---
title: Running AutoScientist | Adaption
description: Create, monitor, and handle an AutoScientist training run with the Python SDK.
---

## Create a run

The SDK reads `ADAPTION_API_KEY` from the environment. Call [`autoscientist.create`](/api/python/resources/autoscientist/methods/create/index.md) with the prepared dataset:

```
from adaption import Adaption


client = Adaption()


dataset_id = "dataset_abc123"
run = client.autoscientist.create(dataset_id=dataset_id)
print(run.id, run.status)  # pending
```

| Parameter                   | Required | Notes                                                                                                                                                                                  |
| --------------------------- | -------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `dataset_id`                | Yes      | An adapted dataset ID, or a [raw dataset](/autoscientist/run-on-non-adapted-data/index.md).                                                                                            |
| `max_iterations`            | No       | From 1 through 10.                                                                                                                                                                     |
| `target_win_rate`           | No       | Greater than 0 and at most 1. Stops the loop early once reached.                                                                                                                       |
| `model`                     | No       | An ID returned by [`client.autoscientist.list_models()`](/api/python/resources/autoscientist/methods/list_models/index.md). If omitted, AutoScientist selects a model for the dataset. |
| `training_type`             | No       | `lora` by default, or `full` when supported by the model.                                                                                                                              |
| `hyperparams`               | No       | Overrides to the [platform recipe](/autoscientist/recommended-hyperparameters/index.md); unset keys retain platform defaults.                                                          |
| `augmentation_domain_rows`  | No       | [Domain-targeted rows](/autoscientist/data-augmentation/index.md) added before training.                                                                                               |
| `augmentation_general_rows` | No       | General-purpose rows added before training.                                                                                                                                            |
| `idempotency_key`           | No       | Repeated requests with the same key return the existing run.                                                                                                                           |
| `column_mapping`            | No       | Maps columns in the adapted dataset schema. Omit it to let the platform infer the mapping.                                                                                             |

Training is supervised (`sft`); `create` has no `method` parameter.

`column_mapping` is validated against the **adapted dataset’s schema**, not the uploaded file’s original headers. Adaptation can add columns such as `enhanced_prompt`. A mapping to a missing column fails with `Selected column '...' for prompt is not in this dataset`.

## Wait for completion

[Run status](/api/python/resources/autoscientist/methods/get/index.md) is `pending`, `running`, `succeeded`, `failed`, or `cancelled`.

```
from adaption import TrainingTimeout


try:
    run = client.autoscientist.wait_for_completion(run.id)
except TrainingTimeout as exc:
    print(exc.resource_id, exc.last_status)
    raise


if run.status == "failed":
    raise RuntimeError(f"failed: {run.error}")
if run.status != "succeeded":
    raise RuntimeError(f"Run ended with status: {run.status}")


print(run.iterations_completed, run.max_iterations, run.best_win_rate)
```

`wait_for_completion` backs off from 10 to 60 seconds and has a four-hour default timeout. A timeout stops waiting, not training. Call it again with the same run ID to resume tracking.

To control the polling schedule yourself:

```
import time


while True:
    run = client.autoscientist.get(run.id)
    print(run.status, run.iterations_completed, run.best_win_rate)


    if run.status == "succeeded":
        break
    if run.status == "failed":
        raise RuntimeError(f"failed: {run.error}")
    if run.status == "cancelled":
        raise RuntimeError("Run was cancelled")


    time.sleep(30)
```

A successful run either reached `target_win_rate` or used its final iteration. [Compare `best_win_rate` with the target](/autoscientist/interpreting-results/index.md) to determine which occurred. Use [`client.autoscientist.cancel(run.id)`](/api/python/resources/autoscientist/methods/cancel/index.md) to stop a run in flight.

## Constrain model size

In the app, **Tiny AutoScientist** limits selection to models under 10B parameters. With the SDK, review the [supported models](/autoscientist/supported-models/index.md), list the currently available models, and pass an eligible ID explicitly:

```
response = client.autoscientist.list_models()
for model in response.models:
    print(model.id)


model_id = input("Eligible model ID: ").strip()
run = client.autoscientist.create(
    dataset_id=dataset_id,
    model=model_id,
)
```

Model availability and metadata can change. Confirm the current values with [`autoscientist.list_models`](/api/python/resources/autoscientist/methods/list_models/index.md) rather than inferring model size from its ID.