autograd

🚀 Feature

Support to pickle a jitted function (or at least throw a TypeError when using protocol 0 and 1).

Motivation

Trying to pickle a jitted function either raises TypeError: cannot pickle 'torch._C.ScriptFunction' object when protocol>1 or far worse when using protocol=0 or protocol=1 python 3.9.5 dies with:

terminate called after throwing an instance of 'std::runt

Description

Implement a YOLO model and add it to the DJL model zoo

References

• https://arxiv.org/abs/1506.02640

Issue to track tutorial requests:

• Deep Learning with PyTorch: A 60 Minute Blitz - #69
• Sentence Classification - #79

As I was inspecting the MultiControlledX gate, I noticed how it printed out:

MultiControlledX(array([[0, 1],
       [1, 0]]), wires=[0, 1, 2, 3])

And was quite confused as to where the array came from. It turns out MultiControlledX inherits from ControlledQubitUnitary with the Unitary matrix as an X gate. This makes MultiControlledX a parametrized gate.

Worse yet, th

We should support the pseudo-derivative used in Bellec et al. 2018 and subsequent
work by other authors as implemented for example here:

https://github.com/IGITUGraz/eligibility_propagation/blob/efd02e6879c01cda3fa9a7838e8e2fd08163c16e/Figure_3_and_S7_e_prop_tutorials/models.py#L40-L70

It has the advantage that it naturally lead to sparse gradients and therefore is a good alternative
to f

The init module has been deprecated, and the recommend approach for generating initial weights is to use the Template.shape method:

>>> from pennylane.templates import StronglyEntanglingLayers
>>> qml.init.strong_ent_layers_normal(n_layers=3, n_wires=2) # deprecated
>>> np.random.random(StronglyEntanglingLayers.shape(n_layers=3, n_wires=2))  # new approach

We should upd

Okay, so this might not exactly be a "good first issue" - it is a little more advanced, but is still very much accessible to newcomers.

Similar to the mygrad.nnet.max_pool function, I would like there to be a mean-pooling layer. That is, a convolution-style windows is strided over the input, an