Files
vector-search-demo/README.md
T
2026-05-19 12:05:20 +02:00

605 lines
23 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Vector Image Search — PostgreSQL/pgvector vs Oracle 26ai
A comparative demo that vectorizes JPEG photos using the CLIP neural network model
and stores the embeddings in two different databases: **PostgreSQL with pgvector**
and **Oracle AI Database 26ai**. Users search the photo collection by typing
plain-text keywords such as "trees" or "water" and receive results ranked by
semantic similarity.
Three backends are implemented, demonstrating two fundamental approaches to vector
embedding:
| Backend | Port | Embedding location | Model |
|---|---|---|---|
| PostgreSQL + pgvector | 8000 | Python (external) | sentence-transformers CLIP |
| Oracle 26ai (Python embedding) | 8001 | Python (external) | sentence-transformers CLIP |
| Oracle 26ai (in-database embedding) | 8002 | Inside Oracle SQL | Oracle native CLIP_TXT |
The key architectural difference: in the third backend, the text query is embedded
**inside a SQL statement** using Oracle's `VECTOR_EMBEDDING()` function — no Python
ML library is loaded or called at search time.
---
## Architecture overview
```
116 JPEG photos
┌───────────────────────────────┐
│ CLIP model (clip-ViT-B-32) │
│ sentence-transformers lib │
│ → 512-dimensional float vec │
└──────────────┬────────────────┘
┌──────────────┴──────────────┐
│ │
▼ ▼
┌──────────────────────┐ ┌──────────────────────┐ ┌───────────────────────┐
│ PostgreSQL 18 │ │ Oracle 26ai │ │ Oracle 26ai │
│ + pgvector 0.8.2 │ │ (version 23.26.1) │ │ (version 23.26.1) │
│ database: │ │ PDB: FREEPDB1 │ │ PDB: FREEPDB1 │
│ vectors_demo │ │ user: vectors_user │ │ schema: VECTOR │
│ HNSW index │ │ HNSW index │ │ HNSW not needed │
└────────┬─────────────┘ └──────────┬───────────┘ └──────────┬────────────┘
│ │ │
▼ ▼ │
Python CLIP encode Python CLIP encode Text stays in Oracle SQL
(search query) (search query) VECTOR_EMBEDDING(CLIP_TXT
USING :q AS data)
│ │ │
▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────────┐
│ FastAPI │ │ FastAPI │ │ FastAPI │
│ main.py │ │ main_oracle │ │ main_oracle_ │
│ port 8000 │ │ port 8001 │ │ indb.py │
└──────┬───────┘ └──────┬───────┘ │ port 8002 │
│ │ └────────┬─────────┘
▼ ▼ ▼
/ui/ (pgvector) /ui/ (Oracle 26ai) /ui/ (Oracle In-DB)
```
---
## Project structure
```
vector-search-demo/
├── start.sh # Start all three backends
├── stop.sh # Stop all three backends
├── photos/ # 116 JPEG photos (gitignored)
├── pgvector-demo/
│ ├── sql/
│ │ └── setup.sql # Create table and HNSW index
│ ├── backend/
│ │ ├── .env # PostgreSQL credentials, photo path
│ │ ├── db.py # PostgreSQL connection factory
│ │ ├── embedder.py # CLIP model wrapper
│ │ ├── index_images.py # One-time indexing script
│ │ └── main.py # FastAPI app (port 8000)
│ └── frontend/
│ └── index.html # Search UI (served at /ui/)
└── oravector-demo/
├── sql/
│ ├── setup_vectors_user.sql # Create vectors_user, table and HNSW index
│ └── setup_vector_schema.sql # Create VECTOR user, load ONNX models, FOTO_VEKTOR table
├── backend/
│ ├── .env # Oracle credentials, photo path
│ ├── db_oracle.py # Oracle connection factory
│ ├── embedder.py # CLIP model wrapper (identical to pgvector)
│ ├── index_images_oracle.py # One-time indexing script (Python embedding)
│ ├── main_oracle.py # FastAPI app — Python embedding (port 8001)
│ └── main_oracle_indb.py # FastAPI app — in-database embedding (port 8002)
└── frontend/
├── index.html # Search UI (Oracle 26ai, served at port 8001 /ui/)
└── indb/
└── index.html # Search UI (Oracle In-DB, served at port 8002 /ui/)
```
---
## System components
### PostgreSQL (Docker)
| Property | Value |
|---|---|
| Image | `pgvector/pgvector:pg18` |
| Version | PostgreSQL 18 |
| pgvector version | 0.8.2 |
| Host port | 5433 (mapped to 5432 inside container) |
| Database | `vectors_demo` |
| User | `dl` |
| Compose file | `~/docker/postgresql/docker-compose.yml` |
**Start PostgreSQL:**
```bash
cd ~/docker/postgresql && docker compose up -d
```
The `pgvector/pgvector:pg18` image includes pgvector pre-installed. See the
[Setup from scratch](#setup-from-scratch) section for first-time database setup.
