diff --git a/vsb/workloads/dataset.py b/vsb/workloads/dataset.py index 60211b0..3ca0af7 100644 --- a/vsb/workloads/dataset.py +++ b/vsb/workloads/dataset.py @@ -13,7 +13,7 @@ class Dataset: """ Represents a Dataset used as the source of documents and/or queries for - Pinecone index operations. + Vector Search operations. The set of datasets are taken from the Pinecone public datasets (https://docs.pinecone.io/docs/using-public-datasets), which reside in a Google Cloud Storage bucket and are downloaded on-demand on first access, diff --git a/vsb/workloads/parquet_workload/parquet_workload.py b/vsb/workloads/parquet_workload/parquet_workload.py index a76e51f..28defbb 100644 --- a/vsb/workloads/parquet_workload/parquet_workload.py +++ b/vsb/workloads/parquet_workload/parquet_workload.py @@ -1,10 +1,12 @@ +from abc import ABC + from ..base import VectorWorkload from ..dataset import Dataset from ...databases.base import Index from ...vsb_types import Record, SearchRequest -class ParquetWorkload(VectorWorkload): +class ParquetWorkload(VectorWorkload, ABC): """A static workload which is implemented by reading records and query from two sets of parquet files. The initial records for the workload are loaded from one set of parquet