> ## Documentation Index
> Fetch the complete documentation index at: https://docs.priorlabs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Improving Performance

> Practical strategies to improve TabPFN performance beyond the default configuration.

TabPFN works well out of the box and handles many tasks natively that traditional ML pipelines require. We recommend feeding in data as raw as possible, as additional processing can hurt performance.
Avoid additional scaling with `StandardScaler` / `MinMaxScaler`, imputation of missing values, or one-hot encoding of categoricals.

## Escalation Path

When the default TabPFN does not meet your needs, try these approaches in roughly this order:

<Steps>
  <Step title="Feature engineering">
    Add domain features, extract datetime components, encode text meaningfully. This is usually the highest-impact change. See [Feature Engineering](/improving-performance/feature-engineering).
  </Step>

  <Step title="Feature selection">
    If you have many features (100+), try filtering to the most informative ones. See [Feature Selection](/improving-performance/feature-selection).
  </Step>

  <Step title="Metric tuning">
    Use `eval_metric` and `tuning_config` to optimize for your specific evaluation metric. See [Model Parameters](/improving-performance/model-parameters).
  </Step>

  <Step title="Preprocessing transforms">
    Experiment with different `PREPROCESS_TRANSFORMS` and target transforms. See [Preprocessing Transforms](/improving-performance/preprocessing).
  </Step>

  <Step title="Fine-tuning">
    [Fine-tune](/capabilities/fine-tuning) the pretrained model on your data when you have a specialized domain or distribution shift.
  </Step>
</Steps>

## Guides

<CardGroup cols={2}>
  <Card title="Feature Engineering" icon="wand-magic-sparkles" href="/improving-performance/feature-engineering">
    Encode domain knowledge into features that TabPFN cannot learn from raw columns alone.
  </Card>

  <Card title="Feature Selection" icon="filter" href="/improving-performance/feature-selection">
    Reduce feature count to improve attention efficiency and predictive power.
  </Card>

  <Card title="Preprocessing Transforms" icon="sliders" href="/improving-performance/preprocessing">
    Configure TabPFN's internal preprocessing pipeline for maximum ensemble diversity.
  </Card>

  <Card title="Model Parameters" icon="gear" href="/improving-performance/model-parameters">
    Tune softmax temperature, metric optimization, and class imbalance handling.
  </Card>
</CardGroup>

## Related

<Card title="Fine-Tuning" icon="wrench" href="/capabilities/fine-tuning">
  Adapt TabPFN's pretrained weights to your domain.
</Card>