(l-onnx-docai-onnx-ml-LinearRegressor)=

# ai.onnx.ml - LinearRegressor


(l-onnx-opai-onnx-ml-linearregressor-1)=

## LinearRegressor - 1 (ai.onnx.ml)

### Version

- **name**: [LinearRegressor (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators-ml.md#ai.onnx.ml.LinearRegressor)
- **domain**: `ai.onnx.ml`
- **since_version**: `1`
- **function**: `False`
- **support_level**: `SupportType.COMMON`
- **shape inference**: `False`

This version of the operator has been available
**since version 1 of domain ai.onnx.ml**.

### Summary

Generalized linear regression evaluation.<br>
If targets is set to 1 (default) then univariate regression is performed.<br>
If targets is set to M then M sets of coefficients must be passed in as a sequence
and M results will be output for each input n in N.<br>
The coefficients array is of length n, and the coefficients for each target are contiguous.
Intercepts are optional but if provided must match the number of targets.

### Attributes

* **coefficients - FLOATS** :

  Weights of the model(s).

* **intercepts - FLOATS** :

  Weights of the intercepts, if used.

* **post_transform - STRING** (default is `'NONE'`):

  Indicates the transform to apply to the regression output vector.<br>One of 'NONE,' 'SOFTMAX,' 'LOGISTIC,' 'SOFTMAX_ZERO,' or 'PROBIT'

* **targets - INT** (default is `'1'`):

  The total number of regression targets, 1 if not defined.

### Inputs

- **X** (heterogeneous) - **T**:

  Data to be regressed.

### Outputs

- **Y** (heterogeneous) - **tensor(float)**:

  Regression outputs (one per target, per example).

### Type Constraints

* **T** in ( `tensor(double)`, `tensor(float)`, `tensor(int32)`, `tensor(int64)` ):

  The input must be a tensor of a numeric type.