{"id":47706110,"url":"https://github.com/oceanbase/ob-mldr-test","last_synced_at":"2026-04-02T17:58:30.028Z","repository":{"id":326352521,"uuid":"1094674125","full_name":"oceanbase/ob-mldr-test","owner":"oceanbase","description":null,"archived":false,"fork":false,"pushed_at":"2026-01-23T12:02:27.000Z","size":126,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-01-24T03:55:10.908Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/oceanbase.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":"NOTICE","maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-11-12T02:53:24.000Z","updated_at":"2026-01-23T12:02:31.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/oceanbase/ob-mldr-test","commit_stats":null,"previous_names":["oceanbase/ob-mldr-test"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/oceanbase/ob-mldr-test","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oceanbase%2Fob-mldr-test","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oceanbase%2Fob-mldr-test/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oceanbase%2Fob-mldr-test/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oceanbase%2Fob-mldr-test/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/oceanbase","download_url":"https://codeload.github.com/oceanbase/ob-mldr-test/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oceanbase%2Fob-mldr-test/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31312744,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-02T12:59:32.332Z","status":"ssl_error","status_checked_at":"2026-04-02T12:54:48.875Z","response_time":89,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-04-02T17:58:29.346Z","updated_at":"2026-04-02T17:58:30.021Z","avatar_url":"https://github.com/oceanbase.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"[English](README.md) | [中文](README_CN.md) \n\n# MLDR dataset testing framework\n\nA modified Multi-Language Document Retrieval (MLDR) testing framework based on the [Infinity MLDR Benchmark](https://github.com/infiniflow/infinity/tree/main/python/benchmark/mldr_benchmark).\n\n[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE)\n\n---\n\n## Project Introduction\n\nThis project is a comprehensive MLDR (Multi-Lingual Document Retrieval) dataset testing framework designed to evaluate and test the performance of multi-language document retrieval systems. It supports various query types, including BM25, Dense Vector, and hybrid retrieval.\n\n## Functional Features\n\n### 🗄️ Supported database backends\n- **OceanBase**: supports full-text search, vector search and hybrid search\n- **seekdb**: supports full-text search, vector search and hybrid search\n\n### 🔍 Supported query types\n\n| Query Type | Description | Supported Backends |\n|---------|------|---------|\n| `bm25` | BM25 Full-Text Search | OceanBase, seekdb |\n| `dense` | Dense Vector Search | OceanBase, seekdb |\n| `hybrid_dense_bm25` | Dense+BM25 Hybrid Search | OceanBase, seekdb |\n\n### 📊 Evaluation Metrics \n- **Recall@10**: Recall rate of the top 10 results\n- **NDCG@10**: Normalized Discounted Cumulative Gain\n- **Average Query Time**: Average response time per query\n\n## Environmental requirements\n\n### System Requirements\n- **Python**: 3.11 or above\n- **Java**: JDK 11 or above (for pyserini)\n\n### Dependent Services\n- **OceanBase Database**: Versions supporting vector retrieval and full-text search (4.4.1 and above)\n- **seekdb Database**: Supports vector retrieval and full-text search\n\n---\n\n## Installation Guide\n\n### 1. Install Java environment\n\n**Linux (Alibaba Cloud Linux / CentOS):**\n```bash\nsudo dnf install java-11-openjdk-devel -y\nexport JAVA_HOME=/usr/lib/jvm/java-11-openjdk\nexport JVM_PATH=$JAVA_HOME/lib/server/libjvm.so\n```\n\n### 2. Create a Python virtual environment\n\n```bash\n# Download and install Conda \nmkdir -p ~/miniconda3\nwget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh\nbash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3\nrm ~/miniconda3/miniconda.sh\n\n# Reopen the terminal and initialize the Conda environment \nsource ~/miniconda3/bin/activate\nconda init --all\n\n# Create and initialize the Python environment required for mdlr\nconda create -n test python=3.11\nconda activate test\n```\n\n### 3. Install Python dependencies\n\n```bash\n# Upgrade pip\npip install --upgrade pip\n\n# Install dependencies\npip install -r requirements.txt\n```\n\n### 4. Prepare the database\n\n**Start OceanBase or seekdb database:**\n```bash\n# Ensure the database is started and the connection information is configured\n# Configure the connection parameters for the corresponding database in config.yaml\n```\n\n---\n\n## Quick Start\n\n\u003e **⚠️ Important Note: Please configure the database information before use**\n\u003e \n\u003e Before running the test, please complete the following configuration steps:\n\u003e \n\u003e ```bash\n\u003e # 1. Copy the example configuration file\n\u003e cp config.yaml.example config.yaml\n\u003e \n\u003e # 2. Edit the configuration file and replace the database information with your own configuration\n\u003e vim config.yaml  \n\u003e ```\n\u003e \n\u003e The main configuration items that need to be modified are:\n\u003e - `oceanbase.host`: Database host address\n\u003e - `oceanbase.port`: Database port number\n\u003e - `oceanbase.user`: Database username\n\u003e - `oceanbase.password`: Database password\n\u003e - `oceanbase.database`: Database name\n\u003e - `embedding.vector_download_url`: Vector file download URL (if vector retrieval is used)\n\n### Run a complete test with one click\n\n```bash\n# English mixed retrieval test\npython mldr_test_runner.py --lang en --query-type hybrid_dense_bm25\n\n# Skip data insertion (data already exists)\npython mldr_test_runner.py --lang en --query-type hybrid_dense_bm25 --skip-insert\n```\n\n### Test process description\n\nThe testing framework will automatically execute the following steps:\n\n1. **Data insertion**: Insert 200,000 test data entries and create indexes\n2. **Warm-up test** (optional): Perform a search warm-up\n3. **Formal testing**: Perform a specified number of searches and evaluations\n4. **Result Output**: Display the average Recall@10, NDCG@10, and average latency\n\n---\n\n## Detailed Usage\n\n### Command line parameters\n\n```bash\npython mldr_test_runner.py [OPTIONS]\n```\n\n**Basic parameters:**\n\n| Parameter | Type | Default | Description |\n|------|------|--------|------|\n| `--lang` | str | `en` | Test language (e.g. `en`, `zh`, etc.) |\n| `--backend` | str | `oceanbase` | Database backend (`oceanbase` or `seekdb`) |\n| `--query-type` | str | `bm25` | Query type, optional values: `hybrid_dense_bm25` (hybrid retrieval), `dense` (vector retrieval), `bm25` (full-text retrieval) |\n| `--skip-insert` | flag | `False` | Skip the data insertion step (used when the data already exists)  |\n| `--result-dir` | str | `/tmp/` | directory for saving results |\n| `--config` | str | `None` | Path to the configuration file (in YAML format), defaults to `config.yaml` in the current directory |\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foceanbase%2Fob-mldr-test","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Foceanbase%2Fob-mldr-test","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foceanbase%2Fob-mldr-test/lists"}