> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to BigML

The BigML connector lets your AI agent act in each user's BigML account. Each user connects their own BigML API key once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 60 tools.

**Authentication:** Username + API Key
**Categories:** AI, Analytics
**Tools:** 60: 28 read, 21 write, 11 destructive
**Users sign in with:** Username and password
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=BIGML)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the BigML connection

   In **AgentKit > Connections**, create a BigML connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'bigml_anomalies_list',
  toolInput: {},
  connector: 'bigml',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="bigml_anomalies_list",
    tool_input={},
    connection_name="bigml",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `bigml_anomalies_list`

List Anomaly Detectors · Read-only

List your anomaly detectors, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (anomaly/<id>), names, status and creation dates. Use bigml_anomaly_get for one anomaly detector.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_anomaly_get`

Get Anomaly Detector · Read-only

Get one anomaly detector with its status (code 5 means finished) and full details. Returns the anomaly detector object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `anomaly_id` (`string`, required): ID of the anomaly detector: the 24-character part after anomaly/ (for example 5f3a1b2c4d5e6f7890123456 from anomaly/5f3a1b2c4d5e6f7890123456). Get it from bigml_anomalies_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"anomaly_id":"<anomaly_id>"}`

### `bigml_anomaly_score_get`

Get Anomaly Score · Read-only

Get one anomaly score with its status (code 5 means finished) and full details. Returns the anomaly score object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `anomaly_score_id` (`string`, required): ID of the anomaly score: the 24-character part after anomalyscore/ (for example 5f3a1b2c4d5e6f7890123456 from anomalyscore/5f3a1b2c4d5e6f7890123456). Get it from bigml_anomaly_scores_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"anomaly_score_id":"<anomaly_score_id>"}`

### `bigml_batch_prediction_get`

Get Batch Prediction · Read-only

Get one batch prediction with its status (code 5 means finished) and full details. Returns the batch prediction object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `batch_prediction_id` (`string`, required): ID of the batch prediction: the 24-character part after batchprediction/ (for example 5f3a1b2c4d5e6f7890123456 from batchprediction/5f3a1b2c4d5e6f7890123456). Get it from bigml_batch_predictions_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"batch_prediction_id":"<batch_prediction_id>"}`

### `bigml_batch_predictions_list`

List Batch Predictions · Read-only

List your batch predictions, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (batchprediction/<id>), names, status and creation dates. Use bigml_batch_prediction_get for one batch prediction.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_centroid_get`

Get Centroid · Read-only

Get one centroid with its status (code 5 means finished) and full details. Returns the centroid object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `centroid_id` (`string`, required): ID of the centroid: the 24-character part after centroid/ (for example 5f3a1b2c4d5e6f7890123456 from centroid/5f3a1b2c4d5e6f7890123456). Get it from bigml_centroids_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"centroid_id":"<centroid_id>"}`

### `bigml_cluster_get`

Get Cluster · Read-only

Get one cluster with its status (code 5 means finished) and full details. Returns the cluster object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cluster_id` (`string`, required): ID of the cluster: the 24-character part after cluster/ (for example 5f3a1b2c4d5e6f7890123456 from cluster/5f3a1b2c4d5e6f7890123456). Get it from bigml_clusters_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"cluster_id":"<cluster_id>"}`

### `bigml_clusters_list`

List Clusters · Read-only

List your clusters, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (cluster/<id>), names, status and creation dates. Use bigml_cluster_get for one cluster.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_dataset_get`

Get Dataset · Read-only

Get one dataset with its status (code 5 means finished) and full details. Returns the dataset object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): ID of the dataset: the 24-character part after dataset/ (for example 5f3a1b2c4d5e6f7890123456 from dataset/5f3a1b2c4d5e6f7890123456). Get it from bigml_datasets_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"dataset_id":"<dataset_id>"}`

### `bigml_datasets_list`

List Datasets · Read-only

List your datasets, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (dataset/<id>), names, status and creation dates. Use bigml_dataset_get for one dataset.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_deepnet_get`

Get Deepnet · Read-only

Get one deepnet with its status (code 5 means finished) and full details. Returns the deepnet object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deepnet_id` (`string`, required): ID of the deepnet: the 24-character part after deepnet/ (for example 5f3a1b2c4d5e6f7890123456 from deepnet/5f3a1b2c4d5e6f7890123456). Get it from bigml_deepnets_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"deepnet_id":"<deepnet_id>"}`

### `bigml_deepnets_list`

List Deepnets · Read-only

List your deepnets, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (deepnet/<id>), names, status and creation dates. Use bigml_deepnet_get for one deepnet.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_ensemble_get`

