https://github.com/climateimpactlab/impactlab-api-mockup
Prototype library for distributed computing with embedded mathematical documentation
https://github.com/climateimpactlab/impactlab-api-mockup
Last synced: 2 months ago
JSON representation
Prototype library for distributed computing with embedded mathematical documentation
- Host: GitHub
- URL: https://github.com/climateimpactlab/impactlab-api-mockup
- Owner: ClimateImpactLab
- Created: 2016-09-23T08:37:41.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2016-10-11T20:55:59.000Z (over 9 years ago)
- Last Synced: 2025-09-10T03:52:51.737Z (7 months ago)
- Language: Jupyter Notebook
- Size: 188 KB
- Stars: 0
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Prototype Climate Impact Lab Interface
## Workspace Setup
Import some python modules
```python
%matplotlib inline
from matplotlib import pyplot as plt
import prototype
```
## Connect to the API
Authenticate with the [prototype] API, and define how you want to use the system
```python
api = prototype.ClimateImpactLabDataAPI(
username='mdelgado',
password='*********')
api.configure(sample='median', location='OSDC', priority='low')
```
API configuration updated
## Retrieve formatted Climate Impact Lab variables
```python
# mortality regression parameters
alpha = api.get_variable('alpha')
gamma1 = api.get_variable('gamma1')
gamma2 = api.get_variable('gamma2')
gamma3 = api.get_variable('gamma3')
# GCP-wide socioeconomic parameters
avg_days_per_bin = api.get_variable('avg_days_per_bin')
gdp_per_capita = api.get_variable('gdppc')
pop_density = api.get_variable('popdens')
# Climate variables
temp = api.get_variable('temp')
```
```python
# Get variable attributes
temp.attrs
```
``OrderedDict([('symbol', 'T'), ('description', 'NASA downscaled climate data')])``
## Do math with Impact Lab variables
```python
# From the NAS mortality presentation...
betahat = alpha+gamma1*avg_days_per_bin+gamma2*(gdp_per_capita.ln())+gamma3*(pop_density.ln())
```
```python
# define how your new variable should look
betahat.symbol = '\\hat{{\\beta}}'
```
```python
# show LaTeX for the math we just did
betahat.display()
```
![betahat][]
[betahat]: http://latex.codecogs.com/gif.latex?%24%5Cbeta_%7Bbins%2Cadm2%2Ctime%7D%20%3D%20%5Calpha_%7Bbins%7D%20+%20%5Cleft%28%7B%7B%5Cgamma_1%7D%7D_%7Bbins%7D%5Cright%29%5Cleft%28AvgDaysPerBin_%7Bbins%2Cadm2%2Ctime%7D%5Cright%29%20+%20%5Cleft%28%7B%7B%5Cgamma_2%7D%7D_%7Bbins%7D%5Cright%29%5Cleft%28%5Cln%7B%5Cleft%28GdpPC_%7Badm2%2Ctime%7D%5Cright%29%7D%5Cright%29%20+%20%5Cleft%28%7B%7B%5Cgamma_3%7D%7D_%7Bbins%7D%5Cright%29%5Cleft%28%5Cln%7B%5Cleft%28PopDensity_%7Badm2%2Ctime%7D%5Cright%29%7D%5Cright%29%24
```python
# Also, show the result of the calculation
betahat.value
```
array([[[ 0.13070758, 0.27115357, 0.51208013, ..., 0.75042949,
0.97468369, 0.38654286],
[ 0.45469313, -0.58699997, 0.89999766, ..., 0.0932762 ,
0.27909178, 0.25180471],
[ 0.78249511, 0.6007064 , 0.31142451, ..., 0.7384293 ,
0.50259999, 0.89219784],
...,
[ 0.09726135, 0.53286143, -0.23998856, ..., 0.22746566,
1.01581255, 0.27642734],
[ 0.94900355, 0.95514836, 0.81664952, ..., 0.18979607,
0.64628862, 0.35505172],
[-0.28351735, 0.29136219, 0.96207262, ..., -0.39525938,
0.52056722, 0.26820543]],
[[-0.24982154, 0.25459062, 0.18381526, ..., 0.86366869,
1.15629761, 0.52607199],
[ 0.04221185, -0.69654775, 0.49941067, ..., -0.2538949 ,
0.19605259, 0.12404486],
[ 0.68108903, 0.38304013, 0.29892101, ..., 0.66033363,
0.65525907, 0.64782637],
...,
[ 0.10886929, 0.11420064, -0.76898607, ..., 0.52976564,
0.81398773, -0.02587275],
[ 0.49060749, 0.65472517, 0.90772419, ..., 0.11534893,
0.19195818, 0.27063844],
[-0.82047299, -0.03915025, 0.62043444, ..., -0.33573926,
0.61454831, 0.20731075]],
[[ 0.29334982, -0.1410393 , 0.06668666, ..., 0.19058335,
0.38035987, -0.76053658],
[ 0.24590676, -0.19187054, 0.33738993, ..., -0.03573067,