### Oracle 26ai (Podman container)
| Property | Value |
|---|---|
| Product | Oracle AI Database 26ai Free |
| Version | 23.26.1.0.0 |
| Container name | `oracle.free` |
| Host port | 37611 (mapped to 1521 inside container) |
| Pluggable Database | FREEPDB1 |
| Schema users | `vectors_user`, `VECTOR` |
**Oracle vector memory** — the HNSW index is held entirely in the SGA's Vector
Memory Area. This is already configured:
```sql
ALTER SYSTEM SET vector_memory_size = 512M SCOPE=SPFILE;
```
After restart, the SGA confirms: `Vector Memory Area: 536870912 bytes (512 MB)`.
### Python packages
| Package | Version | Used by | Purpose |
|---|---|---|---|
| `sentence-transformers` | 5.3.0 | both | CLIP model loading and inference |
| `torch` | 2.11.0 | both | Neural network runtime for CLIP |
| `Pillow` | 10.2.0 | both | JPEG loading and colour conversion |
| `fastapi` | 0.135.2 | both | REST API framework |
| `uvicorn` | 0.42.0 | both | ASGI server |
| `python-dotenv` | 1.0.1 | both | `.env` file support |
| `psycopg2-binary` | 2.9.11 | pgvector only | PostgreSQL driver |
| `oracledb` | 3.4.2 | Oracle only | Oracle driver (thin mode, no client libs needed) |
**Install packages:**
```bash
pip3 install -r pgvector-demo/backend/requirements.txt --break-system-packages
pip3 install -r oravector-demo/backend/requirements.txt --break-system-packages
```
---
## Vectorization
### Model: CLIP (clip-ViT-B-32)
CLIP (Contrastive LanguageImage Pretraining) is a neural network model developed
by OpenAI. It was trained on hundreds of millions of imagetext pairs and maps both
images and text into the **same 512-dimensional vector space**. This enables
searching images by plain-text query without any manual labelling or tagging.
| Property | Value |
|---|---|
| Architecture | Vision Transformer ViT-B/32 |
| Output dimension | 512 floats |
| Similarity metric | Cosine similarity |
| Weights source | Hugging Face Hub: `sentence-transformers/clip-ViT-B-32` |
| Downloaded to | `~/.cache/huggingface/hub/` on first run |
**Why cosine similarity?** CLIP vectors have varying magnitudes. Cosine similarity
normalises for magnitude and measures only the direction — the angle between two
vectors — which reliably captures semantic relatedness regardless of vector scale.
The `embedder.py` module is identical in both projects. It lazily loads the model
on first call and exposes two functions:
| Function | Input | Output |
|---|---|---|
| `embed_image(path)` | Filesystem path to a JPEG | `list[float]` — 512 values |
| `embed_text(text)` | Plain-text query string | `list[float]` — 512 values |
At search time, the text query is embedded into the same vector space as the photos.