Get Ensemble · Read-only

Get one ensemble with its status (code 5 means finished) and full details. Returns the ensemble object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ensemble_id` (`string`, required): ID of the ensemble: the 24-character part after ensemble/ (for example 5f3a1b2c4d5e6f7890123456 from ensemble/5f3a1b2c4d5e6f7890123456). Get it from bigml_ensembles_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"ensemble_id":"<ensemble_id>"}`

### `bigml_ensembles_list`

List Ensembles · Read-only

List your ensembles, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (ensemble/<id>), names, status and creation dates. Use bigml_ensemble_get for one ensemble.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_evaluation_get`

Get Evaluation · Read-only

Get one evaluation with its status (code 5 means finished) and full details. Returns the evaluation object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `evaluation_id` (`string`, required): ID of the evaluation: the 24-character part after evaluation/ (for example 5f3a1b2c4d5e6f7890123456 from evaluation/5f3a1b2c4d5e6f7890123456). Get it from bigml_evaluations_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"evaluation_id":"<evaluation_id>"}`

### `bigml_evaluations_list`

List Evaluations · Read-only

List your evaluations, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (evaluation/<id>), names, status and creation dates. Use bigml_evaluation_get for one evaluation.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_logistic_regression_get`

Get Logistic Regression · Read-only

Get one logistic regression with its status (code 5 means finished) and full details. Returns the logistic regression object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `logistic_regression_id` (`string`, required): ID of the logistic regression: the 24-character part after logisticregression/ (for example 5f3a1b2c4d5e6f7890123456 from logisticregression/5f3a1b2c4d5e6f7890123456). Get it from bigml_logistic_regressions_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"logistic_regression_id":"<logistic_regression_id>"}`

### `bigml_logistic_regressions_list`

List Logistic Regressions · Read-only

List your logistic regressions, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (logisticregression/<id>), names, status and creation dates. Use bigml_logistic_regression_get for one logistic regression.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_model_get`

Get Model · Read-only

Get one model with its status (code 5 means finished) and full details. Returns the model object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `model_id` (`string`, required): ID of the model: the 24-character part after model/ (for example 5f3a1b2c4d5e6f7890123456 from model/5f3a1b2c4d5e6f7890123456). Get it from bigml_models_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"model_id":"<model_id>"}`

### `bigml_models_list`

List Models · Read-only

List your models, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (model/<id>), names, status and creation dates. Use bigml_model_get for one model.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_optiml_get`

Get OptiML · Read-only

Get one optiml with its status (code 5 means finished) and full details. Returns the optiml object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `optiml_id` (`string`, required): ID of the optiml: the 24-character part after optiml/ (for example 5f3a1b2c4d5e6f7890123456 from optiml/5f3a1b2c4d5e6f7890123456). Get it from bigml_optimls_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"optiml_id":"<optiml_id>"}`

### `bigml_prediction_get`

Get Prediction · Read-only

Get one prediction with its status (code 5 means finished) and full details. Returns the prediction object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `prediction_id` (`string`, required): ID of the prediction: the 24-character part after prediction/ (for example 5f3a1b2c4d5e6f7890123456 from prediction/5f3a1b2c4d5e6f7890123456). Get it from bigml_predictions_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"prediction_id":"<prediction_id>"}`

### `bigml_predictions_list`

List Predictions · Read-only

List your predictions, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (prediction/<id>), names, status and creation dates. Use bigml_prediction_get for one prediction.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_project_get`

Get Project · Read-only

Get one project with its status (code 5 means finished) and full details. Returns the project object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): ID of the project: the 24-character part after project/ (for example 5f3a1b2c4d5e6f7890123456 from project/5f3a1b2c4d5e6f7890123456). Get it from bigml_projects_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"project_id":"<project_id>"}`

### `bigml_projects_list`

List Projects · Read-only

List your projects, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (project/<id>), names, status and creation dates. Use bigml_project_get for one project.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_source_get`

Get Source · Read-only

Get one source with its status (code 5 means finished) and full details. Returns the source object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `source_id` (`string`, required): ID of the source: the 24-character part after source/ (for example 5f3a1b2c4d5e6f7890123456 from source/5f3a1b2c4d5e6f7890123456). Get it from bigml_sources_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"source_id":"<source_id>"}`

### `bigml_sources_list`

List Sources · Read-only

List your sources, newest first, with paging, ordering and filters by name, tags or project. Returns resource objects with IDs (source/<id>), names, status and creation dates. Use bigml_source_get for one source.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Number of resources to return (up to 200). Defaults to 20. Example: 20.
- `name_contains` (`string`, optional): Only resources whose name contains this word (case-insensitive). Example: iris.
- `offset` (`integer`, optional): Position to start listing from, for paging. Defaults to 0. Example: 0.
- `order_by` (`string`, optional): Sortable field to order by; prefix with - for descending. Defaults to -created. Example: -created.
- `project` (`string`, optional): Only resources in this project (project/<id>). Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`string`, optional): Only resources tagged with any of these comma-separated tags. Example: fraud,test.

