PennyLane v0.18 is out of the oven! β¨οΈπ₯
Piping hot features like
-Major updates to lightning.qubit β‘
-
@PyTorch
quantum simulator π¦
-Superpowered Hamiltonians πͺ
-Beefed-up Rotosolve optimizer πͺοΈ
-Custom batch & gradient transforms πͺ
And much more!
β‘ lightning-qubit reloaded! β‘
π’ Integrated high-performance simulations built-in when you upgrade PennyLane
β‘ High-performance adjoint gradients in C++
π§ Supports all operations and observables of default.qubit
And more!
Powered by recent upgrades, you can now use
@PyTorch
as a quantum simulation backend π‘
Full support for quantum gradients via backpropagation! π
Shout out to Slimane Thabet, Esteban Payares, and Arshpreet Singh for this mega contribution from
#unitaryHACK
Hamiltonians have received new upgrades across the board ποΈππ‘
- Hamiltonian coefficients now trainable π€οΈ
- Runs natively in your circuit (no more ExpvalCost!) π
- Stores information about commuting groups π₯
All this translates into big speedups for simulations π
π We've got more quantum circuit transforms!
- Batch transforms to process multiple quantum circuits at the same time π―ββοΈ
- Gradient transforms to register custom quantum gradientsβ°οΈ
Transforms support classical processing and are fully differentiable π€
Once again, thanks to all our users, contributors, and fans for all your support! πͺ
PennyLane is completely open-source. If you like what we're doing, give us a π on GitHub, tell your friends, join one of our events, and help spread the word! π£