data-morph
data-morph copied to clipboard
stefmolin
Morph an input dataset of 2D points into select shapes, while preserving the summary statistics to a given number of decimal points through simulated annealing.
Consider switching the internal logic from pandas to polars, but still allow pandas `DataFrames` to be used as `Datasets`.
Currently, the docs are only built for the latest version of the code (HEAD), and previous versions are preserved automatically by virtue of publishing via the `gh-pages` branch. I'd like...
Fixes #217 **Describe your changes** - If a dataset was added/changed, morph it to shapes: `bullseye heart rectangle star slant_up` - If a shape was added/changed, morph starting from dataset...
**Is your feature request related to a problem? Please describe.** It'll be hard to parallelize a single morph, but when doing multiple at the same time, like with `data-morph --start-shape...
Some of the details are lost with the current padding (see the gorilla in #221): I would like to reduce the vertical and horizontal spacing so that we have a...
**Is your feature request related to a problem? Please describe.** When testing a new dataset or shape that someone proposes, I have to run the transformations on every possible combination...
Doing some benchmarks, I've noticed that the I/O parts are done in the main loop, which causes an increase in the total runtime. As an example, we have: ```plaintext time...
On local machines, both the multiprocessing logic introduced in https://github.com/stefmolin/data-morph/pull/273 and the GNU `parallel` command (https://github.com/stefmolin/data-morph/pull/222) provide significant speedups over running serially. However, this doesn't seem to work when running...