Example input: `{}`

### `bigml_time_series_get`

Get Time Series · Read-only

Get one time series with its status (code 5 means finished) and full details. Returns the time series object. Resources are built in the background, so poll this tool after creating one.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `time_series_id` (`string`, required): ID of the time series: the 24-character part after timeseries/ (for example 5f3a1b2c4d5e6f7890123456 from timeseries/5f3a1b2c4d5e6f7890123456). Get it from bigml_time_series_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"time_series_id":"<time_series_id>"}`

### `bigml_anomaly_create`

Create Anomaly Detector · Write

Create an anomaly detector (isolation forest) from a dataset. Returns the new resource with its ID (anomaly/<id>) and status; poll bigml_anomaly_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented anomaly detector arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `forest_size` (`integer`, optional): The number of trees used by the anomaly detector. Must be a number greater than or equal to 2 and less than or equal to 1000. Example: 128.
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]
- `top_n` (`integer`, optional): The number of instances that will be returned together with the anomaly detector that were scored as most anomalous. The minimum number is 1 and the maximum is 1024. Example: 10.

Example input: `{"dataset":"<dataset>"}`

### `bigml_anomaly_score_create`

Create Anomaly Score · Write

Create an anomaly score between 0 and 1 for the given input data. Returns the new resource with its ID (anomalyscore/<id>) and status; poll bigml_anomaly_score_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `anomaly` (`string`, required): Anomaly detector ID (anomaly/<id>). Get IDs from bigml_anomalies_list. Example: anomaly/5f3a1b2c4d5e6f7890123456.
- `input_data` (`object`, required): Input values as an object keyed by field ID or field name, for example {"000000": 5.1, "petal width": 0.2}. Get field IDs from the model's fields. Example: {"000000": 5.1, "000001": 3.5}
- `extra_arguments` (`object`, optional): Any other documented anomaly score arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.

Example input: `{"anomaly":"<anomaly>","input_data":{"000000":5.1,"000001":3.5}}`

### `bigml_batch_prediction_create`

Create Batch Prediction · Write

Create a batch prediction that scores every row of a dataset with a supervised model and writes the results to a new dataset or CSV. Returns the new resource with its ID (batchprediction/<id>) and status; poll bigml_batch_prediction_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `all_fields` (`boolean`, optional): Whether all the fields from the dataset should be part of the generated csv file together with the predictions. Example: true.
- `confidence` (`boolean`, optional): Whether the confidence for each prediction for the model or non-boosted ensemble should be added to the each csv file. For logistic regressions, it is accepted but deprecated in favor of probability. Example: true.
- `deepnet` (`string`, optional): Deepnet ID (deepnet/<id>). Get IDs from bigml_deepnets_list. Example: deepnet/5f3a1b2c4d5e6f7890123456.
- `ensemble` (`string`, optional): Ensemble ID (ensemble/<id>). Get IDs from bigml_ensembles_list. Example: ensemble/5f3a1b2c4d5e6f7890123456.
- `extra_arguments` (`object`, optional): Any other documented batch prediction arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `fusion` (`string`, optional): Fusion ID (fusion/<id>). Example: fusion/5f3a1b2c4d5e6f7890123456.
- `logisticregression` (`string`, optional): Logistic regression ID (logisticregression/<id>). Get IDs from bigml_logistic_regressions_list. Example: logisticregression/5f3a1b2c4d5e6f7890123456.
- `model` (`string`, optional): Model ID (model/<id>). Get IDs from bigml_models_list. Example: model/5f3a1b2c4d5e6f7890123456.
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `output_dataset` (`boolean`, optional): Whether a dataset with the results should be automatically created or not. Example: true.
- `output_fields` (`array`, optional): Specifies the fields to be included in the csv file. It can be a list of field ids or names. Example: ["000000"]
- `probability` (`boolean`, optional): Whether the probability for each prediction for the classification task should be added. Example: true.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.

