Docs: Add tags
Browse files
README.md
CHANGED
|
@@ -1,4 +1,14 @@
|
|
| 1 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
|
| 3 |
SpaceV, initially published by Microsoft, is arguably the best dataset for large-scale Vector Search benchmarks.
|
| 4 |
It's large enough to stress-test indexing engines running across hundreds of CPU or GPU cores, significantly larger than the traditional [Big-ANN](https://big-ann-benchmarks.com/), which generally operates on just 10 million vectors.
|
|
@@ -95,4 +105,4 @@ from usearch.index import Index, BatchMatches
|
|
| 95 |
index = Index(ndim=100, metric="l2sq", dtype="i8")
|
| 96 |
index.add(ids, base)
|
| 97 |
matches: BatchMatches = index.search(queries)
|
| 98 |
-
```
|
|
|
|
| 1 |
+
---
|
| 2 |
+
task_categories:
|
| 3 |
+
- feature-extraction
|
| 4 |
+
tags:
|
| 5 |
+
- vector-search
|
| 6 |
+
- retrieval
|
| 7 |
+
size_categories:
|
| 8 |
+
- 100M<n<1B
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# SpaceV 100M
|
| 12 |
|
| 13 |
SpaceV, initially published by Microsoft, is arguably the best dataset for large-scale Vector Search benchmarks.
|
| 14 |
It's large enough to stress-test indexing engines running across hundreds of CPU or GPU cores, significantly larger than the traditional [Big-ANN](https://big-ann-benchmarks.com/), which generally operates on just 10 million vectors.
|
|
|
|
| 105 |
index = Index(ndim=100, metric="l2sq", dtype="i8")
|
| 106 |
index.add(ids, base)
|
| 107 |
matches: BatchMatches = index.search(queries)
|
| 108 |
+
```
|