Skip to content

[BUG] KNNMapperSearcherIT Failure When Building Index Remotely #30

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

Jump to bottom
Open
Tracked by #2553
anntians opened this issue Mar 21, 2025 · 4 comments
Open
Tracked by #2553

[BUG] KNNMapperSearcherIT Failure When Building Index Remotely #30

anntians opened this issue Mar 21, 2025 · 4 comments
Assignees
@anntians
Labels
bug Something isn't working

Comments

@anntians
Copy link

anntians commented Mar 21, 2025 •
edited
Loading

What is the bug?

While testing the KNN plugin KNNMapperSearcherIT against the remote index builder, the tests testKNNResultsUpdateDocAndForceMerge and testKNNResultsWithForceMerge failed due to a read time out after sending a build request to the remote index builder.

Looking at the logs for remote index builder, it rejects (422) several build requests coming from the addKnnDoc method when adding test data, which is expected as remote builder will reject build requests with doc_count <= 1 (ref: #26). Then it accepts the build request coming from the forceMergeKnnIndex method, and then is stuck, causing the tests to time out and fail.

Previously the index builder freezes due to incorrect test data, such as wrong dimensions/data type. However, the .knndid and .knnvec file size look correct with .knnvec 2x the size of .knndid due to the test vector dimensions are 2.

Remote index builder logs:

2025-03-17 22:47:07 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:41794 - "POST /_build HTTP/1.1" 422
2025-03-17 22:47:07 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:41794 - "POST /_build HTTP/1.1" 422
2025-03-17 22:47:07 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:41794 - "POST /_build HTTP/1.1" 422
2025-03-17 22:47:07 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:41794 - "POST /_build HTTP/1.1" 422
2025-03-17 22:47:07 - app.api.routes.build - INFO - Received build request: repository_type=<ObjectStoreType.S3: 's3'> container_name='tommy-test-vector-repo-us-east-1-temp' vector_path='vectors/QzGnC_RESBWDrJckLq-kJQ_vectors/TflKppUBHsUfMG62FBq0_test_vector__5.knnvec' doc_id_path='vectors/QzGnC_RESBWDrJckLq-kJQ_vectors/TflKppUBHsUfMG62FBq0_test_vector__5.knndid' tenant_id='integTestRemoteIndexBuild' dimension=2 doc_count=4 data_type=<DataType.FLOAT: 'float'> engine=<Engine.FAISS: 'faiss'> index_parameters=IndexParameters(algorithm=<Algorithm.HNSW: 'hnsw'>, space_type=<SpaceType.L2: 'l2'>, algorithm_parameters=AlgorithmParameters(ef_construction=100, ef_search=100, m=16))
2025-03-17 22:47:07 - app.services.job_service - INFO - Job id requirements: GPU memory: 8.65757465362549e-07, CPU memory: 5.871057510375977e-07
2025-03-17 22:47:07 - app.services.job_service - INFO - Added job to request store with job id: f51b838d2469af3c4143cd6a949e640916fab617eb7f270fb65846b2b9783a65
2025-03-17 22:47:07 - app.services.job_service - INFO - Worker resource status - GPU: 23.999999134242536 units, CPU: 63.99999941289425 units
2025-03-17 22:47:07 - app.executors.workflow_executor - INFO - Starting execution of job f51b838d2469af3c4143cd6a949e640916fab617eb7f270fb65846b2b9783a65
2025-03-17 22:47:07 - app.services.job_service - INFO - Successfully created workflow with job id: f51b838d2469af3c4143cd6a949e640916fab617eb7f270fb65846b2b9783a65
2025-03-17 22:47:07 - app.services.index_builder - INFO - Integration tests: None
      INFO   172.31.37.21:41794 - "POST /_build HTTP/1.1" 200
2025-03-17 22:47:07 - core.tasks - INFO - Starting task execution for vector path: vectors/QzGnC_RESBWDrJckLq-kJQ_vectors/TflKppUBHsUfMG62FBq0_test_vector__5.knnvec
2025-03-17 22:47:07 - core.tasks - INFO - Downloading vector and doc id blobs for vector path: vectors/QzGnC_RESBWDrJckLq-kJQ_vectors/TflKppUBHsUfMG62FBq0_test_vector__5.knnvec
   WARNING   Unsupported upgrade request.
2025-03-17 22:47:07 - uvicorn.error - WARNING - Unsupported upgrade request.
      INFO   172.31.37.21:41794 - "GET 
             /_status/f51b838d2469af3c4143cd6a949e640916fab617eb7f270fb65846b2b9
             783a65 HTTP/1.1" 200
2025-03-17 22:47:07 - botocore.credentials - INFO - Found credentials in environment variables.
2025-03-17 22:47:08 - core.tasks - INFO - Building GPU index for vector path: vectors/QzGnC_RESBWDrJckLq-kJQ_vectors/TflKppUBHsUfMG62FBq0_test_vector__5.knnvec
2025-03-17 22:47:08 - root - INFO - Setting number of parallel threads for gpu based graph build: 14
2025-03-17 22:47:08 - core.index_builder.create_gpu_index - INFO - Creating GPU Index.. with IVF_PQ
   WARNING   Unsupported upgrade request.
2025-03-17 22:47:08 - uvicorn.error - WARNING - Unsupported upgrade request.
      INFO   172.31.37.21:41794 - "GET 
             /_status/f51b838d2469af3c4143cd6a949e640916fab617eb7f270fb65846b2b9
             783a65 HTTP/1.1" 200
[W] [22:47:09.051201] Intermediate graph degree cannot be larger than dataset size, reducing it to 4
[W] [22:47:09.051233] Graph degree (32) cannot be larger than intermediate graph degree (3), reducing graph_degree.
using ivf_pq::index_params nrows 4, dim 2, n_lits 2, pq_dim 0
2025-03-17 22:47:09 - root - INFO - HNSW value of ef search is 256

