Downloads · 30 days
8
13% of all-time downloads
martian786/agnews-roberta-seed42-fulltext-10pct
agnews-roberta-seed42-fulltext-10pct is a text classification model from martian786. Use it when you need a label for a piece of text. It is set up for transformers. The card lists the license as apache-2.0.
This model is a fine-tuned roberta-base sequence classifier for the AG News four-class classification task.
Downloads · 30 days
8
13% of all-time downloads
All-time downloads
63
Public
Parameters
125M
499 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors499 MB · 99%
From the Hugging Face model README
This model is a fine-tuned roberta-base sequence classifier for the AG News four-class classification task.
roberta-base| Label ID | Label |
|---|---|
| 0 | World |
| 1 | Sports |
| 2 | Business |
| 3 | Sci/Tech |
| Metric | Value |
|---|---|
| Test accuracy | 0.931316 |
| Test macro F1 | 0.931301 |
| Accuracy 95% Wilson CI | 0.925409–0.936787 |
| Misclassified examples | 522 |
| Single-example latency mean | 17.150 ms |
| Training time | 450.40 s |
| Class | F1 |
|---|---|
| World | 0.940990 |
| Sports | 0.982105 |
| Business | 0.894977 |
| Sci/Tech | 0.907132 |
repo_id = "martian786/agnews-roberta-seed42-fulltext-10pct"
Use AutoTokenizer.from_pretrained(repo_id) and AutoModelForSequenceClassification.from_pretrained(repo_id).
This model was fine-tuned on AG News and is intended for controlled academic comparison rather than production deployment.