The database then finds the photos whose vectors point in the most similar direction.
---
## Database schemas
### PostgreSQL + pgvector
```sql
-- database: vectors_demo (PostgreSQL 18)
CREATE EXTENSION vector; -- pgvector 0.8.2
CREATE TABLE images (
id SERIAL PRIMARY KEY,
filename TEXT NOT NULL UNIQUE,
filepath TEXT NOT NULL,
embedding vector(512) -- pgvector type, 512 dimensions
);
CREATE INDEX images_embedding_idx
ON images USING hnsw (embedding vector_cosine_ops);
```
### Oracle 26ai
```sql
-- PDB: FREEPDB1, user: vectors_user
CREATE TABLE images (
id NUMBER GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
filename VARCHAR2(255) NOT NULL UNIQUE,
filepath VARCHAR2(1000) NOT NULL,
embedding VECTOR(512, FLOAT32) -- native Oracle type, typed at definition
);
CREATE VECTOR INDEX images_embedding_idx
ON images(embedding)
ORGANIZATION INMEMORY NEIGHBOR GRAPH -- HNSW (in-memory)
WITH DISTANCE COSINE
WITH TARGET ACCURACY 95
PARAMETERS (type HNSW, neighbors 32, efconstruction 200);
```
**Key schema differences:**
| Aspect | PostgreSQL/pgvector | Oracle 26ai |
|---|---|---|
| Extension needed | `CREATE EXTENSION vector` | Built-in, no extension |
| Vector column | `vector(512)` — dimension only | `VECTOR(512, FLOAT32)` — dimension + element type |
| Primary key | `SERIAL` (auto-increment) | `NUMBER GENERATED ALWAYS AS IDENTITY` |
| Text columns | `TEXT` (unlimited) | `VARCHAR2(n)` (length required) |
| HNSW syntax | `USING hnsw (col vector_cosine_ops)` | `ORGANIZATION INMEMORY NEIGHBOR GRAPH` |
| IVF syntax | `USING ivfflat (col vector_cosine_ops)` | `ORGANIZATION NEIGHBOR PARTITIONS` |
| Accuracy target | Implicit (set via index params) | `WITH TARGET ACCURACY 95` (explicit %) |
| Memory prereq | None | `vector_memory_size > 0` in SGA |
---
## Backend modules
### Connection factories
**`db.py` (PostgreSQL):**
Reads `DB_HOST`, `DB_PORT`, `DB_NAME`, `DB_USER`, `DB_PASSWORD` from `.env` and
returns a `psycopg2` connection.
**`db_oracle.py` (Oracle):**
Reads `ORA_HOST`, `ORA_PORT`, `ORA_SERVICE`, `ORA_USER`, `ORA_PASSWORD` (and
`ORA_USER_INDB`, `ORA_PASSWORD_INDB` for the in-DB backend) from `.env` and
returns an `oracledb` connection. The DSN is assembled as `host:port/service`.
Runs in **thin mode** — no Oracle Instant Client installation is required on the host.
---
### Indexing scripts
Both scripts are idempotent: they check for existing rows and skip already-indexed
photos. Each photo is committed individually so a crash does not lose prior work.
| | `index_images.py` | `index_images_oracle.py` |
|---|---|---|
| Run command | `python3 index_images.py` | `python3 index_images_oracle.py` |
| Vector bind | Python `list` passed directly | `array.array("f", embedding)` required |
| Bind style | `%s` placeholders (psycopg2) | `:1`, `:2`, `:3` positional (oracledb) |
| Runtime (116 photos, CPU) | ~26 seconds | ~16 seconds |
**Why `array.array` for Oracle?**
The `python-oracledb` driver does not accept a plain Python list for a `VECTOR`
column. The data must be a Python `array.array` with typecode `"f"` (32-bit float),
matching the `FLOAT32` declaration in the Oracle column type.