Example input: `{"dataset":"<dataset>"}`

### `bigml_centroid_create`

Create Centroid · Write

Create a centroid: the nearest cluster for the given input data. Returns the new resource with its ID (centroid/<id>) and status; poll bigml_centroid_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `cluster` (`string`, required): Cluster ID (cluster/<id>). Get IDs from bigml_clusters_list. Example: cluster/5f3a1b2c4d5e6f7890123456.
- `input_data` (`object`, required): Input values as an object keyed by field ID or field name, for example {"000000": 5.1, "petal width": 0.2}. Get field IDs from the model's fields. Example: {"000000": 5.1, "000001": 3.5}
- `extra_arguments` (`object`, optional): Any other documented centroid arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.

Example input: `{"cluster":"<cluster>","input_data":{"000000":5.1,"000001":3.5}}`

### `bigml_cluster_create`

Create Cluster · Write

Create a cluster analysis (k-means or g-means) of a dataset. Returns the new resource with its ID (cluster/<id>) and status; poll bigml_cluster_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `critical_value` (`integer`, optional): The clustering algorithm G-means is parameter free except for one, the critical_value parameter. G-means iteratively takes existing clusters and tests whether the cluster's neighborhood appears Gaussian. If it doesn't the cluster is split into two. The critical_value sets how strict the test is when deciding whether data looks Gaussian. The default is to 5, which seems to work well in most cases. A range of 1 - 10 is acceptable. A critical_value of 1 means data must look very Gaussian to... Example: 5.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented cluster arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `k` (`integer`, optional): The number of clusters. Must be null or a number greater than or equal to 1 and less than or equal to 300. Example: 3.
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_dataset_create`

Create Dataset · Write

Create a dataset (a structured, summarised version of the data) from a source, or a filtered or sampled copy of another dataset. Returns the new resource with its ID (dataset/<id>) and status; poll bigml_dataset_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented dataset arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `lisp_filter` (`string`, optional): A string representing a Lisp s-expression to filter rows from the datasource. Example: (> (f "000001") 3)
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `objective_field` (`object`, optional): ID of the field to predict, for example 000004. Defaults to the dataset's objective (usually the last field). Example: 000004.
- `origin_dataset` (`string`, optional): The dataset/id of dataset to be transformed. See the Section on Transformations for more details. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `size` (`integer`, optional): The number of bytes from the source that you want to use. Example: 1048576.
- `source` (`string`, optional): The source ID (source/<id>) to build the dataset from. Get IDs from bigml_sources_list. Example: source/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{}`

### `bigml_dataset_update`

Update Dataset · Write

Update a dataset's name, description, tags or project. Returns the updated dataset (HTTP 202).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): ID of the dataset: the 24-character part after dataset/ (for example 5f3a1b2c4d5e6f7890123456 from dataset/5f3a1b2c4d5e6f7890123456). Get it from bigml_datasets_list. Example: 5f3a1b2c4d5e6f7890123456.
- `category` (`integer`, optional): Category code that best describes the resource (see BigML category codes). Example: 1.
- `description` (`string`, optional): New description. Example: Retrained on the full dataset.
- `name` (`string`, optional): New name. Example: Iris classifier v2.
- `project` (`string`, optional): Project ID (project/<id>) to move the resource to. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): New list of tags; replaces the current tags. Example: ["iris", "prod"]

Example input: `{"dataset_id":"<dataset_id>"}`

### `bigml_deepnet_create`

Create Deepnet · Write

Create a deep neural network from a dataset, with automatic network search by default. Returns the new resource with its ID (deepnet/<id>) and status; poll bigml_deepnet_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented deepnet arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `max_training_time` (`integer`, optional): The maximum training time, in seconds, for which to train the network. Training may stop early if performance on a holdout set does not show improvement. Use -1 to deactivate the max training time. This value takes precedence over training_duration if specified. Example: 1800.
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `objective_field` (`string`, optional): ID of the field to predict, for example 000004. Defaults to the dataset's objective (usually the last field). Example: 000004.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `search` (`boolean`, optional): During the deepnet creation, BigML trains and evaluates over all possible network configurations, returning the best networks found for the problem. The final deepnet returned by the search is a compromise between the top n networks found in the search. Since this option builds several networks, it may be significantly slower than the suggest_structure technique. When it is set to true, only the following modeling properties are applied: default_numeric_value, input_fields, excluded_fields,... Example: true.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_ensemble_create`

