krzysztof/landing_page
1
0
Fork 0
mirror of https://github.com/qdrant/landing_page.git synced 2026-09-29 07:58:31 +02:00
Code Issues Packages Projects Releases Wiki Activity

docs: Query API snippets with Python (#1095)

Browse Source 
Co-authored-by: generall <andrey@vasnetsov.com>
This commit is contained in:
Anush
2024-08-23 22:48:27 +05:30
committed by GitHub
co-authored by generall
parent 4821e18664
commit e9fa9be0a4
17 changed files with 203 additions and 195 deletions
Download Patch File Download Diff File Expand all files Collapse all files
qdrant-landing/content/documentation/concepts/explore.md
+68 -39
View File
@@ -43,11 +43,15 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
client.recommend(
client.query_points(
collection_name="{collection_name}",
positive=[100, 231],
negative=[718, [0.2, 0.3, 0.4, 0.5]],
strategy=models.RecommendStrategy.AVERAGE_VECTOR,
query=models.RecommendQuery(
recommend=models.RecommendInput(
positive=[100, 231],
negative=[718, [0.2, 0.3, 0.4, 0.5]],
strategy=models.RecommendStrategy.AVERAGE_VECTOR,
)
),
query_filter=models.Filter(
must=[
models.FieldCondition(
@@ -245,10 +249,14 @@ POST /collections/{collection_name}/points/query
```
```python
client.recommend(
client.query_points(
collection_name="{collection_name}",
positive=[100, 231],
negative=[718],
query=models.RecommendQuery(
recommend=models.RecommendInput(
positive=[100, 231],
negative=[718],
)
),
using="image",
limit=10,
)
@@ -350,15 +358,18 @@ POST /collections/{collection_name}/points/query
```
```python
client.recommend(
client.query_points(
collection_name="{collection_name}",
positive=[100, 231],
negative=[718],
query=models.RecommendQuery(
recommend=models.RecommendInput(
positive=[100, 231],
negative=[718],
)
),
using="image",
limit=10,
lookup_from=models.LookupLocation(
collection="{external_collection_name}",
vector="{external_vector_name}"
collection="{external_collection_name}", vector="{external_vector_name}"
),
)
```
@@ -520,13 +531,25 @@ filter_ = models.Filter(
)
recommend_queries = [
models.RecommendRequest(
positive=[100, 231], negative=[718], filter=filter_, limit=3
models.QueryRequest(
query=models.RecommendQuery(
recommend=models.RecommendInput(positive=[100, 231], negative=[718])
),
filter=filter_,
limit=3,
),
models.QueryRequest(
query=models.RecommendQuery(
recommend=models.RecommendInput(positive=[200, 67], negative=[300])
),
filter=filter_,
limit=3,
),
models.RecommendRequest(positive=[200, 67], negative=[300], filter=filter_, limit=3),
]
client.recommend_batch(collection_name="{collection_name}", requests=recommend_queries)
client.query_batch_points(
collection_name="{collection_name}", requests=recommend_queries
)
```
```typescript
@@ -776,18 +799,22 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
discover_queries = [
models.DiscoverRequest(
target=[0.2, 0.1, 0.9, 0.7],
context=[
models.ContextExamplePair(
positive=100,
negative=718,
),
models.ContextExamplePair(
positive=200,
negative=300,
),
],
models.QueryRequest(
query=models.DiscoverQuery(
discover=models.DiscoverInput(
target=[0.2, 0.1, 0.9, 0.7],
context=[
models.ContextPair(
positive=100,
negative=718,
),
models.ContextPair(
positive=200,
negative=300,
),
],
)
),
limit=10,
),
]
@@ -948,17 +975,19 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
discover_queries = [
models.DiscoverRequest(
context=[
models.ContextExamplePair(
positive=100,
negative=718,
),
models.ContextExamplePair(
positive=200,
negative=300,
),
],
models.QueryRequest(
query=models.ContextQuery(
context=[
models.ContextPair(
positive=100,
negative=718,
),
models.ContextPair(
positive=200,
negative=300,
),
],
),
limit=10,
),
]
qdrant-landing/content/documentation/concepts/search.md
+54 -64
View File
@@ -11,7 +11,6 @@ Searching for the nearest vectors is at the core of many representational learni
Modern neural networks are trained to transform objects into vectors so that objects close in the real world appear close in vector space.
It could be, for example, texts with similar meanings, visually similar pictures, or songs of the same genre.
