Thomas Dybdahl Ahle

Use separate PQs in each cluster

2

Currently the same product quantizer is used for every cluster in IVF. However, the PQ doesn't use a lot of space (it's just 16 center points), so we might as...

Currently, only SSE is supported. It would be nice to also support AMD chips.

Add numba

1

PyNNDescent (https://github.com/lmcinnes/pynndescent) is able to get a lot of speed (presumably) by using [numba](https://numba.pydata.org/). This should be relatively easy to add to fast_pq as well.

Support more ast types in Signature typed parsing

1

On the website, you [see this example](https://dspy.ai/#__tabbed_3_3): ```python import dspy dspy.configure(lm=dspy.LM('gpt-4o-mini-2024-07-18')) # Define the DSPy module for classification. It will use the hint at training time, if available. signature =...

Added support for more types

1

Adds support for all python `ast` types in string signatures, like `Signature("text -> label:Literal['a', 'b']")`. Fixes https://github.com/stanfordnlp/dspy/issues/1899

In Figure 6 you show that the Pareto frontier for ResNet-50 includes CP and "Standard" convolutions. However, there are at least three more "unnamed" Einconvs which are also on the...

## Summary - correct typos in README notation section - further clarify README by fixing small spelling errors ## Testing - `python -m py_compile sunfish.py sunfish_nnue.py`

Does cotengra support sparse tensors, like convolutions?

7

The [Hayashi et al.](https://arxiv.org/abs/1908.04471) way to write a convolutional layer in a NN is, `einsum('c h w, h h2 i, w w2 j, c i j c2 -> c2 h2...