Create Ensemble · Write

Create an ensemble (random decision forest, bagging or boosted trees) from a dataset. Returns the new resource with its ID (ensemble/<id>) and status; poll bigml_ensemble_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `boosting` (`object`, optional): Gradient boosting options for the ensemble. Required to created an ensemble with boosted trees. See the Gradient Boosting section for more information. Example: {"iterations": 20, "learning_rate": 0.1}
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented ensemble arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `number_of_models` (`integer`, optional): The number of models to build the ensemble. This parameter is ignored for boosted trees. See the Gradient Boosting section for more information. Example: 10.
- `objective_field` (`string`, optional): ID of the field to predict, for example 000004. Defaults to the dataset's objective (usually the last field). Example: 000004.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `randomize` (`boolean`, optional): Setting this parameter to true will consider only a subset of the possible fields when choosing a split. See the Section on Random Decision Forests below. Example: true.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_evaluation_create`

Create Evaluation · Write

Create an evaluation of a model, ensemble, deepnet, logistic regression or fusion against a test dataset, with accuracy and other metrics. Returns the new resource with its ID (evaluation/<id>) and status; poll bigml_evaluation_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `deepnet` (`string`, optional): Deepnet ID (deepnet/<id>). Get IDs from bigml_deepnets_list. Example: deepnet/5f3a1b2c4d5e6f7890123456.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `ensemble` (`string`, optional): Ensemble ID (ensemble/<id>). Get IDs from bigml_ensembles_list. Example: ensemble/5f3a1b2c4d5e6f7890123456.
- `extra_arguments` (`object`, optional): Any other documented evaluation arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `fusion` (`string`, optional): Fusion ID (fusion/<id>). Example: fusion/5f3a1b2c4d5e6f7890123456.
- `logisticregression` (`string`, optional): Logistic regression ID (logisticregression/<id>). Get IDs from bigml_logistic_regressions_list. Example: logisticregression/5f3a1b2c4d5e6f7890123456.
- `model` (`string`, optional): Model ID (model/<id>). Get IDs from bigml_models_list. Example: model/5f3a1b2c4d5e6f7890123456.
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_forecast_create`

Create Forecast · Write

Create a forecast of future values from a time series model. Returns the new resource with its ID (forecast/<id>) and status; poll bigml_forecast_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `input_data` (`object`, required): Input values as an object keyed by field ID or field name, for example {"000000": 5.1, "petal width": 0.2}. Get field IDs from the model's fields. Example: {"000000": 5.1, "000001": 3.5}
- `timeseries` (`string`, required): Time series ID (timeseries/<id>). Get IDs from bigml_time_series_list. Example: timeseries/5f3a1b2c4d5e6f7890123456.
- `extra_arguments` (`object`, optional): Any other documented forecast arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.

Example input: `{"input_data":{"000000":5.1,"000001":3.5},"timeseries":"<timeseries>"}`

### `bigml_logistic_regression_create`

Create Logistic Regression · Write

Create a logistic regression classifier from a dataset. Returns the new resource with its ID (logisticregression/<id>) and status; poll bigml_logistic_regression_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `c` (`number`, optional): The inverse of the regularization strength. Must be greater than 0. Example: 1.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented logistic regression arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `objective_field` (`string`, optional): ID of the field to predict, for example 000004. Defaults to the dataset's objective (usually the last field). Example: 000004.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `regularization` (`string`, optional): Either l1 or l2, which selects the norm to minimize when regularizing the solution. Regularizing with respect to the l1 norm causes more coefficients to be zero, using the l2 norm forces the magnitudes of all coefficients towards zero. Example: l2.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_model_create`

Create Model · Write

Create a decision tree model from a dataset, predicting the objective field (the last field by default). Returns the new resource with its ID (model/<id>) and status; poll bigml_model_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `excluded_fields` (`array`, optional): IDs of fields to leave out. Example: ["000003"]
- `extra_arguments` (`object`, optional): Any other documented model arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `input_fields` (`array`, optional): IDs of the fields to use as inputs. Defaults to all fields. Example: ["000000", "000001", "000002"]
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `objective_field` (`string`, optional): ID of the field to predict, for example 000004. Defaults to the dataset's objective (usually the last field). Example: 000004.
- `optimize` (`boolean`, optional): Whether the model should be built with the automatic optimization. When it is set to true, only the following modeling properties are applied: default_numeric_value, excluded_fields, input_fields, max_training_time, missing_splits, number_of_model_candidates, objective_field, objective_weights, sample_rate, and weight_field. Example: true.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `sample_rate` (`number`, optional): A real number between 0 and 1 specifying the sample rate. See the Section on Sampling for more details. Example: 0.8.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_model_update`