{{< figure src="/docs/encoders.png" caption="This is how vector similarity works" width="70%" >}}
## Query API
@@ -36,7 +35,6 @@ Depending on the `query` parameter, Qdrant might prefer different strategies for
| [Multi-Stage Search](../hybrid-queries/#multi-stage-queries) | Optimize performance for large embeddings |
| [Random Sampling](#random-sampling) | Get random points from the collection |
**Nearest Neighbors Search**
```http
@@ -100,8 +98,8 @@ using Qdrant.Client;
var client = new QdrantClient("localhost", 6334);
await client.QueryAsync(
collectionName: "{collection_name}",
query: new float[] { 0.2f, 0.1f, 0.9f, 0.7f }
collectionName: "{collection_name}",
query: new float[] { 0.2f, 0.1f, 0.9f, 0.7f }
);
```
@@ -168,8 +166,8 @@ using Qdrant.Client;
var client = new QdrantClient("localhost", 6334);
await client.QueryAsync(
collectionName: "{collection_name}",
query: Guid.Parse("43cf51e2-8777-4f52-bc74-c2cbde0c8b04")
collectionName: "{collection_name}",
query: Guid.Parse("43cf51e2-8777-4f52-bc74-c2cbde0c8b04")
);
```
@@ -181,10 +179,10 @@ The choice of metric depends on the vectors obtained and, in particular, on the
Qdrant supports these most popular types of metrics:
* Dot product: `Dot` - https://en.wikipedia.org/wiki/Dot_product
* Cosine similarity: `Cosine` - https://en.wikipedia.org/wiki/Cosine_similarity
* Euclidean distance: `Euclid` - https://en.wikipedia.org/wiki/Euclidean_distance
* Manhattan distance: `Manhattan`* - https://en.wikipedia.org/wiki/Taxicab_geometry <i><sup>*Available as of v1.7</sup></i>
* Dot product: `Dot` - <https://en.wikipedia.org/wiki/Dot_product>
* Cosine similarity: `Cosine` - <https://en.wikipedia.org/wiki/Cosine_similarity>
* Euclidean distance: `Euclid` - <https://en.wikipedia.org/wiki/Euclidean_distance>
* Manhattan distance: `Manhattan`*- <https://en.wikipedia.org/wiki/Taxicab_geometry> <i><sup>*Available as of v1.7</sup></i>
The most typical metric used in similarity learning models is the cosine metric.
@@ -233,8 +231,9 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
client.search(
client.query_points(
collection_name="{collection_name}",
query=[0.2, 0.1, 0.9, 0.7],
query_filter=models.Filter(
must=[
models.FieldCondition(
@@ -246,7 +245,6 @@ client.search(
]
),
search_params=models.SearchParams(hnsw_ef=128, exact=False),
query_vector=[0.2, 0.1, 0.9, 0.7],
limit=3,
)
```
@@ -385,9 +383,10 @@ from qdrant_client import QdrantClient
client = QdrantClient(url="http://localhost:6333")
client.search(
client.query_points(
collection_name="{collection_name}",
query_vector=("image", [0.2, 0.1, 0.9, 0.7]),
query=[0.2, 0.1, 0.9, 0.7],
using="image",
limit=3,
)
```
@@ -466,7 +465,7 @@ You can still use payload filtering and other features of the search API with sp
There are however important differences between dense and sparse vector search:
| Index| Sparse Query | Dense Query |
| --- | --- | --- |
| --- | --- | --- |
| Scoring Metric | Default is `Dot product`, no need to specify it | `Distance` has supported metrics e.g. Dot, Cosine |
| Search Type | Always exact in Qdrant | HNSW is an approximate NN |
| Return Behaviour | Returns only vectors with non-zero values in the same indices as the query vector | Returns `limit` vectors |
@@ -490,15 +489,13 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
client.search(
client.query_points(
collection_name="{collection_name}",
query_vector=models.NamedSparseVector(
name="text",
vector=models.SparseVector(
indices=[1, 7],
values=[2.0, 1.0],
),
query=models.SparseVector(
indices=[1, 7],
values=[2.0, 1.0],
),
using="text",
limit=3,
)
```
@@ -598,9 +595,9 @@ POST /collections/{collection_name}/points/query
```
```python
client.search(
client.query_points(
collection_name="{collection_name}",
query_vector=[0.2, 0.1, 0.9, 0.7],
query=[0.2, 0.1, 0.9, 0.7],
with_vectors=True,
with_payload=True,
)
@@ -669,8 +666,8 @@ await client.QueryAsync(
);
```
You can use `with_payload` to scope to or filter a specific payload subset.
You can even specify an array of items to include, such as `city`,
You can use `with_payload` to scope to or filter a specific payload subset.
