SDK Documentation | Platform Documentation | Usage Examples
The official SDK of MOSTLY AI, a Python toolkit for high-fidelity, privacy-safe Synthetic Data.
- Client mode connects to a remote MOSTLY AI platform for training & generating synthetic data there.
- Local mode trains and generates synthetic data locally on your own compute resources.
- Generators, that were trained locally, can be easily imported to a platform for further sharing.
The SDK allows you to programmatically create, browse and manage 3 key resources:
- Generators - Train a synthetic data generator on your existing tabular or language data assets
- Synthetic Datasets - Use a generator to create any number of synthetic samples to your needs
- Connectors - Connect to any data source within your organization, for reading and writing data
Intent | Primitive | Documentation |
---|---|---|
Train a Generator on tabular or language data | g = mostly.train(config) |
see mostly.train |
Generate any number of synthetic data records | sd = mostly.generate(g, config) |
see mostly.generate |
Live probe the generator on demand | df = mostly.probe(g, config) |
see mostly.probe |
Connect to any data source within your org | c = mostly.connect(config) |
see mostly.connect |
Client mode only
pip install -U mostlyai
Client + Local mode
pip install -U 'mostlyai[local]' # for CPU
#pip install -U 'mostlyai[local-gpu]' # for GPU
NOTE: installing mostlyai[local]
on Linux requires --extra-index-url https://download.pytorch.org/whl/cpu
to be specified.
Optional Connectors
Add any of the following extras for further data connectors support: databricks
, googlebigquery
, hive
, mssql
, mysql
, oracle
, postgres
, snowflake
.
E.g.
pip install -U 'mostlyai[local, databricks, snowflake]'
For client mode, initialize with base_url
and api_key
obtained from your account settings page. For local mode, initialize the client simply with local=True
.
import pandas as pd
from mostlyai.sdk import MostlyAI
# load original data
repo_url = 'https://github.com/mostly-ai/public-demo-data'
df_original = pd.read_csv(f'{repo_url}/raw/dev/census/census.csv.gz')
# initialize the SDK in local or client mode
mostly = MostlyAI(local=True)
# mostly = MostlyAI(base_url='https://app.mostly.ai', api_key='YOUR_API_KEY')
# train a synthetic data generator
g = mostly.train(config={
'name': 'US Census Income', # name of the generator
'tables': [{ # provide list of table(s)
'name': 'census', # name of the table
'data': df_original, # the original data as pd.DataFrame
'tabular_model_configuration': { # tabular model configuration (optional)
'max_training_time': 1, # - limit training time (in minutes)
# model, max_epochs,,.. # further model configurations (optional)
'differential_privacy': { # differential privacy configuration (optional)
'max_epsilon': 5.0, # - max epsilon value, used as stopping criterion
'delta': 1e-5, # - delta value
}
},
# columns, keys, compute,.. # further table configurations (optional)
}]
},
start=True, # start training immediately (default: True)
wait=True, # wait for completion (default: True)
)
Once the generator has been trained, you can use it to generate synthetic data samples. Either via probing:
# probe for some representative synthetic samples
df_samples = mostly.probe(g, size=100)
df_samples
or by creating a synthetic dataset entity for larger data volumes:
# generate a large representative synthetic dataset
sd = mostly.generate(g, size=100_000)
df_synthetic = sd.data()
df_synthetic
or by conditionally probing / generating synthetic data:
# create 100 seed records of 24y old Mexicans
df_seed = pd.DataFrame({
'age': [24] * 100,
'native_country': ['Mexico'] * 100,
})
# conditionally probe, based on provided seed
df_samples = mostly.probe(g, seed=df_seed)
df_samples
- Broad Data Support
- Mixed-type data (categorical, numerical, geospatial, text, etc.)
- Single-table, multi-table, and time-series
- Multiple Model Types
- TabularARGN for SOTA tabular performance
- Fine-tune HuggingFace-based language models
- Efficient LSTM for text synthesis from scratch
- Advanced Training Options
- GPU/CPU support
- Differential Privacy
- Progress Monitoring
- Automated Quality Assurance
- Quality metrics for fidelity and privacy
- In-depth HTML reports for visual analysis
- Flexible Sampling
- Up-sample to any data volumes
- Conditional generation by any columns
- Re-balance underrepresented segments
- Context-aware data imputation
- Statistical fairness controls
- Rule-adherence via temperature
- Seamless Integration
- Connect to external data sources (DBs, cloud storages)
- Fully permissive open-source license
Please consider citing our project if you find it useful:
@software{mostlyai,
author = {{MOSTLY AI}},
title = {{MOSTLY AI SDK}},
url = {https://github.com/mostly-ai/mostlyai},
year = {2025}
}