my-favorite-things 0.5

Convenient functions and classes I use too often.

My Favorite Things

Convenient functions and classes I use too often. If Coltrane was a programmer (shudder) and much worse.

Installation

Install with

pip install my-favorite-things

Current Methods (by file)

save

save(name, savedir="", savepath="", stype="npz", absolute=False, parents=0, overwrite=False, append=False, dryrun=False, save_kwargs={}, **kwargs)

This method is used for saving data to a file. You can save as an .npz/.npy file for numpy array(s) or as a .pkl file for dictionaries and other odd python objects. By default, it will not overwrite existing files but instead append a number onto the end of file name (the keywords being, by default, overwite=False and append=True). You can save relative to your current directory (absolute=False) or as an absolute path (absolute=True). Addtionally, double check that you're saving to the correct directory with dryrun=True. Check the doc string for more info.

---

ddicts

nested_ddict(depth, endtype)

This method allows for creating a nested defaultdictionary. This is useful if you have data that is dependent on multiple parameters that are heirarchical. For example, if we do

d = nested_ddict(3, list)

then we can use it as

d['zero']['one']['two']['three'].append(datum)

format_ddict(ddict, make_nparr=True, sort_lists=False)

This method will format your (nested) defaultdictionary into dictionaries. Additionally, it can turns lists in numpy arrays and/or sort the lists too.

pprint_nested_dict(d, tab=2, k_format="", v_format="", sort=True, indentation=0)

Similar to Python's pprint, this will print out a dictionary such that each successive nested layer is more indented. It also has a means to format the final keys and values either with a method that they are passed to or an f-string.

---

plots

cumulative_bins(*arrs, num_bins)

This is similar to the previous method but for a linear scale. When plotting multiple data sets on the same plot, they may have different ranges and, thus, bin sizes. So this method will create the bin values so that the bars of the histogram will all have the same width.

log_bins(*arrs, num_bins)

This method is used for binning for histograms logarithmically. In plt.hist, setting the keyword bins to the output of this function (where arrs are the arrays being plotted) and ax.set_xscale("log") will give equally spaced bins (for multiple data sets over the logarithmic x-axis).

bar_count(ax, counts, labels, label_bars, sort_type, *, bar_params, **kwargs)

This method will create a bar plot for the data passed using strings as labels (either as the keys of a dictionary passed for counts or as a list passed from labels) with various conveniences like specifying the format of the label strings or the order of the plotted data.

histbar(ax, xs, ys, label_type=None, capends=None, fill=False, **kwargs)

Creates a step plot at the between the values in xs for the values in ys. So the former has $n+1$ values if the latter has $n$ values. Can fill in the plot via Matplotlib's fill_between using the boolean keyword fill and label the specific values of xs if they aren't uniform using label_type.

ScientificLocator(step=0.1)

This is a Locator class that will equally space ticks as step$\cdot 10^p$ where $p$ is the order of magnitude of the axis. That is, this is like MultipleLocator modulo the order of magnitude.

ScientificFormatter(useMathText=True)

An accompanying Formatter class to ScientificLocator. This class will force the axis to be in scientific notation.

---

colors

fader(color1, color2, fraction)

This method will return a color in hex code as a fraction between the two given colors color1 and color2.

multifader(colors, fractions)

Like above, but intermediate colors can also be defined in the colors list. These colors are equally spaced.

---

style

format_plots()

Changes the default values for Matplotlib plots.

format_numpy(precision=6, linewidth=1000, sign=" ")

Changes the default print options for Numpy.