krzysztof/landing_page
1
0
Fork 0
mirror of https://github.com/qdrant/landing_page.git synced 2026-10-04 18:38:30 +02:00
Code Issues Packages Projects Releases Wiki Activity

docs: Use local inference modern-sparse-neural-retrieval.md (#1600)

Browse Source 
* docs: Update to use local inference

Signed-off-by: Anush008 <anushshetty90@gmail.com>

* chore: Review updates

Signed-off-by: Anush008 <anushshetty90@gmail.com>

* chore: Formatting

Signed-off-by: Anush008 <anushshetty90@gmail.com>

---------

Signed-off-by: Anush008 <anushshetty90@gmail.com>
This commit is contained in:
Anush
2025-05-15 19:37:07 +05:30
committed by GitHub
parent 9577b88aac
commit 57fb83b9bc
5 changed files with 72 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
qdrant-landing/content/articles/modern-sparse-neural-retrieval.md
+65
View File
@@ -438,6 +438,7 @@ metadata = [{"movie_name": "The Passion of Joan of Arc", "movie_watch_time_min":
</details>
Upload embedded descriptions with movie metadata into the collection.
```python
qdrant_client.upsert(
collection_name="movies",
@@ -456,6 +457,34 @@ qdrant_client.upsert(
],
)
```
<aside role="status">
You can also implicitly generate sparse vectors using built-in FastEmbed integration.
</aside>
<details>
<summary>Implicitly generate sparse vectors (Click to expand)</summary>
```python
qdrant_client.upsert(
collection_name="movies",
points=[
models.PointStruct(
id=idx,
payload=metadata[idx],
vector={
"film_description": models.Document(
text=description, model=sparse_model_name
)
},
)
for idx, description in enumerate(descriptions)
],
)
```
</details>
#### Querying
Let’s query our collection!
@@ -472,6 +501,24 @@ response = qdrant_client.query_points(
)
print(response)
```
<details>
<summary>Implicitly generate sparse vectors (Click to expand)</summary>
```python
response = qdrant_client.query_points(
collection_name="movies",
query=models.Document(text="A movie about music", model=sparse_model_name),
using="film_description",
limit=1,
with_vectors=True,
with_payload=True,
)
print(response)
```
</details>
Output looks like this:
```bash
points=[ScoredPoint(
@@ -586,6 +633,24 @@ response = qdrant_client.query_points(
print(get_tokens_and_weights(response.points[0].vector['film_description'], tokenizer))
```
<details>
<summary>Implicitly generate sparse vectors (Click to expand)</summary>
```python
response = qdrant_client.query_points(
collection_name="movies",
query=models.Document(text="A movie about music", model=sparse_model_name),
using="film_description",
limit=1,
with_vectors=True,
with_payload=True,
)
print(get_tokens_and_weights(response.points[0].vector["film_description"], tokenizer))
```
</details>
And that's how SPLADE++ expanded the answer.
```python