pycytominer

pycytominer copied to clipboard

[Enhancement] Improve performance/scaling by enabling an optional install of modin.

2

Certain pycytominer functions run into scaling issues for extremely large datasets. This can be solved by enabling an optional install of [modin](https://github.com/modin-project/modin) which is a drop-in replacement for pandas that...

kenibrewer

Standardize performance issue intake

1

Discussed in https://github.com/cytomining/pycytominer/issues/195#issuecomment-1217265110 The project could benefit from templating or standard description of issue reports (for example, machine/environment details, code samples, etc). For ex., see [Pandas Github issue templates](https://github.com/pandas-dev/pandas/tree/main/.github/ISSUE_TEMPLATE).

gwaybio

Provide Guidance or Tooling for Local Development Testing and Linting

1

This repo utilizes various testing and linting (including code-style) frameworks which are also sometimes incorporated as part of automated CI/CD workflows (for example, pytest). Developers could benefit from better understanding...

d33bs

Locate other performance bottlenecks, document, and add testing constraints

2

Discussed in https://github.com/cytomining/pycytominer/issues/195#issuecomment-1217265110 Add Memory/Time Testing Constraints for .merge_single_cells() (or others! 🙂 ) (Something like [pytest-memray](https://github.com/bloomberg/pytest-memray) or similar): automate targeted performance tests to ensure operations within this project are usable...

gwaybio

Collate.py fails with less number of compartments

4

# Issue: not having all compartments affects SingleCells and collate.py Sometimes we don't have all the compartments in an assay, for example having only stained the cells with a dye,...

fefossa

Running `consensus(operation="modz")` will silently drop groups (as specified in the argument `replicate_columns`) if one of the groups contains `NA` in one of the `replicate_columns`. The correct behavior would be to...

gwaybio

- Save results at intermediate steps, so that long calculations can't crash! - solve inconsistency around index and npz file metadata. Decide on one and stick to it - Build...

michaelbornholdt

Set epsilon to zero in `RobustMAD`

5

@gwaygenomics You may already know that we sometimes have [had issues](https://github.com/jump-cellpainting/pilot-analysis/issues/15#issuecomment-670640802)(private repo) with `mad_robustize` where one or few features end up being significantly larger than the other features, and then...

niranjchandrasekaran

Add feature generation to calculate heterogeneity

2

described in broadinstitute/cell-health#17

gwaybio

Detect compartments in the single cell input file and throw a warning if they are not used

2

In #111 - I add a method in the `cells.py` class to merge tables using predefined "linking" columns. If there are more than the default compartments in the provided input...

gwaybio