README.md

metadata

language:
  - en
license: apache-2.0
library_name: transformers

English Text Classifier

This repository contains a runnable sequence-classification model for English text.

Files Needed for Inference

• config.json
• model.safetensors
• tokenizer.json
• tokenizer_config.json
• special_tokens_map.json

Minimal Usage

import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer

model_path = "."
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForSequenceClassification.from_pretrained(model_path)
model.eval()

texts = [
    "I disagree with this argument, but the evidence is interesting.",
    "You are awful and should leave.",
]

inputs = tokenizer(
    texts,
    padding=True,
    truncation=True,
    return_tensors="pt",
)

with torch.no_grad():
    logits = model(**inputs).logits
    scores = torch.sigmoid(logits)

print(scores.tolist())

The model returns a 14-dimensional score vector for each input text.