en_textcat_gender

Description

Profile text data classification using SpaCy textcat components. A super tiny model with no vectors. Internally uses an ensemle of BOW and single-headed attention. I've got better performance scores with transformer-based models but they are much slower to train/run on CPU and much larger. Sticking to the basics so far.

The model is trained on specifically formatted profile fields [name] ([login], [email]) in [location] so it's critical to format the data with the provided format_input helper before applying a model to the text.

Usage

from en_textcat_gender import load as load_gender_model
from en_textcat_gender.utils import format_input

gender_model = load_gender_model()

class PredictGenderByProfileInput(BaseModel):
  login: str
  name: str
  location: str | None = None
  email: str | None = None

type Gender = Literal["Male", "Female", "Neutral"]

# Example of model use in the context of FastAPI route
@router.post("/predictGenderByProfile")
async def predictGenderByProfile(body: list[PredictGenderByProfileInput]) -> list[Gender]:
  texts = [format_input({
    "login": input.login,
    "name": input.name,
    "location": input.location,
    "email": input.email,
  }) for input in body]
  docs = gender_model.pipe(texts)
  genders: list[GenderAlt] = []
  for doc in docs:
    match max(doc.cats, key=doc.cats.get):
      case "FEMALE": genders.append("Female")
      case "MALE": genders.append("Male")
      case "NEUTRAL": genders.append("Neutral")
      case _: raise HTTPException(500, "Invalid enum value")
  return genders

Label Scheme

• MALE is a prediction of male human profiles
• FEMALE is a prediction for female human profiles
• NEUTRAL is a prediction for gender-neutral names, unintelligible sequences ("Foo Bar") and non-human profiles (organizations, companies)

Accuracy