Downloads · 30 days
3
3% of all-time downloads
dshirls/food-truck-moderation-v1
food-truck-moderation-v1 is a text classification model from dshirls. Use it when you need a label for a piece of text. It is set up for transformers. The card lists the license as mit.
Binary text classifier for a food-truck game: allowed vs blocked user dish descriptions.
Downloads · 30 days
3
3% of all-time downloads
All-time downloads
99
Public
Parameters
67M
1.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors268 MB · 100%
From the Hugging Face model README
Binary text classifier for a food-truck game: allowed vs blocked user dish descriptions.
Fine-tuned from distilbert-base-uncased on game-specific examples (gross food humour allowed; slurs/threats blocked).
| Label | Meaning |
|---|---|
allowed | Silly/gross food humour (e.g. cockroach salad) |
blocked | Profanity, hate, threats, sexual content |
Local / app (recommended for custom model):
TEXT_MODERATION_PROVIDER=local-model # uses ml/text-moderation/output/model
Dedicated Inference Endpoint (for Heroku / cloud):
HUGGINGFACE_INFERENCE_ENDPOINT to the endpoint URL in web/.envThe shared
hf-inferencerouter does not host custom Hub models — only a curated catalog.
from transformers import pipeline
clf = pipeline("text-classification", model="dshirls/food-truck-moderation-v1")
clf("cockroach salad") # allowed
clf("fuck you cake") # blocked