KiCad Schematic Converter

Python utilities for converting KiCad schematics (.kicad_sch) to and from JSON/YAML formats. Built for the open-schematics dataset on Hugging Face.

About open-schematics Dataset

The open-schematics dataset is a collection of open-source KiCad schematics designed for training and fine-tuning AI models on electronic circuit design. Each entry contains:

Field	Description
schematic	Raw KiCad schematic file content (.kicad_sch)
image	Rendered schematic image
components_used	List of components in the design
json	Simplified JSON representation
yaml	Simplified YAML representation
name	Project name
description	Project description
type	File type

Requirements

pip install kiutils pyyaml datasets

Scripts Overview

Script	Direction	Description
kicad_to_json.py	KiCad → JSON	Convert schematic to simplified JSON
kicad_to_yaml.py	KiCad → YAML	Convert schematic to simplified YAML
json_to_kicad.py	JSON → KiCad	Reconstruct schematic from JSON
yaml_to_kicad.py	YAML → KiCad	Reconstruct schematic from YAML

Usage

Working with the Dataset

Load the dataset from Hugging Face:

from datasets import load_dataset

dataset = load_dataset("bshada/open-schematics")

# Access a schematic entry
entry = dataset['train'][0]
print(entry['name'])           # Project name
print(entry['description'])    # Project description
print(entry['components_used']) # List of components
print(entry['json'])           # Simplified JSON format
print(entry['yaml'])           # Simplified YAML format

KiCad → JSON/YAML

Convert existing KiCad schematics to the simplified format used in the dataset:

# Convert to JSON
python kicad_to_json.py my_project.kicad_sch

# Convert to YAML
python kicad_to_yaml.py my_project.kicad_sch

# Specify output path
python kicad_to_json.py my_project.kicad_sch -o output.json
python kicad_to_yaml.py my_project.kicad_sch -o output.yaml

JSON/YAML → KiCad

Reconstruct KiCad schematics from the simplified format. This is useful for:

• Generating schematics from AI model outputs
• Converting dataset entries back to editable KiCad files

# From JSON (uses kicad-symbols folder by default)
python json_to_kicad.py schematic.json

# From YAML  
python yaml_to_kicad.py schematic.yaml

# Specify custom symbols path
python json_to_kicad.py schematic.json -s /path/to/symbols

# Specify output path
python json_to_kicad.py schematic.json -o output.kicad_sch

The -s/--symbols flag defaults to kicad-symbols folder in the project directory. You can override it to point to a different location if needed.

Example: Dataset Entry to KiCad File

from datasets import load_dataset
import json

# Load dataset
dataset = load_dataset("bshada/open-schematics")
entry = dataset['train'][0]

# Save JSON to file
with open('schematic.json', 'w') as f:
    f.write(entry['json'])

# Then convert to KiCad format
# python json_to_kicad.py schematic.json -o project.kicad_sch

Data Format

The simplified JSON/YAML format captures essential schematic elements:

libSymbols:
  - "Connector:USB_C_Receptacle"
  - "Device:R"
  - "Device:C"

schematicSymbols:
  - libraryNickname: "Device"
    entryName: "R"
    position: {x: 100, y: 50, angle: 0}
    properties:
      - {key: "Reference", value: "R1", position: {...}}
      - {key: "Value", value: "10k", position: {...}}

junctions:
  - position: {x: 100, y: 100, angle: 0}

graphicalItems:
  - type: "wire"
    points:
      - {x: 100, y: 50}
      - {x: 100, y: 100}

globalLabels:
  - text: "VCC"
    shape: "input"
    position: {x: 50, y: 25, angle: 0}

License

MIT