---
### FastAPI applications
All three apps expose identical endpoints:
| Endpoint | Description |
|---|---|
| `GET /search?q=<text>&limit=<n>` | Embed query, run nearest-neighbour search, return ranked results |
| `GET /stats` | Return count of indexed photos |
| `GET /photos/<filename>` | Serve original JPEG from the photos directory |
| `GET /ui/` | Serve the search frontend (HTML) |
**Search query comparison:**
PostgreSQL (`main.py`, port 8000):
```sql
SELECT filename, 1 - (embedding <=> $1::vector) AS score
FROM images
ORDER BY embedding <=> $1::vector
LIMIT $2
```
Oracle 26ai (`main_oracle.py`, port 8001):
```sql
SELECT filename,
1 - VECTOR_DISTANCE(embedding, :vec, COSINE) AS score
FROM images
ORDER BY VECTOR_DISTANCE(embedding, :vec, COSINE)
FETCH FIRST :lim ROWS ONLY
```
**Key query differences:**
| Aspect | PostgreSQL/pgvector | Oracle 26ai |
|---|---|---|
| Distance operator | `<=>` (cosine distance operator) | `VECTOR_DISTANCE(col, val, COSINE)` |
| Cast required | `$1::vector` — explicit cast | No cast, column type is enforced |
| Top-N clause | `LIMIT n` | `FETCH FIRST n ROWS ONLY` |
| Bind style | `$1`, `$2` positional (psycopg2) | `:name` named binds (dict) |
| Repeated param | `$1` can appear multiple times | Same `:name` can appear multiple times |
| Score formula | `1 - (embedding <=> val)` | `1 - VECTOR_DISTANCE(...)` |
In both cases `1 distance` converts cosine distance (0 = identical) into a
similarity score (1.0 = identical), displayed as a percentage in the frontend.
---
## Frontend
Three single-file HTML frontends, each served by its own backend at `/ui/`:
| | pgvector | Oracle 26ai | Oracle In-DB |
|---|---|---|---|
| URL | `http://localhost:8000/ui/` | `http://localhost:8001/ui/` | `http://localhost:8002/ui/` |
| Badge colour | Blue | Red | Purple |
| File | `pgvector-demo/frontend/index.html` | `oravector-demo/frontend/index.html` | `oravector-demo/frontend/indb/index.html` |
Features: search box, Enter-key support, suggestion chips (trees, water, people,
buildings, sky, street, night, cars), result grid with thumbnails and similarity
scores in percent.
---
## Configuration (.env files)
Each backend reads its credentials and paths from a `.env` file in its `backend/`
directory. These files are gitignored — copy the `.env.example` template and fill
in the values before running.
### pgvector-demo/backend/.env
```bash
cp pgvector-demo/backend/.env.example pgvector-demo/backend/.env
```
| Variable | Description | Example |
|---|---|---|
| `DB_HOST` | PostgreSQL host | `localhost` |
| `DB_PORT` | PostgreSQL host port | `5433` |
| `DB_NAME` | Database name | `vectors_demo` |
| `DB_USER` | Database user | `dl` |
| `DB_PASSWORD` | Database password | — |
| `PHOTOS_DIR` | Absolute path to the photos folder | `/home/user/photos` |
### oravector-demo/backend/.env
```bash
cp oravector-demo/backend/.env.example oravector-demo/backend/.env
```
| Variable | Description | Example |
|---|---|---|
| `ORA_HOST` | Oracle host | `localhost` |
| `ORA_PORT` | Oracle host port | `37611` |
| `ORA_SERVICE` | Oracle service name (PDB) | `FREEPDB1` |
| `ORA_USER` | User for Python-embedding backend | `vectors_user` |