Update Model · Write

Update a model's name, description, tags or project. Returns the updated model (HTTP 202).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `model_id` (`string`, required): ID of the model: the 24-character part after model/ (for example 5f3a1b2c4d5e6f7890123456 from model/5f3a1b2c4d5e6f7890123456). Get it from bigml_models_list. Example: 5f3a1b2c4d5e6f7890123456.
- `category` (`integer`, optional): Category code that best describes the resource (see BigML category codes). Example: 1.
- `description` (`string`, optional): New description. Example: Retrained on the full dataset.
- `name` (`string`, optional): New name. Example: Iris classifier v2.
- `project` (`string`, optional): Project ID (project/<id>) to move the resource to. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): New list of tags; replaces the current tags. Example: ["iris", "prod"]

Example input: `{"model_id":"<model_id>"}`

### `bigml_optiml_create`

Create OptiML · Write

Create an OptiML: an automatic search over model types and parameters that returns the best-performing models for a dataset. Returns the new resource with its ID (optiml/<id>) and status; poll bigml_optiml_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `extra_arguments` (`object`, optional): Any other documented optiml arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `max_training_time` (`integer`, optional): The maximum training time allowed for the optimization, in seconds, as a strictly positive integer. Example: 1800.
- `metric` (`string`, optional): The metric used to evaluate models used. For classification problems, you can use one of these: accuracy, area_under_pr_curve, area_under_roc_curve, f_measure, kendalls_tau_b, ks_statistic, max_phi, phi_coefficient, precision, recall, and spearmans_rho. For regression problems, r_squared is the only supported metric. Deepnet optimization is an independent process that uses a combination of several metrics, hence it is not affected by the configuration of the metric. Example: accuracy.
- `model_types` (`array`, optional): A list of the types of model to create during the search. Available model types are model, ensemble, linearregression, logisticregression, deepnet. If not specified, all model types are explored according the objective field type. Example: ["model", "ensemble", "deepnet"]
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `number_of_model_candidates` (`integer`, optional): The number of model candidates evaluated over the course of the optimization. Maximum 200 candidates. Example: 128.
- `objective_field` (`string`, optional): ID of the field to predict, for example 000004. Defaults to the dataset's objective (usually the last field). Example: 000004.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_prediction_create`

Create Prediction · Write

Create a single prediction for the given input data from a model, ensemble, deepnet, logistic regression, linear regression or fusion. Returns the new resource with its ID (prediction/<id>) and status; poll bigml_prediction_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `input_data` (`object`, required): Input values as an object keyed by field ID or field name, for example {"000000": 5.1, "petal width": 0.2}. Get field IDs from the model's fields. Example: {"000000": 5.1, "000001": 3.5}
- `deepnet` (`string`, optional): Deepnet ID (deepnet/<id>). Get IDs from bigml_deepnets_list. Example: deepnet/5f3a1b2c4d5e6f7890123456.
- `ensemble` (`string`, optional): Ensemble ID (ensemble/<id>). Get IDs from bigml_ensembles_list. Example: ensemble/5f3a1b2c4d5e6f7890123456.
- `explain` (`boolean`, optional): Whether you want to request an explanation for the prediction. Example: true.
- `extra_arguments` (`object`, optional): Any other documented prediction arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `fusion` (`string`, optional): Fusion ID (fusion/<id>). Example: fusion/5f3a1b2c4d5e6f7890123456.
- `linearregression` (`string`, optional): Linear regression ID (linearregression/<id>). Example: linearregression/5f3a1b2c4d5e6f7890123456.
- `logisticregression` (`string`, optional): Logistic regression ID (logisticregression/<id>). Get IDs from bigml_logistic_regressions_list. Example: logisticregression/5f3a1b2c4d5e6f7890123456.
- `missing_strategy` (`integer`, optional): Specifies the method that should be used when a missing split is found for a model, ensemble or fusion. That is, when a missing value is found in the input data for a decision node. The options are:0: last prediction predicts based on the subset of the data which reached the parent of the missing split.1: proportional evaluates all the subtrees of a missing split and recombines their predictions based on the proportion of data in each subtree. Example: 0.
- `model` (`string`, optional): Model ID (model/<id>). Get IDs from bigml_models_list. Example: model/5f3a1b2c4d5e6f7890123456.
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.