0.05934987, -0.36174642],
[ 0.14058797, 0.22807228, 0.02445355, ..., 0.38002263,
-0.05175731, 0.18988073],
...,
[ 0.26074214, 0.27289792, -0.56491045, ..., -0.66360662,
0.47412097, -0.02909703],
[ 0.23397792, 0.41959969, 0.36822878, ..., -1.09986828,
0.2577988 , -0.15395848],
[-0.01366349, -0.41443745, 0.41451397, ..., -0.53642207,
0.06734171, 0.14076383]],
...,
[[-1.22909687, -0.16171115, -0.35495874, ..., 0.03058367,
0.83641502, 0.6679868 ],
[-0.64060209, -2.40039114, 0.61196964, ..., -0.91890489,
-0.67590221, -0.14152328],
[ 0.72354879, 0.41586095, -0.04011696, ..., 0.4473447 ,
0.15337609, 0.45083991],
...,
[-0.93966302, -0.38880867, -1.61066037, ..., 0.12229467,
0.84788684, -0.44109869],
[ 0.9050935 , 0.50790203, 0.37243353, ..., 0.70626393,
-0.1332156 , -0.12347863],
[-2.13993911, -0.13466855, 0.82092427, ..., -2.03681629,
0.16071452, -0.66681163]],
[[-1.78127977, -0.64995827, -0.70599218, ..., -0.08563787,
0.53846099, -0.03527768],
[-0.96737409, -3.3405469 , 0.12035706, ..., -1.24401243,
-1.18152117, -1.02750102],
[ 0.43426372, 0.06269273, -0.77531073, ..., -0.07509919,
-0.01719537, 0.41981726],
...,
[-1.69569958, -0.67028133, -2.24069923, ..., -0.74557355,
0.78917575, -0.96980891],
[ 0.54463694, 0.35235451, 0.05831266, ..., -0.09088725,
-0.45957768, -0.72981302],
[-2.66118488, -0.83070028, 0.461331 , ..., -2.61814683,
-0.45347083, -1.47672656]],
[[ 0.05461305, 0.79718925, 0.60464898, ..., 0.8691447 ,
1.29582956, 1.03214796],
[ 0.50555223, -0.341628 , 1.15139969, ..., 0.33088638,
0.35190861, 0.4681457 ],
[ 1.23498927, 0.70501668, 0.74332358, ..., 1.03057786,
0.87111314, 1.15050731],
...,
[ 0.26777543, 0.43157869, -0.0971003 , ..., 0.81305677,
1.24930213, 0.59565844],
[ 0.94444546, 1.06483473, 0.81508081, ..., 0.84482893,
0.82771585, 0.52695798],
[-0.36370791, 0.70637385, 1.00317956, ..., 0.02623079,
0.67527143, 0.47231838]]])
Coordinates:
* bins (bins) int32 0 1 2 3 4 5 6 7 8 9 10 11
* adm2 (adm2) int32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
* time (time) int32 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 ...
Attributes:
symbol: \hat{{\beta}}
## Use functions from xArray and the rest of python
```python
ax = betahat.value.plot()
plt.show(ax)
```
## The next step in the calculation...
```python
# simplify the representation of betahat:
betahat.symbolic = '\\hat{{\\beta}}'
```
```python
# sum betahat*temp across bins
mortality = (betahat * temp).sum(dim='bins')
# declare a symbol for mortality
mortality.symbol = 'M'
```
```python
mortality.display()
```
![mortality][]
[mortality]: http://latex.codecogs.com/gif.latex?%24M_%7Badm2%2Ctime%7D%20%3D%20%5Csum_%7Bbins%7D%7B%5Cleft%7B%5Cleft%28%5Chat%7B%7B%5Cbeta%7D%7D%7Bbins%2Cadm2%2Ctime%7D%5Cright%29%5Cleft%28T%7Bbins%2Cadm2%2Ctime%7D%5Cright%29%5Cright%7D%7D%24
Note that the dimensionality of mortality is less than beta and temp in the LaTeX and data:
```python
mortality.value
```
array([[ -2.92124395e+00, -1.99362992e+00, -8.93652801e-01, ...,
1.10621390e+00, 2.57830688e+00, -2.29306181e+00],
[ -1.38629855e+00, -7.76493092e+00, 2.03159987e+00, ...,
-3.41893630e+00, -4.84213829e+00, -2.45017356e+00],
[ 1.56519757e+00, -2.41087398e-01, -4.24496624e-01, ...,
1.36587685e+00, -5.93521188e-01, 2.76255544e+00],
...,
[ -3.31492736e+00, 6.77445891e-03, -5.47755159e+00, ...,
-6.10378895e+00, 3.93503051e+00, -3.73371494e+00],
[ 3.06004915e+00, 2.91296446e+00, 3.59191665e+00, ...,
-7.61977078e+00, 5.19491826e-01, -1.90245230e+00],
[ -7.99817007e+00, -4.96106865e+00, 3.18680168e+00, ...,
-8.19163235e+00, -8.54668938e-01, -8.54509732e-01]])
Coordinates:
* adm2 (adm2) int32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ...
* time (time) int32 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 ...
Attributes:
symbol: M
```python
ax = mortality.value.mean(dim='adm2').plot()
plt.show(ax)
```