How can one reproduce the bug?

  1. Clone the branch https://github.com/anntians/k-NN/tree/remote-index-build-debug
  2. Create S3 bucket {bucket_name}
  3. Setup remote index builder
// 1. Pull GPU docker image
docker pull rchitale7/remote-index-build-service:api

// 2. Set Environment Variables
export AWS_ACCESS_KEY_ID=
export AWS_SECRET_ACCESS_KEY=
export AWS_SESSION_TOKEN=

// 3. Run docker image

//GPU docker run command
docker run --gpus all -p 80:80 -e AWS_ACCESS_KEY_ID=${AWS_ACCESS_KEY_ID} -e AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY} -e AWS_SESSION_TOKEN=${AWS_SESSION_TOKEN} rchitale7/remote-index-build-service:api
  1. Run the test
./gradlew :integTestRemoteIndexBuild -Ds3.enabled=true -Dtest.remoteBuild=s3 -Dtest.bucket={bucket_name} -Dtest.base_path=vectors -Daccess_key=${AWS_ACCESS_KEY_ID} -Dsecret_key=${AWS_SECRET_ACCESS_KEY} -Dsession_token=${AWS_SESSION_TOKEN} --tests "org.opensearch.knn.index.KNNMapperSearcherIT.testKNNResultsUpdateDocAndForceMerge" -Dnproc.count=3

What is the expected behavior?

The expected behavior is the remote index builds successfully and uploads the .faiss file to S3 bucket for KNN to download

What is your host/environment?

g5.4xlarge EC2 instance, with Deep Learning Base OSS Nvidia Driver GPU AMI (Amazon Linux 2023) AMI

Do you have any screenshots?

If applicable, add screenshots to help explain your problem.

Do you have any additional context?

