---
metadata:
  - name: generator
    content: Diplodoc Platform v5.50.3
alternate:
  - https://catboost.ai/docs/en/concepts/output-data_feature-analysis_feature-interaction-strength.md
---
> **Documentation Index:** Fetch the complete configuration index at https://catboost.ai/docs/en/llms.txt

# Feature interaction strength

The following types of [feature interaction strength](https://catboost.ai/docs/en/concepts/fstr.md) files are created depending on the task and the execution parameters:
- [Interaction](#per-feature-interaction-strength)
- [InternalInteraction](#internal-interaction-strength)

## Interaction {#per-feature-interaction-strength}

#### Contains

<!-- source: en/_includes/work_src/reusage-formats/feature-interaction-strength.md -->
The value of the feature interaction strength for each pair of features.
<!-- endsource: en/_includes/work_src/reusage-formats/feature-interaction-strength.md -->

#### Format

- The rows are sorted in descending order of the feature interaction strength value.
- Each row contains information related to one pair of features.

    Format:
    ```
    <feature interaction strength><\t><feature name 1><\t><feature name 2>
    ```

    - `feature interaction strength` is the value of the feature interaction strength.

    - `feature name` is the zero-based index of the feature.

    An alphanumeric identifier is used instead if specified in the corresponding `Num` or `Categ` column of the [input data](https://catboost.ai/docs/en/concepts/input-data_column-descfile.md).

    For example, let's assume that the [columns description](https://catboost.ai/docs/en/concepts/input-data_column-descfile.md) file has the following structure:
    ```
    0<\t>Label value<\t>
    1<\t>Num
    2<\t>Num<\t>ratio
    3<\t>Categ
    4<\t>Auxiliary
    5<\t>Num
    ```

    The input [dataset](https://catboost.ai/docs/en/concepts/input-data_values-file.md) file contains the following line:
    ```
    120<\t>80<\t>0.8<\t>rock<\t>some useless information<\t>12
    ```

    The table below shows the compliance between the given feature values and the feature indices.

    <!-- source: en/_includes/concepts/feature-interaction-strength_table.md -->
    | Feature value       | Feature index or name|
    |---------------------|----------------------|
    |          80         | 0                    |
    |          0.8        | ratio                |
    |          rock       | 2                    |
    |          12         | 3                    |
    <!-- endsource: en/_includes/concepts/feature-interaction-strength_table.md -->


#### Example

```
0.5<\t>0<\t>1
30<\t>1<\t>2
```

## InternalInteraction {#internal-interaction-strength}

#### Contains

<!-- source: en/_includes/work_src/reusage-formats/internal-feature-interaction-strength.md -->
The value of the feature interaction strength for each pair of features that are used in the model. Internally the model uses feature combinations as separate features. All feature combinations that are used in the model are listed separately. For example, if the model contains a feature named `F1` and a combination of features `{F2, F3}`, the interaction between `F1` and the combination of features `{F2, F3}` is listed in the output file.
<!-- endsource: en/_includes/work_src/reusage-formats/internal-feature-interaction-strength.md -->


#### Format

- The rows are sorted in descending order of the feature interaction strength value.
- Each row contains information related to one pair of features and/or their combinations.

    Format:
    ```
    <feature interaction strength><\t><feature name 1><\t><feature name 2>
    ```

    - `feature interaction strength` is the value of the internal feature interaction strength.
    - `feature name` is the zero-based index of the feature.

    An alphanumeric identifier is used instead if specified in the corresponding `Num` or `Categ` column of the [input data](https://catboost.ai/docs/en/concepts/input-data_column-descfile.md).

    For example, let's assume that the [columns description](https://catboost.ai/docs/en/concepts/input-data_column-descfile.md) file has the following structure:
    ```
    0<\t>Label value<\t>
    1<\t>Num
    2<\t>Num<\t>ratio
    3<\t>Categ
    4<\t>Auxiliary
    5<\t>Num
    ```

    The input [dataset](https://catboost.ai/docs/en/concepts/input-data_values-file.md) file contains the following line:
    ```
    120<\t>80<\t>0.8<\t>rock<\t>some useless information<\t>12
    ```

    The table below shows the compliance between the given feature values and the feature indices.

    <!-- source: en/_includes/concepts/feature-interaction-strength_table.md -->
    | Feature value       | Feature index or name|
    |---------------------|----------------------|
    |          80         | 0                    |
    |          0.8        | ratio                |
    |          rock       | 2                    |
    |          12         | 3                    |
    <!-- endsource: en/_includes/concepts/feature-interaction-strength_table.md -->

#### Example

```
0.4004860988<\t>15<\t>13
0.1134764975<\t>{4} prior_num=0 prior_denom=1 targetborder=0 type=Borders<\t>15
```