Carlos Martin

@lazka [This Sourcegraph query](https://sourcegraph.com/search?q=context:global+%28context%7Cctx%29%5C.arc%5C%28+lang:python&patternType=regexp&sm=0&groupBy=repo) can give some idea of how common it is to use `arc` to just draw a circle.

@mikesol I like this idea, and was looking for an implementation myself. @chiamp I think stacked RNNs are quite commonly used.

@vroulet I meant something like the example shown in the second link (more examples [here](https://www.cvxpy.org/examples/index.html)). Here's another example: ```python3 def find_nash_equilibrium(u): """Find the Nash equilibrium of a two-player zero-sum normal-form...

@jakevdp Recalling what we did for [jax.numpy.fill_diagonal](https://jax.readthedocs.io/en/latest/_autosummary/jax.numpy.fill_diagonal.html) with `inplace` (vs. [numpy.fill_diagonal](https://numpy.org/doc/stable/reference/generated/numpy.fill_diagonal.html)): - https://github.com/google/jax/issues/2680#issuecomment-1763273560 - https://github.com/google/jax/pull/18180 I think that, since the API will be identical/compatible when this functionality is added to...

@jakevdp Thanks for your response! Is there a namespace for useful array functionality that has not yet been incorporated into NumPy? Perhaps [`jax.lax`](https://jax.readthedocs.io/en/latest/jax.lax.html) or some theoretical `jax.numpy_experimental`? --- Somewhat off-topic,...

@mattjj One example is how PyTorch makes it easier to chain array operations. Consider the following: ```python3 jnp.sqrt(jnp.square(abs(x.sum(2)).max(1)).sum(0)) ``` Notice the "back-and-forth" or "spiral" pattern of the function calls. Compare...

Thanks for the link. It's indeed very relevant. It also brings up another advantage I forgot to include: > With fluent API in common, users may more easily write library-agnostic...

> I understand the choice the numpy devs have made. [...] They’ve made their opinions pretty clear But do you agree with it? Does the user community agree with it?...

@hawkinsp ```shell python3 -m pip install --upgrade "jax[cuda11_pip]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html ````

@pengzhi1998 I did a clean reinstall of the CUDA driver and toolkit at some point. That seems to have solved the issue.