| `ORA_PASSWORD` | Password for `ORA_USER` | — |
| `ORA_USER_INDB` | User for in-database embedding backend | `vector` |
| `ORA_PASSWORD_INDB` | Password for `ORA_USER_INDB` | — |
| `PHOTOS_DIR` | Absolute path to the photos folder | `/home/user/photos` |
---
## Setup from scratch
### 0. Python dependencies
Install all required packages for both backends:
```bash
pip3 install -r pgvector-demo/backend/requirements.txt --break-system-packages
pip3 install -r oravector-demo/backend/requirements.txt --break-system-packages
```
### 1. PostgreSQL
**Start the container:**
```bash
cd ~/docker/postgresql && docker compose up -d
```
**Create the database:**
```bash
docker exec postgresql-database-1 psql -U dl -d pgdl -c "CREATE DATABASE vectors_demo;"
```
**Run the setup script** (creates the pgvector extension, `images` table, and HNSW index):
```bash
docker exec -i postgresql-database-1 psql -U dl -d vectors_demo -f - \
< pgvector-demo/sql/setup.sql
```
**Copy photos and index them:**
```bash
cd pgvector-demo/backend && python3 index_images.py
```
---
### 2. Oracle 26ai — Python embedding backend
**Configure vector memory** (once, requires Oracle restart):
```bash
podman exec oracle.free bash -c "sqlplus -s / as sysdba <<'EOF'
ALTER SYSTEM SET vector_memory_size = 512M SCOPE=SPFILE;
SHUTDOWN ABORT;
STARTUP;
EXIT;
EOF"
```
**Run the setup script** (creates `vectors_user`, the `images` table, and HNSW index):
Copy the script into the container and run it as SYSDBA:
```bash
podman cp oravector-demo/sql/setup_vectors_user.sql oracle.free:/tmp/
podman exec oracle.free bash -c "sqlplus -s / as sysdba @/tmp/setup_vectors_user.sql"
```
**Index the photos:**
```bash
cd oravector-demo/backend && python3 index_images_oracle.py
```
---
### 3. Oracle 26ai — in-database embedding backend
This backend requires CLIP ONNX models loaded into the Oracle database. The setup
is more involved and is intended to be done once by an administrator.
**Prerequisites:**
- CLIP ONNX model files (`clip_txt.onnx`, `clip_img.onnx`) present in the Oracle
VEC_DUMP directory inside the container (typically `/opt/oracle/dbs/vec_dump/`)
- The `clip_txt.onnx` model must use **CLS-token pooling** (position 0), not the
standard EOS-token pooling — Oracle's ONNX validator rejects models that use
`ArgMax` on `input_ids`. See the [Oracle in-database embedding](#oracle-in-database-embedding)
section for details.
**Run the setup script** (creates `VECTOR` user, loads ONNX models, creates `FOTO_VEKTOR` table):
```bash
podman cp oravector-demo/sql/setup_vector_schema.sql oracle.free:/tmp/
podman exec oracle.free bash -c "sqlplus -s / as sysdba @/tmp/setup_vector_schema.sql"
```
**Populate `FOTO_VEKTOR`** with images and their vectors (run as VECTOR user in SQL):
```sql
-- Example: insert one photo with its CLIP_IMG embedding
INSERT INTO vector.foto_vektor (filename, foto, foto_vek)
VALUES (
'photo.jpg',
TO_BLOB(BFILENAME('VEC_DUMP', 'photo.jpg')),
VECTOR_EMBEDDING(CLIP_IMG USING TO_BLOB(BFILENAME('VEC_DUMP', 'photo.jpg')) AS data)
);
COMMIT;
```
---
## Running the applications
### Start all backends
```bash
./start.sh
```
This starts all three backends concurrently. Press Ctrl+C to stop all.
