Getting Started with Qdrant

Qdrant is a vector database and a similarity search engine. With Qdrant, embeddings or neural network encoders can be turned into full-fledged applications for matching, searching, recommending, and much more!

In this basic tutorial, you will use Qdrant to generate semantic search results and recommendations from a sample dataset.

[python]

This is valuable knowledge, thank you for sharing. I like that Qdrant is written in Rust as well!

[BuiHungHoward]

could you update for method qdrant client -> method recommend -> argument: query_vector seem be changed?

[sriramananthakrishnan]

This is an excellent tutorial ! Just one suggestion, can you please, the next time, use the words from the payload to create the embeddings for the vectors and then pad it as well to create vectors of n=100 dimensions and then show the same example. It will help people like me understand easy to use next steps to embed complete documents.

[fabsync]

Love your teaching style! One question.. if I have millions of data points and I want to create and save some query results.. so that I can show my clients something while I am creating/retrieving the new data query results…how would you go about doing that ? Would you do that with Qdrant only or will you combine it with another db?

[greendsnow]

More human way of adding your vectors is like this:

points = []
for i in range(len(vectors))
point = models.PointStruct(
id=1,
vector={
"image": [0.9, 0.1, 0.1, 0.2],
"text": [0.4, 0.7, 0.1, 0.8, 0.1, 0.1, 0.9, 0.2],
},
),
points.append(point)

client.upsert(
collection_name="{collection_name}",
points=points
)

[darogajee]

very compliicated tutorial ever
from 1:00 time you are writitng everything in docker

[wsjczqz]

from qdrant_client import QdrantClient
from qdrant_client.http import models

client.upsert(
collection_name="point_example",
points=models.Batch(
ids=[1, 2, 3],
vectors=[
[0.9, 0.1, 0.1, 0.5, 0.6],
[0.1, 0.9, 0.1, 0.2, 0.4],
[0.1, 0.1, 0.9, 0.5, 0.2],
]
),
)

After run this code,

collection_name = "insert_for_point_examle"
vector_id = 3

# 특정 ID의 벡터 조회
client.retrieve(
collection_name=collection_name,
ids=[1],
with_vectors = True
)

i checked my data id=1, but result is [Record(id=1, payload={}, vector=[0.9878784, 0.10976427, 0.10976427])]

Why not [0.9, 0.1, 0.1, 0.5, 0.6]..?