- Ready-to-use descriptors that wrap a specific model,
- A general interface to call other suitable models you select.
- You know how to use descriptors to evaluate text data.
Imports
Toy data to run the example
Toy data to run the example
To generate toy data and create a Dataset object:
Built-in ML evals
Available descriptors. Check all available built-in LLM evals in the reference table.
Custom ML evals
You can also add any custom checks directly as a Python function.
HuggingFace()
descriptor to call a specific named model. The model you use must return a numerical score or a category for each text in a column.
For example, to evaluate “curiousity” expressed in a text:

Sample models
Here are some models you can call using theHuggingFace()
descriptor.
Model | Example use | Parameters |
---|---|---|
Emotion classification
| HuggingFace("response", model="SamLowe/roberta-base-go_emotions", params={"label": "disappointment"}, alias="disappointment") | Required:
|
Zero-shot classification
| HuggingFace("response", model="MoritzLaurer/DeBERTa-v3-large-mnli-fever-anli-ling-wanli", params={"labels": ["science", "physics"], "threshold":0.5}, alias="Topic") | Required:
|
GPT-2 text detection
| HuggingFace("response", model="openai-community/roberta-base-openai-detector", params={"score_threshold": 0.7}, alias="fake") | Optional:
|
- Output a single number (e.g., predicted score for a label) or a label, not an array of values.
- Can process raw text input directly.
-
Name labels using
label
orlabels
fields. -
Use methods named
predict
orpredict_proba
for scoring.