milvus: [Bug]: Milvus crashed during building ANNOY index

Is there an existing issue for this?

  • I have searched the existing issues

Environment

- Milvus version: both v1.x and v2.0
- Deployment mode(standalone or cluster):
- SDK version(e.g. pymilvus v2.0.0rc2):
- OS(Ubuntu or CentOS): 
- CPU/Memory: 
- GPU: 
- Others:

Current Behavior

Run this script on 2.0:

from pymilvus import (
    connections,
    list_collections,
    FieldSchema, CollectionSchema, DataType,
    Collection, Index
)
import numpy as np

# configure milvus hostname and port
print(f"\nCreate connection...")
connections.connect()

# List all collection names
print(f"\nList collections...")
print(list_collections())

# Create a collection named 'demo_film_tutorial'
print(f"\nCreate collection...")
dim = 512
field1 = FieldSchema(name="release_year", dtype=DataType.INT64, description="int64", is_primary=True)
field2 = FieldSchema(name="embedding", dtype=DataType.FLOAT_VECTOR, description="float vector", dim=dim, is_primary=False)
schema = CollectionSchema(fields=[field1, field2], description="collection description")
collection = Collection(name='demo_film_tutorial', data=None, schema=schema)

print(f"\nInsert...")
num = 50000
data = [
    [i for i in range(num)],
    np.random.random([num, dim]).tolist(),
]
collection.insert(data)
print(collection.num_entities)

print(f"\nCreate index...")
index_params = {"index_type": "ANNOY", "metric_type": "IP", "params": {"n_trees": 10}}
index = Index(collection, "embedding", index_params)
print(index.params)

print([index.params for index in collection.indexes])

Milvus crashed in annoy lib.

Expected Behavior

No response

Steps To Reproduce

No response

Anything else?

No response

About this issue

  • Original URL
  • State: closed
  • Created 3 years ago
  • Comments: 42 (35 by maintainers)

Most upvoted comments

working on it