Issues with pennylane lightning installation

[isaacdevlugt]

Hey @imakash! It might be worth updating to the most-recent versions of everything — some of your package versions are out-of-sync (you’re pretty up-to-speed with lightning-gpu but not pl-lightning or pennylane)

@imakash

When the device is “lightning.qubit”, I get the following warning message

You can try building pl-lightning from source via pip install pybind11 pennylane-lightning --no-binary :all:.

@imakash

And when the device is “lightning.gpu”, I get the following error message.

This forum post might be helpful!

Pennylane-Lightning GPU

Hi @George_LA and @mlxd and @isaacdevlugt, Even I tried the solution given by @mlxd, however, I was not able to run the script. So, instead, I did this: import torch device = torch.device("cuda:0" if (torch.cuda.is_available()) else "cpu") I created the quantum circuit under a class. Then, I sent it to the GPU. It was able to utilize it. class QCNN(nn.Module): def CONVCircuit(phi, wires, i=0): """ quantum convolution Node """ # parameter theta = np.pi / 2 …