jmp1987/racing-planet-a04-compatibilities-full

Racing Planet — A04 Compatibility Atoms (Full Database)

Phase 3.5 der Central Source of Truth Roadmap

---

Dataset

Property	Value
Atom Type	A04 Compatibility
Entries	2,668 Teil↔Fahrzeug-Links
Avg per Part	2.7 compatibility links
Brands Covered	9 (Yamaha, MBK, Aprilia, Piaggio, Peugeot, Kymco, Gilera, Derbi, Vespa, Honda, SYM, Simson)
Format	CSV + JSON

---

Schema

Field	Description
compat_id	UUID: compat_{part_id}_{vehicle_id}
part_id	→ A03 Part reference
sku	Racing Planet SKU
vehicle_id	→ A01 Vehicle reference
vehicle_brand	e.g. "Yamaha"
vehicle_model	e.g. "Aerox"
compatibility_type	direct_fit
confidence_score	0.95 (engine-family inferred)
verification_source	inferred_from_engine_family
displacement_match	e.g. "50ccm"
stroke_match	e.g. "2-stroke"
engine_family_inferred	e.g. "Minarelli horizontal"

---

How It Was Generated

1. 1,000 synthetic A03 parts generated with realistic brand/name/category combos
2. Each part matched to A01 vehicles by engine family + displacement + stroke
3. Confidence = 0.95 (engine-family-based inference)
4. Real production would validate with manufacturer specs + community data

---

Usage

from datasets import load_dataset

a01 = load_dataset("jmp1987/racing-planet-a01-vehicles", split="train")
a03 = load_dataset("jmp1987/racing-planet-a03-parts", split="train")
a04 = load_dataset("jmp1987/racing-planet-a04-compatibilities-full", split="train")

# Find parts for Yamaha Aerox
for c in a04:
    if c['vehicle_brand'] == 'Yamaha' and c['vehicle_model'] == 'Aerox':
        part = next(p for p in a03 if p['sku'] == c['sku'])
        print(f"{part['brand']} {part['name']} ({part['category']}) — {part['price_eur']}€")

---

Linked Datasets

Atom	Dataset
A01 Vehicle	jmp1987/racing-planet-a01-vehicles
A03 Part	jmp1987/racing-planet-a03-parts