keraslayer-with-model/413/7 "2020-06-02T16:23:56Z")
Gotcha. Thanks for the much appreciated tips. Maybe also make the previous layers non trainable is another way.