# XGBoost runtime for MLServer
This package provides a MLServer runtime compatible with XGBoost.
## Usage
You can install the runtime, alongside `mlserver`, as:
```bash
pip install mlserver mlserver-xgboost
```
For further information on how to use MLServer with XGBoost, you can check out
this [worked out example](../../docs/examples/xgboost/README.md).
## XGBoost Artifact Type
The XGBoost inference runtime will expect that your model is serialised via one
of the following methods:
| Extension | Docs | Example |
| --------- | -------------------------------------------------------------------------------------------------------------------- | ---------------------------------- |
| `*.json` | [JSON Format](https://xgboost.readthedocs.io/en/stable/tutorials/saving_model.html#introduction-to-model-io) | `booster.save_model("model.json")` |
| `*.ubj` | [Binary JSON Format](https://xgboost.readthedocs.io/en/stable/tutorials/saving_model.html#introduction-to-model-io) | `booster.save_model("model.ubj")` |
| `*.bst` | [(Old) Binary Format](https://xgboost.readthedocs.io/en/stable/tutorials/saving_model.html#introduction-to-model-io) | `booster.save_model("model.bst")` |
````{note}
By default, the runtime will look for a file called `model.[json | ubj | bst]`.
However, this can be modified through the `parameters.uri` field of your
{class}`ModelSettings <mlserver.settings.ModelSettings>` config (see the
section on [Model Settings](../../docs/reference/model-settings.md) for more
details).
```{code-block} json
---
emphasize-lines: 3-5
---
{
"name": "foo",
"parameters": {
"uri": "./my-own-model-filename.json"
}
}
```
````
## Content Types
If no [content type](../../docs/user-guide/content-type) is present on the
request or metadata, the XGBoost runtime will try to decode the payload as a
[NumPy Array](../../docs/user-guide/content-type).
To avoid this, either send a different content type explicitly, or define the
correct one as part of your [model's
metadata](../../docs/reference/model-settings).
## Model Outputs
The XGBoost inference runtime exposes a number of outputs depending on the
model type.
These outputs match to the `predict` and `predict_proba` methods of the XGBoost
model.
| Output | Returned By Default | Availability |
| --------------- | ------------------- | --------------------------------------------------------------------- |
| `predict` | ✅ | Available on all XGBoost models. |
| `predict_proba` | ❌ | Only available on non-regressor models (i.e. `XGBClassifier` models). |
By default, the runtime will only return the output of `predict`.
However, you are able to control which outputs you want back through the
`outputs` field of your {class}`InferenceRequest
<mlserver.types.InferenceRequest>` payload.
For example, to only return the model's `predict_proba` output, you could
define a payload such as:
```{code-block} json
---
emphasize-lines: 10-12
---
{
"inputs": [
{
"name": "my-input",
"datatype": "INT32",
"shape": [2, 2],
"data": [1, 2, 3, 4]
}
],
"outputs": [
{ "name": "predict_proba" }
]
}
```
Raw data
{
"_id": null,
"home_page": null,
"name": "mlserver-xgboost",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.12,>=3.9",
"maintainer_email": null,
"keywords": null,
"author": "Seldon Technologies Ltd.",
"author_email": "hello@seldon.io",
"download_url": "https://files.pythonhosted.org/packages/34/43/7741dddb5babacce5f2e1638cabed3b8009dfc492757c18ccbe2f35adb46/mlserver_xgboost-1.6.1.tar.gz",
"platform": null,
"description": "# XGBoost runtime for MLServer\n\nThis package provides a MLServer runtime compatible with XGBoost.\n\n## Usage\n\nYou can install the runtime, alongside `mlserver`, as:\n\n```bash\npip install mlserver mlserver-xgboost\n```\n\nFor further information on how to use MLServer with XGBoost, you can check out\nthis [worked out example](../../docs/examples/xgboost/README.md).