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

# set_baseline

<!-- source: en/_includes/work_src/reusage-python/set_baseline__desc.md -->
Set initial formula values for all input objects. The training starts from these values for all input objects instead of starting from zero.
<!-- endsource: en/_includes/work_src/reusage-python/set_baseline__desc.md -->


## Method call format {#call-format}

```python
set_baseline(baseline)
```

## Parameters {#parameters}

### baseline

#### Description

<!-- source: en/_includes/work_src/reusage-input-data/baseline__shortdesc.md -->
The initial formula values for all input objects.
<!-- endsource: en/_includes/work_src/reusage-input-data/baseline__shortdesc.md -->


**Possible types**

- list
- numpy.ndarray
- pandas.DataFrame
- pandas.Series
- [polars.Series](https://docs.pola.rs/api/python/stable/reference/series/index.html)
- [polars.DataFrame](https://docs.pola.rs/api/python/stable/reference/dataframe/index.html)

**Default value**

Required parameter

## Example {#example}

```python
import numpy as np
from catboost import Pool

train_data = [[76, 'blvd', 41, 50, 7],
              [75, 'today', 57, 0, 48],
              [70, 'letters', 33, 17, 7],
              [72, 'now', 43, 29, 12],
              [60, 'back', 2, 0, 1]]

label_values = [1, 0, 0, 1, 4]

input_pool = Pool(data = train_data,
                  label = label_values,
                  cat_features = [1])

input_pool.set_baseline([1, 3, 2, 1, 2])

input_pool.get_baseline()
```