Cong123779/oracle-ai-vector-snapshot

---
language:
- vi
- en
tags:
- oracle-26ai
- vector-database
- semantic-search
- amazon-reviews
size_categories:
- 100K<n<1M
---

# 🤖 Oracle 26ai Vector Search Snapshot (290K Records)

This repository contains a high-performance database snapshot (Data Pump Export) of an optimized Oracle 26ai environment. It is pre-indexed and ready for production-grade semantic search experiments.

## 📊 Dataset Statistics
- **Total Records**: ~290,000 processed customer reviews.
- **Vector Dimensions**: 1024 (FLOAT32).
- **Oracle Version**: 26ai (23.26.1).
- **Indexed**: Yes (IVF-Flat Neighbor Partitions).

## 🚀 Quick Setup (New Machine)
To restore this environment on a fresh machine:

1. **Clone the Source Code**:
```bash
git clone https://github.com/congkx123789/He_quan_tri.git
cd He_quan_tri
```

2. **Pull the Snapshot**:
```bash
python python/hf_snapshot_sync.py pull
```

3. **Restore**:
```bash
bash scripts/auto_restore.sh
```

## 🏗️ Architecture
- **Ingestion**: Blackwell GPU Accelerated (RTX 5060 Ti).
- **Storage**: Optimized 12GB Clean Tablespace.
- **Search Engine**: Native Oracle AI Vector Search.

*Crafted by Antigravity for Enterprise-grade AI Demonstrations.*