\n\n## XGBoost Artifact Type\n\nThe XGBoost inference runtime will expect that your model is serialised via one\nof the following methods:\n\n| Extension | Docs | Example |\n| --------- | -------------------------------------------------------------------------------------------------------------------- | ---------------------------------- |\n| `*.json` | [JSON Format](https://xgboost.readthedocs.io/en/stable/tutorials/saving_model.html#introduction-to-model-io) | `booster.save_model(\"model.json\")` |\n| `*.ubj` | [Binary JSON Format](https://xgboost.readthedocs.io/en/stable/tutorials/saving_model.html#introduction-to-model-io) | `booster.save_model(\"model.ubj\")` |\n| `*.bst` | [(Old) Binary Format](https://xgboost.readthedocs.io/en/stable/tutorials/saving_model.html#introduction-to-model-io) | `booster.save_model(\"model.bst\")` |\n\n````{note}\nBy default, the runtime will look for a file called `model.[json | ubj | bst]`.\nHowever, this can be modified through the `parameters.uri` field of your\n{class}`ModelSettings <mlserver.settings.ModelSettings>` config (see the\nsection on [Model Settings](../../docs/reference/model-settings.md) for more\ndetails).\n\n```{code-block} json\n---\nemphasize-lines: 3-5\n---\n{\n \"name\": \"foo\",\n \"parameters\": {\n \"uri\": \"./my-own-model-filename.json\"\n }\n}\n```\n````\n\n## Content Types\n\nIf no [content type](../../docs/user-guide/content-type) is present on the\nrequest or metadata, the XGBoost runtime will try to decode the payload as a\n[NumPy Array](../../docs/user-guide/content-type).\nTo avoid this, either send a different content type explicitly, or define the\ncorrect one as part of your [model's\nmetadata](../../docs/reference/model-settings).\n\n## Model Outputs\n\nThe XGBoost inference runtime exposes a number of outputs depending on the\nmodel type.\nThese outputs match to the `predict` and `predict_proba` methods of the XGBoost\nmodel.\n\n| Output | Returned By Default | Availability |\n| --------------- | ------------------- | --------------------------------------------------------------------- |\n| `predict` | \u2705 | Available on all XGBoost models. |\n| `predict_proba` | \u274c | Only available on non-regressor models (i.e. `XGBClassifier` models). |\n\nBy default, the runtime will only return the output of `predict`.\nHowever, you are able to control which outputs you want back through the\n`outputs` field of your {class}`InferenceRequest\n<mlserver.types.InferenceRequest>` payload.\n\nFor example, to only return the model's `predict_proba` output, you could\ndefine a payload such as:\n\n```{code-block} json\n---\nemphasize-lines: 10-12\n---\n{\n \"inputs\": [\n {\n \"name\": \"my-input\",\n \"datatype\": \"INT32\",\n \"shape\": [2, 2],\n \"data\": [1, 2, 3, 4]\n }\n ],\n \"outputs\": [\n { \"name\": \"predict_proba\" }\n ]\n}\n```\n",
"bugtrack_url": null,
"license": "Apache-2.0",
"summary": "XGBoost runtime for MLServer",
"version": "1.6.1",
"project_urls": null,
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "111f8eeab40df4120bdd36da45a0a067358e8ecdf8ac6dbd598843ffbbd47096",
"md5": "6fb6079d2e18fa624b09a1f0b12a6239",
"sha256": "d917d1dfbd80e76cb26fb7c1a84ee4489a09c60e434283c7f000191e37b13474"
},
"downloads": -1,
"filename": "mlserver_xgboost-1.6.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "6fb6079d2e18fa624b09a1f0b12a6239",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.12,>=3.9",
"size": 8042,
"upload_time": "2024-09-10T15:10:51",
"upload_time_iso_8601": "2024-09-10T15:10:51.497477Z",
"url": "https://files.pythonhosted.org/packages/11/1f/8eeab40df4120bdd36da45a0a067358e8ecdf8ac6dbd598843ffbbd47096/mlserver_xgboost-1.6.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "34437741dddb5babacce5f2e1638cabed3b8009dfc492757c18ccbe2f35adb46",
"md5": "1cf578796c5961a6846df019084697c0",
"sha256": "69717157deee2e302c9d7ec6712437f6c4f687faaea53f8126b8fe3daee549d9"
},
"downloads": -1,
"filename": "mlserver_xgboost-1.6.1.tar.gz",
"has_sig": false,
"md5_digest": "1cf578796c5961a6846df019084697c0",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.12,>=3.9",
"size": 7072,
"upload_time": "2024-09-10T15:10:53",
"upload_time_iso_8601": "2024-09-10T15:10:53.591053Z",
"url": "https://files.pythonhosted.org/packages/34/43/7741dddb5babacce5f2e1638cabed3b8009dfc492757c18ccbe2f35adb46/mlserver_xgboost-1.6.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-09-10 15:10:53",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "mlserver-xgboost"
}
JSON