CIC Check-In Relevance Classifier
A BERT-based text classification model that categorizes daily check-ins by relevance to the Coding in Color (CIC) / myVillage project ecosystem.
Labels
| Label | Description |
|---|---|
relevant |
Directly relates to CIC/myVillage work: client calls, myVillage OS, student platforms, Director workflows, contract management, lessons |
not_relevant |
Personal projects, schoolwork, unrelated side projects, external courses |
vague |
Mentions tools/workflows that could be CIC-related but lacks specificity |
Usage
from transformers import pipeline
classifier = pipeline('text-classification', model='Dc-4nderson/cic-checkin-relevance-classifier2')
result = classifier('Today I met with the JCC team about workflow integration.')
print(result)
# [{'label': 'relevant', 'score': 0.95}]
Training
- Base model:
bert-base-uncased - Training samples: ~366 labeled check-ins
- Weighted cross-entropy loss for class imbalance
- Early stopping on validation F1
- Downloads last month
- 29