Example input: `{"input_data":{"000000":5.1,"000001":3.5}}`

### `bigml_project_create`

Create Project · Write

Create a project that groups related resources. Returns the new resource with its ID (project/<id>) and status; poll bigml_project_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `name` (`string`, required): Name of the resource. Example: Iris classifier.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `extra_arguments` (`object`, optional): Any other documented project arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"name":"<name>"}`

### `bigml_project_update`

Update Project · Write

Update a project's name, description, tags or project. Returns the updated project (HTTP 202).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): ID of the project: the 24-character part after project/ (for example 5f3a1b2c4d5e6f7890123456 from project/5f3a1b2c4d5e6f7890123456). Get it from bigml_projects_list. Example: 5f3a1b2c4d5e6f7890123456.
- `category` (`integer`, optional): Category code that best describes the resource (see BigML category codes). Example: 1.
- `description` (`string`, optional): New description. Example: Retrained on the full dataset.
- `name` (`string`, optional): New name. Example: Iris classifier v2.
- `tags` (`array`, optional): New list of tags; replaces the current tags. Example: ["iris", "prod"]

Example input: `{"project_id":"<project_id>"}`

### `bigml_source_create`

Create Source · Write

Create a source from a remote URL (http, https, s3, gs, azure, odata...) or inline CSV data; BigML parses it in the background. Returns the new resource with its ID (source/<id>) and status; poll bigml_source_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `data` (`string`, optional): Inline data in CSV format, with a header row. Use instead of remote. Example: a,b,c
1,2,3
4,5,6.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `extra_arguments` (`object`, optional): Any other documented source arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `remote` (`string`, optional): URL of the data to import: http(s), s3, gs, azure, dropbox or odata. Use instead of data. Example: https://static.bigml.com/csv/iris.csv.
- `source_parser` (`object`, optional): How to parse the data: header (true/false), separator, quote, missing_tokens, locale. Example: {"header": true, "separator": ","}
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{}`

### `bigml_source_update`

Update Source · Write

Update a source's name, description, tags or project. Returns the updated source (HTTP 202).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `source_id` (`string`, required): ID of the source: the 24-character part after source/ (for example 5f3a1b2c4d5e6f7890123456 from source/5f3a1b2c4d5e6f7890123456). Get it from bigml_sources_list. Example: 5f3a1b2c4d5e6f7890123456.
- `category` (`integer`, optional): Category code that best describes the resource (see BigML category codes). Example: 1.
- `description` (`string`, optional): New description. Example: Retrained on the full dataset.
- `name` (`string`, optional): New name. Example: Iris classifier v2.
- `project` (`string`, optional): Project ID (project/<id>) to move the resource to. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): New list of tags; replaces the current tags. Example: ["iris", "prod"]

Example input: `{"source_id":"<source_id>"}`

### `bigml_time_series_create`

Create Time Series · Write

Create a time series (exponential smoothing) model from a dataset of numeric fields. Returns the new resource with its ID (timeseries/<id>) and status; poll bigml_time_series_get until status code is 5 (finished). Pass any other documented argument in extra_arguments.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset` (`string`, required): A dataset ID (dataset/<id>). Get IDs from bigml_datasets_list. Example: dataset/5f3a1b2c4d5e6f7890123456.
- `description` (`string`, optional): Description (up to 8192 characters). Example: Model trained on the iris dataset.
- `extra_arguments` (`object`, optional): Any other documented time series arguments as a JSON object, merged into the request body. Example: {"category": 1}
- `horizon` (`integer`, optional): The time series model includes a forecast with intervals for all computed ETS models with this horizon. Max value: 400. Example: 50.
- `name` (`string`, optional): Name of the resource. Example: Iris classifier.
- `objective_fields` (`array`, optional): One or more identifiers of numeric fields for which to fit ETS time series models. Non-numeric fields will be ignored, and if not present, the right-most valid field in the dataset will be used. Example: ["000001"]
- `period` (`integer`, optional): The number of data points per period. The period needs to be set taking into account the time interval of your instances and the seasonal frequency. For example, for monthly data and annual seasonality, the period should be 12, for daily data and weekly seasonality, the period should be 7. It can take values from 0 to 60. If the period is set to 1, there is no seasonality. If the period is 0, or not given, BigML will automatically learn the period in your data. Example: 12.
- `project` (`string`, optional): Project ID (project/<id>) to put the resource in. Get IDs from bigml_projects_list. Example: project/5f3a1b2c4d5e6f7890123456.
- `tags` (`array`, optional): Tags that help classify and index the resource. Example: ["iris", "2026"]

