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

# Position Weight Generator

> Sample sequences from position-specific logit distributions

<div class="page-hero">
  <img class="page-hero-banner" src="https://proto-bio.github.io/proto-assets/images/generator/position-weight/hero.png" alt="Position Weight Generator" />
</div>

<p class="entity-disclaimer">This generator is open source. Any third-party models, product names, or trademarks referenced are the property of their respective owners, and Proto is not affiliated with them.</p>

<hr class="entity-rule" />

<div class="tool-tab-bar entity-source-bar"><span class="tool-tab-wrap"><span class="tool-tab badge-source entity-source-tab"><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><polyline points="16 18 22 12 16 6" /><polyline points="8 6 2 12 8 18" /></svg> Source</span></span></div>

<a href="https://github.com/evo-design/proto-language/blob/d3b7822f74ea64747cc751a3b2ab1aa6b799ac47/proto_language/generator/position_weight_generator.py#L89" target="_blank" class="tab-panel source-panel entity-source-panel">
  <div class="source-info">
    <img noZoom src="https://github.com/evo-design.png?size=40" class="source-avatar" width="36" height="36" />

    <span class="source-path">evo-design/proto-language<span class="source-subpath">/proto\_language/generator/position\_weight\_generator.py</span></span>
  </div>

  <span class="panel-goto-btn source-goto-btn"><span><svg width="14" height="14" viewBox="0 0 24 24" fill="currentColor"><path d="M12 0C5.37 0 0 5.37 0 12c0 5.31 3.435 9.795 8.205 11.385.6.105.825-.255.825-.57 0-.285-.015-1.23-.015-2.235-3.015.555-3.795-.735-4.035-1.41-.135-.345-.72-1.41-1.23-1.695-.42-.225-1.02-.78-.015-.795.945-.015 1.62.87 1.845 1.23 1.08 1.815 2.805 1.305 3.495.99.105-.78.42-1.305.765-1.605-2.67-.3-5.46-1.335-5.46-5.925 0-1.305.465-2.385 1.23-3.225-.12-.3-.54-1.53.12-3.18 0 0 1.005-.315 3.3 1.23.96-.27 1.98-.405 3-.405s2.04.135 3 .405c2.295-1.56 3.3-1.23 3.3-1.23.66 1.65.24 2.88.12 3.18.765.84 1.23 1.905 1.23 3.225 0 4.605-2.805 5.625-5.475 5.925.435.375.81 1.095.81 2.22 0 1.605-.015 2.895-.015 3.3 0 .315.225.69.825.57A12.02 12.02 0 0024 12c0-6.63-5.37-12-12-12z" /></svg> View source</span></span>
</a>

<div class="entity-contributors"><span class="entity-contributors-label">Generator contributors</span><span class="entity-contributors-people"><a class="entity-contributor" href="https://github.com/dguo8412" target="_blank" rel="noopener" title="dguo8412: 3 commits"><img noZoom class="entity-contributor-avatar" src="https://avatars.githubusercontent.com/u/46211285?v=4&s=64" alt="" loading="lazy" /><span class="entity-contributor-login">dguo8412</span></a></span></div>
Convert logit distributions into discrete proposal sequences.

Reads `seq.logits` from each proposal sequence, applies softmax, and
writes the decoded string back to `seq.sequence`. Supports deterministic
argmax decoding or stochastic categorical sampling.

Designed for gradient-based optimizers that update `seq.logits` directly
and need discrete sequences for handoff or tracking.

## API Reference

<div class="api-model-section api-model-static api-config-section">
  <div class="api-model-header"><span class="api-model-badge api-config-badge">Config</span><span class="api-model-name">PositionWeightGeneratorConfig</span><a href="https://github.com/evo-design/proto-language/blob/d3b7822f74ea64747cc751a3b2ab1aa6b799ac47/proto_language/generator/position_weight_generator.py#L16" target="_blank" class="func-table-btn func-source-btn api-model-source"><svg width="12" height="12" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><polyline points="16 18 22 12 16 6" /><polyline points="8 6 2 12 8 18" /></svg> Source</a></div>

  Configuration for position-specific sequence sampling.

  This generator is intended for optimizers that own position-specific
  sequence distributions and need to materialize discrete proposal sequences
  for the rest of the framework.

  <ParamField path="sampling_mode" type="enum" default="argmax">
    How to convert position-specific distributions into discrete proposals.

    Options: `argmax`, `categorical`
  </ParamField>

  <ParamField path="temperature" type="number" default="1.0">
    Softmax temperature on logits before decoding. Below 1 sharpens; above 1 flattens. Must be > 0.
  </ParamField>

  <ParamField path="sequence_bias" type="SequenceLogitBiasConfig">
    Optional declarative sequence-symbol bias applied before decoding.
  </ParamField>

  <ParamField path="logit_scale" type="number" default="1.0">
    Optional scale factor applied to logits before adding sequence\_bias and decoding.
  </ParamField>

  <ParamField path="peak_probability_positions" type="array">
    Position indices used in the mean per-position peak-probability metric; None averages over all.
  </ParamField>
</div>

## Usage

```python python icon="python" theme={null}
>>> segment = Segment(sequence="ACGT", sequence_type="dna")
>>> gen = PositionWeightGenerator(PositionWeightGeneratorConfig(sampling_mode="argmax"))
>>> gen.assign(segment)
>>> segment.proposal_sequences[0].logits = np.array([[5, 0, 0, 0], [0, 4, 0, 0], [0, 0, 3, 0], [0, 0, 0, 2]])
>>> gen.sample()
>>> segment.proposal_sequences[0].sequence
'ACGT'
```

## Metadata

| Property                 | Value                     |
| ------------------------ | ------------------------- |
| Key                      | `position-weight`         |
| Class                    | `PositionWeightGenerator` |
| Category                 | `gradient`                |
| Input Type               | `logits`                  |
| Uses GPU                 | `False`                   |
| Supported Sequence Types | `dna`, `rna`, `protein`   |
| Allows Empty Start       | `False`                   |