pragmatiq
Reference

CLI

The pragmatiq command-line interface — a thin wrapper over the Python API.

pragmatiq is an independent implementation inspired by the PRAGMA paper (arXiv 2604.08649) and is not affiliated with or endorsed by Revolut.

The CLI is intentionally thin: each command parses arguments and calls a function in pragmatiq/api.py, so the CLI, notebooks, and production callers all use the same library surface. Run any command with --help for its full options.

CommandWhat it does
pragmatiq synth generateGenerate a synthetic dataset (--config, --n-users, --seed, --n-workers).
pragmatiq synth calibrateFit generator priors to shareable aggregate statistics (no raw data).
pragmatiq validateCheck a dataset against the data contract.
pragmatiq tokenizeFit the tokenizer and write tokenized shards + index.
pragmatiq pretrainPretrain a model (--model-size, --config, --resume auto).
pragmatiq embedEmbed every user with a trained model → parquet.
pragmatiq probeProbe a model on a label table (--probe-model gbdt|logistic|lightgbm).
pragmatiq finetuneLoRA fine-tune a model's adapters + head on a label table.
pragmatiq upliftEvaluate communication-campaign uplift.
pragmatiq gnnRun the AML transfer-graph GraphSAGE ablation.
pragmatiq runs list / compareInspect and compare runs.
pragmatiq benchmarkThroughput benchmark for embedding.
pragmatiq quickstartEnd-to-end smoke: synth → tokenize → pretrain → probe.

A typical session

pragmatiq synth generate --out data/synth --config configs/data/synthetic.yaml
pragmatiq tokenize data/synth --out data/tok --n-workers 8
pragmatiq pretrain data/tok --name demo --model-size small --config configs/pretrain.yaml
pragmatiq embed data/tok --run runs/demo --out embeddings.parquet
pragmatiq probe data/tok --run runs/demo --label data/synth/labels/default_12m.parquet

Each command maps 1:1 onto a function documented in the Python API reference.

On this page