1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
|
Converting arrays
- NumPy
- Python objects
- JSON
- ROOT via Uproot
- Parquet via Arrow
- Pandas: Awkward columns
- Pandas: as MultiIndex
- Zarr
Creating arrays
- ArrayBuilder (easy & general)
- Layout (faster)
- Lazy arrays and Dask
Examining arrays
- Data type
- Single item detail
- Listing fields/keys/columns
- Simple slicing
- Checking validity
Numerical math
- NumPy functions
- Awkward broadcasting
- Reducing (sum/min/any/all)
- Statistics (mean/var/std)
- Using argmin/argmax
- On GPUs
Filtering data
- By number of items
- Cuts vs. masks
- Slicing lists within arrays
- Slices with missing values
Restructuring data
- Zip/unzip and project
- Adding fields to records
- Renaming records
- Flattening for plots
- Padding/clipping for machine learning
- Concatenating and interleaving
- Sorting
Combinatorics
- Cartesian/cross product and "n choose k"
- Best match between collections
Specialized behavior
- Subclassing Array/Record
- Overriding NumPy functions
- In Numba
- For physics: Lorentz vectors
|
