Text Classification
Transformers
Safetensors
English
Chinese
bert
nil
hover
reject
nli
mbert
text-embeddings-inference
Instructions to use aarontseng/nil-hover-reject-mbert with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use aarontseng/nil-hover-reject-mbert with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="aarontseng/nil-hover-reject-mbert")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("aarontseng/nil-hover-reject-mbert") model = AutoModelForSequenceClassification.from_pretrained("aarontseng/nil-hover-reject-mbert", device_map="auto") - Notebooks
- Google Colab
- Kaggle
nil-hover-reject-mbert
Fine-tuned mBERT for hover reject-unsuitable (EN↔ZH).
- Not the pick-best model (
aarontseng/nil-hover-mbert). - Training data:
aarontseng/nil-hover-reject-cmn-hani - Checkpoint:
checkpoint-16000(bestitem_accuracy/ set_acc = 0.2746) - Early-stop metric: exact reject-set match (
item_accuracy) - Inference: reject when
contradiction > entailment(logits); keep the rest.
- Downloads last month
- 46