Accuracy improvement for House price prediction using QNN


I am doing a case study on Boston House data to predict price using QNN with following steps

  • PCA to select features, selected upto 8 features
  • Standardsxalwer to scale data
  • AngleEmbedding to encode data
  • StronglyEntanglingLayers as the quantum layer alongwith 2 Keras layers
  • Adam optimizer
  • default.qubit and as the device

But the Accuracy is very low , please advise any pointers by which the accuracy could be improved any better QNN layers or Optimizer or architecture ?

Also please advise is their any physical device one can use for experimentation ?

Hi! I would suggest first to fit a classical model (linear regression, NN, etc.) on your PCA-selected features and use that as a benchmark. This could potentially isolate the problem to either the feature-engineering side or the QNN itself.

Hi @pratjz, to complement @_zy’s answer I would add that you can use physical devices from a lot of different providers. In our plugins page you can find in blue the different providers that we can connect to. You can choose the provider depending on how many qubits you need and what kind of hardware you want to try.

I would also suggest to check our demos page, including the community demos. Some of the most recent community demos use QNNs with Keras so looking at them might give you ideas on how to improve your results.

Please let me know if this helps!

Thank You @_zy @CatalinaAlbornoz for the Quick Pointers, will try those

No problem! Let us know how it goes :smiley:.