File: README.md

package info (click to toggle)
networkx 3.4.2-4
  • links: PTS, VCS
  • area: main
  • in suites: sid
  • size: 11,700 kB
  • sloc: python: 105,310; xml: 544; makefile: 131; javascript: 120; sh: 34
file content (26 lines) | stat: -rw-r--r-- 866 bytes parent folder | download | duplicates (3)
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
## NetworkX ASV Benchmarks

Currently we have two broad categories for benchmarks:

- `benchmarks_classes.py`: This covers benchmarks for performance of core
  NetworkX classes (`Graph`, `DiGraph`, `MultiGraph`, `MultiDiGraph`).
  We test performance of basic graph operations (`add_node`, `add_edge`, `subgraph`
  and others). Code that touches the core classes should run these benchmarks
  to make sure if there is a performance hit or gain.
- `benchmark_algorithms.py`: This runs a tiny subset of algorithms which cover
  different parts of our codebase. We use graphs with different density, and a
  real world dataset from SNAP datasets to run the algorithms.

## Running the benchmark suite

To run the benchmark on the current HEAD commit:

```
asv run
```

Run the benchmark to compare two commits:

```
asv continuous base_commit_hash test_commit_hash
```