dataxid 0.1.0

The Synthetic Data API — privacy-preserving synthetic data generation

Dataxid Python SDK

Privacy-preserving synthetic data generation, built on a privacy-by-architecture principle. Your raw data never leaves your machine — only abstract embeddings are shared with the API.

Installation

pip install dataxid

Quick Start

import dataxid
import pandas as pd

dataxid.api_key = "dx_..."

df = pd.read_csv("data.csv")
synthetic = dataxid.synthesize(data=df, n_samples=1000)

Full Control

import dataxid
import pandas as pd

dataxid.api_key = "dx_..."

df = pd.read_csv("data.csv")

model = dataxid.Model.create(data=df)
synthetic = model.generate(n_samples=1000)
model.delete()

Error Handling

import dataxid

try:
    synthetic = dataxid.synthesize(data=df)
except dataxid.AuthenticationError:
    print("Invalid API key")
except dataxid.QuotaExceededError as e:
    print(f"Quota exceeded. Upgrade: {e.upgrade_url}")
except dataxid.RateLimitError as e:
    print(f"Rate limited. Retry after: {e.retry_after}s")
except dataxid.DataxidError as e:
    print(f"Error: {e}")

How It Works

Dataxid is built on a privacy-by-architecture principle. Data encoding and decoding happen entirely on your machine; only abstract embeddings are shared with the API for model training. Raw data never leaves your environment.

Configuration

Parameter	Default	Description
embedding_dim	64	Embedding size (larger = more expressive)
model_size	"medium"	Model capacity: "small", "medium", "large"
max_epochs	100	Maximum training epochs
batch_size	256	Training batch size
privacy_enabled	False	Add noise to embeddings for privacy
privacy_noise	0.1	Noise scale (Gaussian std)

model = dataxid.Model.create(
    data=df,
    config=dataxid.ModelConfig(
        embedding_dim=128,
        model_size="large",
        max_epochs=50,
    ),
)

Plain dict also works for quick experiments:

model = dataxid.Model.create(
    data=df,
    config={"embedding_dim": 128, "max_epochs": 50},
)

Links

• Documentation
• API Reference
• GitHub
• Examples