The other tests such as testKNNResultsWithoutForceMerge passes because remote build was not triggered without the forceMergeKnnIndex method call. Calls to addKnnDoc will cause 422 build request errors from remote index builder because _refresh is called, leading to build requests with doc_count <= 1
@anntians anntians added bug Something isn't working untriaged labels Mar 21, 2025
@anntians anntians mentioned this issue Mar 21, 2025
Closed
@anntians
Copy link
Author

I would like to work on this issue, thanks

@jed326 jed326 mentioned this issue Mar 5, 2025
Open
13 tasks
@jed326 jed326 removed the untriaged label Mar 26, 2025
@anntians
Copy link
Author

The doc_count is causing the error. When doc_count is 3 or 4, the error occurs. But when the doc_count is 5, 6 or 2, the remote build works and test pass. So if update here https://github.com/anntians/k-NN/blob/main/src/test/java/org/opensearch/knn/index/KNNMapperSearcherIT.java#L40-L52 to 5 docs, test passes

This issue doesn't seem to be related to vector dimension as the same behavior shows when testing vector dimensions 2, 3, and 10.

@anntians
Copy link
Author

The remote index builder is freezing at the method copyTo(cpuIndex) here https://github.com/rchitale7/remote-vector-index-builder/blob/api/remote_vector_index_builder/core/index_builder/create_gpu_index.py#L98. So the issue is stemming from the gpu to cpu index conversion process.

Looking at the logs below, the hyper parameters of the failed and successful build are the same, so the difference is potentially in the indexDataInIndex method here https://github.com/rchitale7/remote-vector-index-builder/blob/api/remote_vector_index_builder/core/index_builder/create_gpu_index.py#L85.

Logs for test run with doc_count 4, build fails

2025-03-27 00:21:39 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:33110 - "POST /_build HTTP/1.1" 422
2025-03-27 00:21:39 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:33110 - "POST /_build HTTP/1.1" 422
2025-03-27 00:21:39 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:33110 - "POST /_build HTTP/1.1" 422
2025-03-27 00:21:39 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:33110 - "POST /_build HTTP/1.1" 422
2025-03-27 00:21:40 - app.api.routes.build - INFO - Received build request: repository_type=<ObjectStoreType.S3: 's3'> container_name='tommy-test-vector-repo-us-east-1-temp' vector_path='vectors/2_Aos2oWS1SlNxloZU6Vqw_vectors/0mf51JUB-9Yjsw4p3ad__test_vector__5.knnvec' doc_id_path='vectors/2_Aos2oWS1SlNxloZU6Vqw_vectors/0mf51JUB-9Yjsw4p3ad__test_vector__5.knndid' tenant_id='integTestRemoteIndexBuild' dimension=2 doc_count=4 data_type=<DataType.FLOAT: 'float'> engine=<Engine.FAISS: 'faiss'> index_parameters=IndexParameters(algorithm=<Algorithm.HNSW: 'hnsw'>, space_type=<SpaceType.L2: 'l2'>, algorithm_parameters=AlgorithmParameters(ef_construction=100, ef_search=100, m=16))
2025-03-27 00:21:40 - app.services.job_service - INFO - Job id requirements: GPU memory: 8.65757465362549e-07, CPU memory: 5.871057510375977e-07
