---
language: en
license: mit
pipeline_tag: text-classification
tags:
- sarcasm
- text-classification
- bert
- custom-code
---

# loyongzhe/sarcasm-classifier-binary

Binary sarcasm classifier.

## Architecture
- Encoder: `bert-base-uncased`
- Pooling: CLS
- Head: MLP hidden size 256
- Labels:
  - 0 = not_sarcastic
  - 1 = sarcastic

## Load
```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification

repo_id = "loyongzhe/sarcasm-classifier-binary"

tokenizer = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
model = AutoModelForSequenceClassification.from_pretrained(
    repo_id,
    trust_remote_code=True,
)
```