### Stop all backends
```bash
./stop.sh
```
### Start backends individually
```bash
# PostgreSQL backend
cd pgvector-demo/backend && uvicorn main:app --host 0.0.0.0 --port 8000
# Oracle backend — Python embedding
cd oravector-demo/backend && uvicorn main_oracle:app --host 0.0.0.0 --port 8001
# Oracle backend — in-database embedding
cd oravector-demo/backend && uvicorn main_oracle_indb:app --host 0.0.0.0 --port 8002
```
### Re-index after adding photos
```bash
# PostgreSQL
cd pgvector-demo/backend && python3 index_images.py
# Oracle (Python embedding)
cd oravector-demo/backend && python3 index_images_oracle.py
# Oracle in-database: re-indexing is done in SQL directly
# (the VECTOR schema's FOTO_VEKTOR table is managed by Oracle)
```
---
## Oracle in-database embedding
The `VECTOR` schema, its ONNX models, and the `FOTO_VEKTOR` table were manually
set up by the administrator — they are **not** part of a standard Oracle 26ai
installation. The setup involved:
1. Creating a `VECTOR` database user
2. Exporting CLIP (ViT-B/32) to ONNX format and loading the models via
`DBMS_VECTOR.LOAD_ONNX_MODEL`
3. Creating and populating the `FOTO_VEKTOR` table with images and their vectors
The resulting models and table are:
| Object | Type | Input | Output | Purpose |
|---|---|---|---|---|
| `VECTOR.CLIP_TXT` | ONNX model | `VARCHAR2` text | `VECTOR(512)` | Embed text queries |
| `VECTOR.CLIP_IMG` | ONNX model | `BLOB` image | `VECTOR(512)` | Embed image data |
| `VECTOR.FOTO_VEKTOR` | Table | — | — | Stores filenames, image BLOBs, and vectors |
These are called with the `VECTOR_EMBEDDING()` SQL function. The table
`VECTOR.FOTO_VEKTOR` stores images as BLOBs alongside their CLIP_IMG-computed
embeddings.
**The complete in-database search query:**
```sql
SELECT filename,
1 - VECTOR_DISTANCE(
foto_vek,
VECTOR_EMBEDDING(CLIP_TXT USING :q AS data),
COSINE
) AS score
FROM VECTOR.FOTO_VEKTOR
ORDER BY VECTOR_DISTANCE(
foto_vek,
VECTOR_EMBEDDING(CLIP_TXT USING :q AS data),
COSINE
)
FETCH FIRST 12 ROWS ONLY
```
The Python FastAPI backend (`main_oracle_indb.py`) passes only the raw text string
to Oracle via a bind variable `:q`. Oracle tokenizes the text, runs the CLIP_TXT
ONNX model internally, produces the 512-dim vector, and performs the similarity
search — all within one SQL statement. No Python ML library is involved at
query time.
**Why Oracle can ship CLIP as an in-database ONNX model:**
Oracle's `DBMS_VECTOR.LOAD_ONNX_MODEL` requires the model's ONNX graph to use
`input_ids` in a single `Gather` node (embedding lookup only). CLIP's standard
export uses `input_ids` additionally in `ArgMax` for EOS-token pooling, which
Oracle's validator rejects. The manually loaded CLIP_TXT model in the `VECTOR`
schema uses CLS-token pooling (position 0) instead, which produces a simpler
graph that Oracle accepts. The cosine similarity between EOS-pooling and
CLS-pooling variants is ~0.70.
---
## Performance comparison
Measured on this installation (CPU only, no GPU):
| Metric | PostgreSQL + pgvector | Oracle 26ai (Python embed) | Oracle 26ai (in-DB embed) |
|---|---|---|---|
| Photos indexed | 116 | 116 | 116 (manually indexed) |
| Indexing time | ~26 seconds | ~16 seconds | 0 (indexed separately by admin) |
| Index type | HNSW (on disk) | HNSW (in-memory) | Full table scan (116 rows) |
| Memory required | None | 512 MB SGA | 512 MB SGA |
| Python CLIP at query time | Yes | Yes | **No** |
| Embedding location | Python process | Python process | Inside Oracle SQL |
| `VECTOR_EMBEDDING()` used | No | No | **Yes** |
Note: indexing time for backends 1 and 2 is dominated by CLIP inference (CPU),
not database write speed. The in-database backend uses the manually loaded CLIP
models in the `VECTOR` schema; their indexing time is not measured here as it
was performed separately by the administrator.