Example input: `{"dataset":"<dataset>"}`

### `bigml_anomaly_delete`

Delete Anomaly Detector · Destructive

Permanently delete a anomaly detector. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `anomaly_id` (`string`, required): ID of the anomaly detector: the 24-character part after anomaly/ (for example 5f3a1b2c4d5e6f7890123456 from anomaly/5f3a1b2c4d5e6f7890123456). Get it from bigml_anomalies_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"anomaly_id":"<anomaly_id>"}`

### `bigml_cluster_delete`

Delete Cluster · Destructive

Permanently delete a cluster. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `cluster_id` (`string`, required): ID of the cluster: the 24-character part after cluster/ (for example 5f3a1b2c4d5e6f7890123456 from cluster/5f3a1b2c4d5e6f7890123456). Get it from bigml_clusters_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"cluster_id":"<cluster_id>"}`

### `bigml_dataset_delete`

Delete Dataset · Destructive

Permanently delete a dataset. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): ID of the dataset: the 24-character part after dataset/ (for example 5f3a1b2c4d5e6f7890123456 from dataset/5f3a1b2c4d5e6f7890123456). Get it from bigml_datasets_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"dataset_id":"<dataset_id>"}`

### `bigml_deepnet_delete`

Delete Deepnet · Destructive

Permanently delete a deepnet. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `deepnet_id` (`string`, required): ID of the deepnet: the 24-character part after deepnet/ (for example 5f3a1b2c4d5e6f7890123456 from deepnet/5f3a1b2c4d5e6f7890123456). Get it from bigml_deepnets_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"deepnet_id":"<deepnet_id>"}`

### `bigml_ensemble_delete`

Delete Ensemble · Destructive

Permanently delete a ensemble. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `ensemble_id` (`string`, required): ID of the ensemble: the 24-character part after ensemble/ (for example 5f3a1b2c4d5e6f7890123456 from ensemble/5f3a1b2c4d5e6f7890123456). Get it from bigml_ensembles_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"ensemble_id":"<ensemble_id>"}`

### `bigml_evaluation_delete`

Delete Evaluation · Destructive

Permanently delete a evaluation. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `evaluation_id` (`string`, required): ID of the evaluation: the 24-character part after evaluation/ (for example 5f3a1b2c4d5e6f7890123456 from evaluation/5f3a1b2c4d5e6f7890123456). Get it from bigml_evaluations_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"evaluation_id":"<evaluation_id>"}`

### `bigml_logistic_regression_delete`

Delete Logistic Regression · Destructive

Permanently delete a logistic regression. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `logistic_regression_id` (`string`, required): ID of the logistic regression: the 24-character part after logisticregression/ (for example 5f3a1b2c4d5e6f7890123456 from logisticregression/5f3a1b2c4d5e6f7890123456). Get it from bigml_logistic_regressions_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"logistic_regression_id":"<logistic_regression_id>"}`

### `bigml_model_delete`

Delete Model · Destructive

Permanently delete a model. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `model_id` (`string`, required): ID of the model: the 24-character part after model/ (for example 5f3a1b2c4d5e6f7890123456 from model/5f3a1b2c4d5e6f7890123456). Get it from bigml_models_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"model_id":"<model_id>"}`

### `bigml_prediction_delete`

Delete Prediction · Destructive

Permanently delete a prediction. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `prediction_id` (`string`, required): ID of the prediction: the 24-character part after prediction/ (for example 5f3a1b2c4d5e6f7890123456 from prediction/5f3a1b2c4d5e6f7890123456). Get it from bigml_predictions_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"prediction_id":"<prediction_id>"}`

### `bigml_project_delete`

Delete Project · Destructive

Permanently delete a project. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): ID of the project: the 24-character part after project/ (for example 5f3a1b2c4d5e6f7890123456 from project/5f3a1b2c4d5e6f7890123456). Get it from bigml_projects_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"project_id":"<project_id>"}`

### `bigml_source_delete`

Delete Source · Destructive

Permanently delete a source. Returns an empty response; a resource in use by another one may be refused with 400.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `source_id` (`string`, required): ID of the source: the 24-character part after source/ (for example 5f3a1b2c4d5e6f7890123456 from source/5f3a1b2c4d5e6f7890123456). Get it from bigml_sources_list. Example: 5f3a1b2c4d5e6f7890123456.

Example input: `{"source_id":"<source_id>"}`

## Related

Other AI connectors ([all 216](/agentkit/connectors/?category=ai)):

- [Chorus](/agentkit/connectors/chorus/): Scalekit connector, Username and password, 9 tools
- [Gong](/agentkit/connectors/gong/): Scalekit connector, OAuth, 69 tools
- [Jiminny](/agentkit/connectors/jiminny/): Scalekit connector, Access token, 27 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
