Quick start#

This section provides a quick introduction to Scipp. For in depth explanations refer to the sections in the user guide.

[1]:

import numpy as np
import scipp as sc

We start by creating some variables:

[2]:

var = sc.Variable(dims=['y', 'x'], values=np.random.rand(4,5))
sc.show(var)

Type the name of a variable at the end of a cell to generate an HTML respresentation:

[3]:

var

[3]:

scipp.Variable (416 Bytes)

(y: 4, x: 5)

float64

𝟙

0.639, 0.041, ..., 0.347, 0.409

Values:
array([[0.63888866, 0.04108235, 0.96318335, 0.3435437 , 0.49736929],
       [0.72490326, 0.64915203, 0.58315774, 0.90415743, 0.17200032],
       [0.68844817, 0.85674765, 0.56840341, 0.45739428, 0.76468145],
       [0.70321757, 0.73213711, 0.94100991, 0.34663324, 0.40933469]])

[4]:

x = sc.Variable(dims=['x'], values=np.arange(5), unit=sc.units.m)
y = sc.Variable(dims=['y'], values=np.arange(4), unit=sc.units.m)

We combine the variables into a data array:

[5]:

array = sc.DataArray(
    data=var,
    coords={'x': x, 'y': y})
sc.show(array)
array

[5]:

scipp.DataArray (1.48 KB)

Dimensions:
- y: 4
- x: 5
Coordinates: (2)
- x
  (x)
  int64
  m
  0, 1, 2, 3, 4
```
Values:
array([0, 1, 2, 3, 4])
```
- y
  (y)
  int64
  m
  0, 1, 2, 3
```
Values:
array([0, 1, 2, 3])
```

Data: (1)

(y, x)

float64

𝟙

0.639, 0.041, ..., 0.347, 0.409

Values:
array([[0.63888866, 0.04108235, 0.96318335, 0.3435437 , 0.49736929],
       [0.72490326, 0.64915203, 0.58315774, 0.90415743, 0.17200032],
       [0.68844817, 0.85674765, 0.56840341, 0.45739428, 0.76468145],
       [0.70321757, 0.73213711, 0.94100991, 0.34663324, 0.40933469]])

Variables can have uncertainties. Scipp stores these as variances (the square of the standard deviation):

[6]:

array.variances = np.square(np.random.rand(4,5))
sc.show(array)

We create a dataset:

[7]:

dataset = sc.Dataset(
    data={'a': var},
    coords={'x': x, 'y': y, 'aux': x})
dataset['b'] = array
dataset['scalar'] = 1.23 * (sc.units.m / sc.units.s)
sc.show(dataset)

We can slice variables, data arrays, and datasets using a dimension label and an index or a slice object like i:j:

[8]:

dataset['c'] = dataset['b']['x', 2]
sc.show(dataset)
dataset

We can also generate table representations (only 0-D and 1-D) and plots:

[9]:

sc.table(dataset['y', 2])

[9]:

		a	b
Coordinates		Data	Data
aux [m]	x [m]	[𝟙]	[𝟙]
0	0	0.688±0.919	0.688±0.919
1	1	0.857±0.748	0.857±0.748
2	2	0.568±0.521	0.568±0.521
3	3	0.457±0.236	0.457±0.236
4	4	0.765±0.748	0.765±0.748

[10]:

sc.plot(dataset['a'])

[10]:

../_images/getting-started_quick-start_17_0.svg

Arithmetic operations can be combined with slicing and handle propagation of uncertainties and units:

[11]:

print(dataset)

<scipp.Dataset>
Dimensions: Sizes[y:4, x:5, ]
Coordinates:
  aux                         int64              [m]  (x)  [0, 1, ..., 3, 4]
  x                           int64              [m]  (x)  [0, 1, ..., 3, 4]
  y                           int64              [m]  (y)  [0, 1, 2, 3]
Data:
  a                         float64  [dimensionless]  (y, x)  [0.638889, 0.0410824, ..., 0.346633, 0.409335]  [0.408395, 0.565323, ..., 0.407081, 0.351352]
  b                         float64  [dimensionless]  (y, x)  [0.638889, 0.0410824, ..., 0.346633, 0.409335]  [0.408395, 0.565323, ..., 0.407081, 0.351352]
  c                         float64  [dimensionless]  (y)  [0.963183, 0.583158, 0.568403, 0.94101]  [0.542325, 0.00451387, 0.271609, 0.00798375]
    Attributes:
        aux                         int64              [m]  ()  [2]
        x                           int64              [m]  ()  [2]
  scalar                    float64            [m/s]  ()  [1.23]

[12]:

dataset['b']['y', 0:2] -= dataset['y', 0:2]['a']['x', 0]
dataset['b'] *= dataset['scalar']
print(dataset)

<scipp.Dataset>
Dimensions: Sizes[y:4, x:5, ]
Coordinates:
  aux                         int64              [m]  (x)  [0, 1, ..., 3, 4]
  x                           int64              [m]  (x)  [0, 1, ..., 3, 4]
  y                           int64              [m]  (y)  [0, 1, 2, 3]
Data:
  a                         float64            [m/s]  (y, x)  [0, -0.735302, ..., 0.426359, 0.503482]  [1.23572, 1.47314, ..., 0.615872, 0.53156]
  b                         float64            [m/s]  (y, x)  [0, -0.735302, ..., 0.426359, 0.503482]  [1.23572, 1.47314, ..., 0.615872, 0.53156]
  c                         float64            [m/s]  (y)  [0.398882, -0.174347, 0.699136, 1.15744]  [1.43834, 0.613908, 0.410917, 0.0120786]
    Attributes:
        aux                         int64              [m]  ()  [2]
        x                           int64              [m]  ()  [2]
  scalar                    float64            [m/s]  ()  [1.23]

Finally, type the imported name of the Scipp module at the end of a cell for a list of all current Scipp objects (variables, data arrays, datasets). Click on entries to expand nested sections:

[13]:

sc

[13]:

<module 'scipp' from '/home/runner/work/scipp/scipp/.tox/docs/lib/python3.8/site-packages/scipp/__init__.py'>