<!--
{
  "availability" : [
    "iOS: 15.0.0 -",
    "iPadOS: 15.0.0 -",
    "macCatalyst: 15.0.0 -",
    "macOS: 10.15.0 -",
    "visionOS: 1.0.0 -"
  ],
  "documentType" : "symbol",
  "framework" : "CreateML",
  "identifier" : "/documentation/CreateML/MLImageClassifier/FeatureExtractorType/custom(_:)",
  "metadataVersion" : "0.1.0",
  "role" : "Case",
  "symbol" : {
    "kind" : "Case",
    "modules" : [
      "Create ML"
    ],
    "preciseIdentifier" : "s:8CreateML17MLImageClassifierV20FeatureExtractorTypeO6customyAeC06CustomeF0VcAEmF"
  },
  "title" : "MLImageClassifier.FeatureExtractorType.custom(_:)"
}
-->

# MLImageClassifier.FeatureExtractorType.custom(_:)

A feature extractor that you provide as a Core ML model file or a layer within that file.

```
case custom(MLImageClassifier.CustomFeatureExtractor)
```

## Parameters

`_`

custom feature extractor to be used for extracting features.

## Discussion

A custom feature extractor is a neural network model you provide to an [`MLImageClassifier`](/documentation/CreateML/MLImageClassifier). The feature
extractor can be a layer within the model or the model itself. In either case, the neural network must take
an image as input and output an <doc://com.apple.documentation/documentation/CoreML/MLMultiArray>.

---

Copyright &copy; 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)