You can even specify an array of items to include, such as `city`,
`village`, and `town`:
```http
@@ -686,9 +683,9 @@ from qdrant_client import QdrantClient
client = QdrantClient(url="http://localhost:6333")
client.search(
client.query_points(
collection_name="{collection_name}",
query_vector=[0.2, 0.1, 0.9, 0.7],
query=[0.2, 0.1, 0.9, 0.7],
with_payload=["city", "village", "town"],
)
```
@@ -786,9 +783,9 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
client.search(
client.query_points(
collection_name="{collection_name}",
query_vector=[0.2, 0.1, 0.9, 0.7],
query=[0.2, 0.1, 0.9, 0.7],
with_payload=models.PayloadSelectorExclude(
exclude=["city"],
),
@@ -866,8 +863,8 @@ await client.QueryAsync(
```
It is possible to target nested fields using a dot notation:
- `payload.nested_field` - for a nested field
- `payload.nested_array[].sub_field` - for projecting nested fields within an array
* `payload.nested_field` - for a nested field
* `payload.nested_array[].sub_field` - for projecting nested fields within an array
Accessing array elements by index is currently not supported.
@@ -940,11 +937,11 @@ filter_ = models.Filter(
)
search_queries = [
models.SearchRequest(vector=[0.2, 0.1, 0.9, 0.7], filter=filter_, limit=3),
models.SearchRequest(vector=[0.5, 0.3, 0.2, 0.3], filter=filter_, limit=3),
models.QueryRequest(query=[0.2, 0.1, 0.9, 0.7], filter=filter_, limit=3),
models.QueryRequest(query=[0.5, 0.3, 0.2, 0.3], filter=filter_, limit=3),
]
client.search_batch(collection_name="{collection_name}", requests=search_queries)
client.query_batch_points(collection_name="{collection_name}", requests=search_queries)
```
```typescript
@@ -1113,9 +1110,9 @@ from qdrant_client import QdrantClient
client = QdrantClient(url="http://localhost:6333")
client.search(
client.query_points(
collection_name="{collection_name}",
query_vector=[0.2, 0.1, 0.9, 0.7],
query=[0.2, 0.1, 0.9, 0.7],
with_vectors=True,
with_payload=True,
limit=10,
@@ -1289,10 +1286,10 @@ POST /collections/{collection_name}/points/query/groups
```
```python
client.search_groups(
client.query_points_groups(
collection_name="{collection_name}",
# Same as in the regular search() API
query_vector=[1.1],
# Same as in the regular query_points() API
query=[1.1],
# Grouping parameters
group_by="document_id", # Path of the field to group by
limit=4, # Max amount of groups
@@ -1448,10 +1445,10 @@ POST /collections/chunks/points/query/groups
```
```python
client.search_groups(
client.query_points_groups(
collection_name="chunks",
# Same as in the regular search() API
query_vector=[1.1],
query=[1.1],
# Grouping parameters
group_by="document_id", # Path of the field to group by
limit=2, # Max amount of groups
@@ -1538,20 +1535,20 @@ using Qdrant.Client.Grpc;
var client = new QdrantClient("localhost", 6334);
await client.SearchGroupsAsync(
collectionName: "{collection_name}",
vector: new float[] { 1.0f },
groupBy: "document_id",
limit: 2,
groupSize: 2,
withLookup: new WithLookup
{
Collection = "documents",
WithPayload = new WithPayloadSelector
{
Include = new PayloadIncludeSelector { Fields = { new string[] { "title", "text" } } }
},
WithVectors = false
}
collectionName: "{collection_name}",
vector: new float[] { 1.0f },
groupBy: "document_id",
limit: 2,
groupSize: 2,
withLookup: new WithLookup
{
Collection = "documents",
WithPayload = new WithPayloadSelector
{
Include = new PayloadIncludeSelector { Fields = { new string[] { "title", "text" } } }
},
WithVectors = false
}
);
```
@@ -1616,7 +1613,6 @@ Random sampling API is a part of [Universal Query API](#query-api) and can be us
}
```
```python
from qdrant_client import QdrantClient, models
@@ -1679,15 +1675,9 @@ client
using Qdrant.Client;
using Qdrant.Client.Grpc;
var client = new QdrantClient("localhost", 6334);
await client.QueryAsync(
collectionName: "{collection_name}",
query: Sample.Random
);
await client.QueryAsync(collectionName: "{collection_name}", query: Sample.Random);
```
## Query planning
qdrant-landing/content/documentation/concepts/vectors.md
+4 -9
View File
@@ -328,16 +328,11 @@ from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
result = client.search(
result = client.query_points(
collection_name="{collection_name}",
query_vector=models.NamedSparseVector(
name="text",
vector=models.SparseVector(
indices=[1, 3, 5, 7],
values=[0.1, 0.2, 0.3, 0.4]
),
)
)
query_vector=models.SparseVector(indices=[1, 3, 5, 7], values=[0.1, 0.2, 0.3, 0.4]),
using="text",
).points
```
```rust