2025-03-27 00:21:40 - app.services.job_service - INFO - Added job to request store with job id: 42837e606317d69b48defdf9d1c165e34e5bcf4300c4636086629338e8c11454
2025-03-27 00:21:40 - app.services.job_service - INFO - Worker resource status - GPU: 23.999999134242536 units, CPU: 63.99999941289425 units
2025-03-27 00:21:40 - app.executors.workflow_executor - INFO - Starting execution of job 42837e606317d69b48defdf9d1c165e34e5bcf4300c4636086629338e8c11454
2025-03-27 00:21:40 - app.services.job_service - INFO - Successfully created workflow with job id: 42837e606317d69b48defdf9d1c165e34e5bcf4300c4636086629338e8c11454
2025-03-27 00:21:40 - app.services.index_builder - INFO - Integration tests: None
      INFO   172.31.37.21:33110 - "POST /_build HTTP/1.1" 200
2025-03-27 00:21:40 - core.tasks - INFO - Starting task execution for vector path: vectors/2_Aos2oWS1SlNxloZU6Vqw_vectors/0mf51JUB-9Yjsw4p3ad__test_vector__5.knnvec
2025-03-27 00:21:40 - core.tasks - INFO - Downloading vector and doc id blobs for vector path: vectors/2_Aos2oWS1SlNxloZU6Vqw_vectors/0mf51JUB-9Yjsw4p3ad__test_vector__5.knnvec
2025-03-27 00:21:40 - botocore.credentials - INFO - Found credentials in environment variables.
2025-03-27 00:21:40 - core.tasks - INFO - Building GPU index for vector path: vectors/2_Aos2oWS1SlNxloZU6Vqw_vectors/0mf51JUB-9Yjsw4p3ad__test_vector__5.knnvec
2025-03-27 00:21:40 - root - INFO - Setting number of parallel threads for gpu based graph build: 14
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.intermediate_graph_degree 64
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.graph_degree 32
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.device 0
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.store_dataset False
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.refine_rate 2.0
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.build_algo 0
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.kmeans_n_iters 20
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.pq_bits 8
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.pq_dim 0
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.n_lists 2
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.kmeans_trainset_fraction 0.5
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.force_random_rotation True
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.conservative_memory_allocation True
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - cagraIndexSearchIVFPQConfig.n_probes 20
2025-03-27 00:21:40 - core.index_builder.create_gpu_index - INFO - Creating GPU Index.. with IVF_PQ
[W] [00:21:41.226083] Intermediate graph degree cannot be larger than dataset size, reducing it to 4
[W] [00:21:41.226113] Graph degree (32) cannot be larger than intermediate graph degree (3), reducing graph_degree.
using ivf_pq::index_params nrows 4, dim 2, n_lits 2, pq_dim 0
2025-03-27 00:21:41 - root - INFO - HNSW value of ef search is 256
2025-03-27 00:21:41 - root - INFO - before cagraIndex.copyTo
2025-03-27 00:21:41 - root - INFO - Made updates 2

