ahh - Andrew Huang Helps? v1.0.3

Functions that I can easily reference, and maybe you too!

Note that this package is now deprecated! Please see https://github.com/ahuang11/holoext for my new project!

HOW TO GET IT:

New-school method:

1. pip install ahh

2. Ensure your packages version (pip list) match with ones listed in requirements.txt

Old-school method:

1. Type git clone https://github.com/ahuang11/ahh.git

2. Go into ahh folder (where setup.py is)

3. Type pip install -e . (may need to be in bash first!)

4. In a Python script, type from ahh import pre, era, vis, sci, ext

MOTIVATION TO GET STARTED:

Perhaps you've wrote these lines one too many times just to create a map that looks like this.

import matplotlib.pyplot as plt
import cartopy.crs as ccrs
import cartopy.feature as cfeature

fig = plt.figure(figsize=(12, 8))
projection = ccrs.PlateCarree()
ax = plt.axes(projection=projection)
im = ax.contourf(da.lon, da.lat, da[0].values, transform=projection, cmap='RdBu_r')
_ = plt.colorbar(im, orientation='horizontal', shrink=0.5, pad=0.05)
_ = ax.gridlines(crs=ccrs.PlateCarree(), draw_labels=True)
_ = ax.coastlines()
plt.grid()

With this package, it's simplified down to just this (and a prettier map)!

from ahh import vis

ax = vis.plot_map(da[0], da.lat, da.lon)

Or perhaps you don't deal too much with maps, but you do a lot with timeseries plots.

import matplotlib.pyplot as plt

plt.plot(x, y, label='Blue', linestyle='--')
plt.legend() # equivalent to below

Compare that with this package's version!

from ahh import vis

vis.plot_line(x, y, label='Red', linestyle='--')

Best of all, the functions are super flexible (and have a lot of cool features built into them!)

from ahh import vis

vis_dict = dict(rows=2, # specify number of subplots
                sidebar_count=2, # specify number of side by side bars
                xlabel='x', title='Vis Demonstration', suptitle=True,
                title_pad=0.9, figsize='na') # set shared settings

vis.set_figsize(15, 8) # set figure size
vis.plot_bar(x, y, label='Red Bars', ylabel='Values', **vis_dict)
vis.plot_bar(x, y2, label='Blue Bars', sidebar_pos=2, color='blue', **vis_dict)

vis.plot_hist(['Chicken'] * 5 + ['Egg'] * 2 + ['Spam'], # random data
              rows=2, pos=2, ptype='bar',
              ylabel='Count', color='orange', # labels and color
              cumsum=True, save='vis_demonstration') # get cumulative count and save

But data exploration doesn't begin with visualizations! Check out ext.ahh() too!

from ahh import ext

ext.ahh(ds=arr_ds, arr=x) # name=data

            Name: ds
          Length: 366
      Dimensions: (366, 73, 144)
   Unnested Type: <class 'numpy.float64'>
Overarching Type: <class 'numpy.ndarray'>
Minimum, Maximum: 188.290, 315.300
 Average, Median: 277.191, 282.670

Snippet of values:
[ 238.1  238.1  238.1  238.1  238.1 ...,  238.1  238.1  238.1  238.1  238.1]


            Name: arr
          Length: 15
      Dimensions: (15,)
   Unnested Type: <class 'numpy.int64'>
Overarching Type: <class 'numpy.ndarray'>
Minimum, Maximum: 1.000, 15.000
 Average, Median: 8.000, 8.000

Snippet of values:
[ 1  2  3  4  5  6  7  8  9 10 11 12 13 14 15]

Interested in seeing more? There's way more in the examples page!

AVAILABLE MODULES:

Documentation here: https://ahuang11.github.io/ahh/

Random, but awesome, tips: https://github.com/ahuang11/ahhsumtips

KNOWN ISSUES AND HACKS:

• Installation of cartopy/basemap can be a bit tedious; easy way to install cartopy and basemap if you have Anaconda: 'conda install -c conda-forge cartopy' and 'conda install -c anaconda basemap'
• Updates to packages listed in requirements may break this package; recommend creating a Python 3 environment and install the specific package versions listed in requirements.txt
• In version 0.6.2, I accidentally bloated the repo so I decided to reset the repo and start semantic versioning beginning at v1.0.0; you may want to rm -rf ahh and git clone https://github.com/ahuang11/ahh.git again (or newly available, pip install ahh)!