syncopy

syncopy copied to clipboard

Add support for trial-averaging in combination w/parallel writing

2

We should rethink, whether trial-averaging must really trigger sequential writing - why not use concurrent writing and average afterwards?

joschaschmiedt

ZETA method to replace PSTH

6

As far as I know doesn't have a PSTH method yet for spike events. My suggestion is not to bother and implement this cool looking method called Zeta IFR instead:...

KatharineShapcott

Make in-place selections the default for data access

8

Explore UI/UX options to have in-place selections be present "inherently" by injecting the selection in all relevant property getters. For instance, if `data` contains 128 channels, 92 of which are...

pantaray

Parallel computation precision

2

At least for the coherence, sequential and parallel computing do not give the same results: ```python import numpy as np import syncopy as spy import syncopy.tests.synth_data as synth_data import dask.distributed...

tensionhead

It's a nice feature used by major packages like matplotlib: https://sphinx-gallery.github.io/stable/index.html Let's think how we could make use of that in our documentation. And when we're at it, we could...

tensionhead

Numpy ufunc and Syncopy data object

1

According to the [Numpy doc](https://numpy.org/doc/stable/reference/arrays.classes.html#numpy.class.__array_ufunc__) we could try to add `__ufunc_array__` methods to our data classes to automatically allow numpy ufuncs like `np.real`, `np.abs` and so on to work with...

tensionhead

Scalar Indexing with more than one dim broken

2

## The Problem Trying to find why the `--full` tests where always failing for `test_csd_arithmetic` I stumbled upon a general issue which surprisingly was not covered by any other test...

tensionhead

## Current State and Problem Right now, we are implicitly (down-)casting everything to double float precision (32bit for real, 64bit for complex). As touched upon in #264, we should at...

tensionhead

## Current Situation At the moment we have an overloaded `time` axis: holding both the actual samples of a trial which actually has a time axis, and the trials even...

tensionhead

There are many high-level concepts and rules within Syncopy, not all of them self explanatory. We should enhance the developer guide: https://syncopy.readthedocs.io/en/latest/developer/developers.html to cover topics like data class interfaces and...

tensionhead