Logs for test run with doc_count 5, build succeeds

2025-03-27 00:25:21 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:60842 - "POST /_build HTTP/1.1" 422
2025-03-27 00:25:21 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:60842 - "POST /_build HTTP/1.1" 422
2025-03-27 00:25:22 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:60842 - "POST /_build HTTP/1.1" 422
2025-03-27 00:25:22 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:60842 - "POST /_build HTTP/1.1" 422
2025-03-27 00:25:22 - testing_api.app.main - INFO - Error while validating parameters: #[{'field': 'body.doc_count', 'message': 'Input should be greater than 1', 'type': 'greater_than'}]
      INFO   172.31.37.21:60842 - "POST /_build HTTP/1.1" 422
2025-03-27 00:25:22 - app.api.routes.build - INFO - Received build request: repository_type=<ObjectStoreType.S3: 's3'> container_name='tommy-test-vector-repo-us-east-1-temp' vector_path='vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec' doc_id_path='vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knndid' tenant_id='integTestRemoteIndexBuild' dimension=2 doc_count=5 data_type=<DataType.FLOAT: 'float'> engine=<Engine.FAISS: 'faiss'> index_parameters=IndexParameters(algorithm=<Algorithm.HNSW: 'hnsw'>, space_type=<SpaceType.L2: 'l2'>, algorithm_parameters=AlgorithmParameters(ef_construction=100, ef_search=100, m=16))
2025-03-27 00:25:22 - app.services.job_service - INFO - Job id requirements: GPU memory: 1.0821968317031862e-06, CPU memory: 7.338821887969972e-07
2025-03-27 00:25:22 - app.services.job_service - INFO - Added job to request store with job id: b13eed1e37514f18bd3aa2626332c8c02bcd0741fab9b05c3a767bd8d9183720
2025-03-27 00:25:22 - app.services.job_service - INFO - Worker resource status - GPU: 23.99999891780317 units, CPU: 63.99999926611781 units
2025-03-27 00:25:22 - app.executors.workflow_executor - INFO - Starting execution of job b13eed1e37514f18bd3aa2626332c8c02bcd0741fab9b05c3a767bd8d9183720
2025-03-27 00:25:22 - app.services.job_service - INFO - Successfully created workflow with job id: b13eed1e37514f18bd3aa2626332c8c02bcd0741fab9b05c3a767bd8d9183720
      INFO   172.31.37.21:60842 - "POST /_build HTTP/1.1" 200
2025-03-27 00:25:22 - app.services.index_builder - INFO - Integration tests: None
2025-03-27 00:25:22 - core.tasks - INFO - Starting task execution for vector path: vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec
2025-03-27 00:25:22 - core.tasks - INFO - Downloading vector and doc id blobs for vector path: vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec
2025-03-27 00:25:22 - botocore.credentials - INFO - Found credentials in environment variables.
2025-03-27 00:25:23 - core.tasks - INFO - Building GPU index for vector path: vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec
2025-03-27 00:25:23 - root - INFO - Setting number of parallel threads for gpu based graph build: 14
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.intermediate_graph_degree 64
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.graph_degree 32
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.device 0
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.store_dataset False
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.refine_rate 2.0
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexConfig.build_algo 0
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.kmeans_n_iters 20
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.pq_bits 8
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.pq_dim 0
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.n_lists 2
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.kmeans_trainset_fraction 0.5
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.force_random_rotation True
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexIVFPQConfig.conservative_memory_allocation True
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - cagraIndexSearchIVFPQConfig.n_probes 20
2025-03-27 00:25:23 - core.index_builder.create_gpu_index - INFO - Creating GPU Index.. with IVF_PQ
[W] [00:25:23.909501] Intermediate graph degree cannot be larger than dataset size, reducing it to 5
[W] [00:25:23.909522] Graph degree (32) cannot be larger than intermediate graph degree (4), reducing graph_degree.
using ivf_pq::index_params nrows 5, dim 2, n_lits 2, pq_dim 0
2025-03-27 00:25:24 - root - INFO - HNSW value of ef search is 256
2025-03-27 00:25:24 - root - INFO - before cagraIndex.copyTo
2025-03-27 00:25:24 - root - INFO - Made updates 2
2025-03-27 00:25:24 - root - INFO - after cagraIndex.copyTo
2025-03-27 00:25:24 - root - INFO - Before faiss.write_index
2025-03-27 00:25:24 - root - INFO - /tmp/tmpnksvqraz/vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec
2025-03-27 00:25:24 - root - INFO - <faiss.swigfaiss.IndexIDMap; proxy of <Swig Object of type 'faiss::IndexIDMapTemplate< faiss::Index > *' at 0x7f0cdd6fc0c0> >
2025-03-27 00:25:24 - root - INFO - After faiss.write_index
2025-03-27 00:25:24 - core.index_builder.create_gpu_index - INFO - Write completed
2025-03-27 00:25:24 - core.tasks - INFO - Uploading index for vector path: vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec
2025-03-27 00:25:24 - core.tasks - INFO - Ending task execution for vector path: vectors/OcfpfblJQDSNpbP2wK4ZQQ_vectors/ofj91JUBIWKcbLMgQ1dV_test_vector__6.knnvec
2025-03-27 00:25:24 - app.services.index_builder - INFO - Index built successfully!
2025-03-27 00:25:24 - app.services.index_builder - INFO - Index path: ofj91JUBIWKcbLMgQ1dV_test_vector__6.faiss
2025-03-27 00:25:24 - app.executors.workflow_executor - INFO - Job b13eed1e37514f18bd3aa2626332c8c02bcd0741fab9b05c3a767bd8d9183720 completed with status JobStatus.COMPLETED

@anntians
Copy link
Author

Faiss Github Issue: facebookresearch/faiss#4260

@rchitale7 rchitale7 mentioned this issue Apr 12, 2025
Merged
@anntians anntians mentioned this issue Apr 14, 2025
Open
5 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@anntians anntians

Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@jed326 @anntians