Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Run queries twice to test for idempotence #159

Merged
merged 4 commits into from
Dec 12, 2023
Merged

Run queries twice to test for idempotence #159

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
211182e
run queries twice to test for idempotence
jparismorgan Dec 11, 2023
57bb6e1
refactor query and check into helper function
jparismorgan Dec 11, 2023
86e5d3a
3 runs of query
jparismorgan Dec 11, 2023
3f0b3b7
add missing query()
jparismorgan Dec 11, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
126 changes: 72 additions & 54 deletions apis/python/test/test_index.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,9 @@
def test_flat_index(tmp_path):
uri = os.path.join(tmp_path, "array")
index = flat_index.create(uri=uri, dimensions=3, vector_type=np.dtype(np.uint8))
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {ind.MAX_UINT64} == set(result_i[0])
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {ind.MAX_UINT64} == set(result_i[0])
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Lets add a util function that runs x queries and asserts results to avoid duplication and make the code more readable.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, updated:

(TileDB-Vector-Search) ~/repo/TileDB-Vector-Search/apis/python pytest test/test_index.py -s
================================================= test session starts =================================================
platform darwin -- Python 3.9.18, pytest-7.4.3, pluggy-1.3.0
rootdir: /Users/parismorgan/repo/TileDB-Vector-Search/apis/python
collected 2 items                                                                                                     

test/test_index.py ..

================================================== 2 passed in 3.33s ==================================================


update_vectors = np.empty([5], dtype=object)
update_vectors[0] = np.array([0, 0, 0], dtype=np.dtype(np.uint8))
Expand All @@ -19,39 +20,47 @@ def test_flat_index(tmp_path):
update_vectors[3] = np.array([3, 3, 3], dtype=np.dtype(np.uint8))
update_vectors[4] = np.array([4, 4, 4], dtype=np.dtype(np.uint8))
index.update_batch(vectors=update_vectors, external_ids=np.array([0, 1, 2, 3, 4]))
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))

index.delete_batch(external_ids=np.array([1, 3]))
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))

update_vectors = np.empty([2], dtype=object)
update_vectors[0] = np.array([1, 1, 1], dtype=np.dtype(np.uint8))
update_vectors[1] = np.array([3, 3, 3], dtype=np.dtype(np.uint8))
index.update_batch(vectors=update_vectors, external_ids=np.array([1, 3]))
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {1, 2, 3}.issubset(set(result_i[0]))

index.delete_batch(external_ids=np.array([1, 3]))
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(np.array([[2, 2, 2]], dtype=np.float32), k=3)
assert {0, 2, 4}.issubset(set(result_i[0]))


def test_ivf_flat_index(tmp_path):
Expand All @@ -60,10 +69,11 @@ def test_ivf_flat_index(tmp_path):
index = ivf_flat_index.create(
uri=uri, dimensions=3, vector_type=np.dtype(np.uint8), partitions=partitions
)
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {ind.MAX_UINT64} == set(result_i[0])
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {ind.MAX_UINT64} == set(result_i[0])

update_vectors = np.empty([5], dtype=object)
update_vectors[0] = np.array([0, 0, 0], dtype=np.dtype(np.uint8))
Expand All @@ -72,52 +82,60 @@ def test_ivf_flat_index(tmp_path):
update_vectors[3] = np.array([3, 3, 3], dtype=np.dtype(np.uint8))
update_vectors[4] = np.array([4, 4, 4], dtype=np.dtype(np.uint8))
index.update_batch(vectors=update_vectors, external_ids=np.array([0, 1, 2, 3, 4]))
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))

index.delete_batch(external_ids=np.array([1, 3]))
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))

update_vectors = np.empty([2], dtype=object)
update_vectors[0] = np.array([1, 1, 1], dtype=np.dtype(np.uint8))
update_vectors[1] = np.array([3, 3, 3], dtype=np.dtype(np.uint8))
index.update_batch(vectors=update_vectors, external_ids=np.array([1, 3]))
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {1, 2, 3}.issubset(set(result_i[0]))

index.delete_batch(external_ids=np.array([1, 3]))
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))

index = index.consolidate_updates()
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))
for _ in range(2):
result_d, result_i = index.query(
np.array([[2, 2, 2]], dtype=np.float32), k=3, nprobe=partitions
)
assert {0, 2, 4}.issubset(set(result_i[0]))
Loading