Model Parameters
Model selection and inference parameters for generative AI models.
Properties
The ModelParameters type has these properties:
| Name | Description | Type | Inherited from | JSON-LD @id |
Aliases |
|---|---|---|---|---|---|
id |
The identifier for this item. | String |
Entity |
schema:id |
- |
modelIds |
The ids of the models to select. | String* |
- | stencila:modelIds |
models, model, model-ids, model_ids, modelId, model-id, model_id |
replicates |
The number of replicate inferences to run per model id. | UnsignedInteger |
- | stencila:replicates |
reps |
qualityWeight |
The relative weighting given to model quality (0-100). | UnsignedInteger |
- | stencila:qualityWeight |
quality, qual, quality-weight, quality_weight |
costWeight |
The relative weighting given to model cost (0-100). | UnsignedInteger |
- | stencila:costWeight |
cost, cost-weight, cost_weight |
speedWeight |
The relative weighting given to model speed (0-100). | UnsignedInteger |
- | stencila:speedWeight |
speed, speed-weight, speed_weight |
minimumScore |
The minimum score for models to be selected (0-100). | UnsignedInteger |
- | stencila:minimumScore |
minimum-score, minimum_score, minScore, min-score, min_score |
temperature |
The temperature option for model inference (0-100). | UnsignedInteger |
- | stencila:temperature |
temp |
randomSeed |
The random seed used for the model (if possible) | Integer |
- | stencila:randomSeed |
random-seed, random_seed, rand-seed, rand_seed, seed |
executeContent |
Automatically execute generated content. | Boolean |
- | stencila:executeContent |
execute-content, execute_content |
executionBounds |
The environment in which code should be executed. | ExecutionBounds |
- | stencila:executionBounds |
execution-bounds, execution_bounds |
maximumRetries |
When executing content, the maximum number of retries. | UnsignedInteger |
- | stencila:maximumRetries |
retries, maximum-retries, maximum_retries |
Related
The ModelParameters type is related to these types:
- Parents:
Entity - Children: none
Formats
The ModelParameters type can be encoded (serialized) to, and/or decoded (deserialized) from, these formats:
| Format | Encoding | Decoding | Support | Notes |
|---|---|---|---|---|
| DOM HTML | 🟢 No loss | |||
| HTML | 🔷 Low loss | |||
| JATS | ||||
| Markdown | ⚠️ High loss | Encoded using implemented function | ||
| Stencila Markdown | ⚠️ High loss | |||
| Quarto Markdown | ⚠️ High loss | |||
| MyST Markdown | ⚠️ High loss | |||
| LLM Markdown | ⚠️ High loss | |||
| LaTeX | 🔷 Low loss | 🔷 Low loss | ||
| 🔷 Low loss | ||||
| Plain text | ⚠️ High loss | |||
| IPYNB | 🔷 Low loss | 🔷 Low loss | ||
| Microsoft Word DOCX | 🔷 Low loss | 🔷 Low loss | ||
| OpenDocument ODT | 🔷 Low loss | 🔷 Low loss | ||
| TeX | 🔷 Low loss | 🔷 Low loss | ||
| JSON | 🟢 No loss | 🟢 No loss | ||
| JSON+Zip | 🟢 No loss | 🟢 No loss | ||
| JSON5 | 🟢 No loss | 🟢 No loss | ||
| JSON-LD | 🟢 No loss | 🟢 No loss | ||
| CBOR | 🟢 No loss | 🟢 No loss | ||
| CBOR+Zstandard | 🟢 No loss | 🟢 No loss | ||
| YAML | 🟢 No loss | 🟢 No loss | ||
| Lexical JSON | 🔷 Low loss | 🔷 Low loss | ||
| Koenig JSON | 🔷 Low loss | 🔷 Low loss | ||
| Pandoc AST | 🔷 Low loss | 🔷 Low loss | ||
| Directory | ||||
| Stencila Web Bundle | ||||
| Debug | 🔷 Low loss |
Bindings
The ModelParameters type is represented in:
- JSON-LD
- JSON Schema
- Python class
ModelParameters - Rust struct
ModelParameters - TypeScript class
ModelParameters
Source
This documentation was generated from ModelParameters.yaml by docs_types.rs.