Downloads · 30 days
14
30% of all-time downloads
muhtasham/web-scale-ner
web-scale-ner is a token classification model from muhtasham. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as apache-2.0.
Downloads · 30 days
14
30% of all-time downloads
All-time downloads
46
Public
Parameters
148M
1.8 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors592 MB · 99%
From the Hugging Face model README
Training and Evaluation Metrics
| Metric | Value |
|---|---|
| Evaluation Recall | 0.8099 |
| Training Runtime | 3761.66 seconds |
| Evaluation Steps/Sec | 0.304 |
| Total FLOPs | 55,842,753,586,544,640 |
| Training Loss | 0.3471 |
| Training Loss (Epoch) | 0.7509 |
| Training Epochs | 5 |
| Evaluation Accuracy | 0.8099 |
| Gradient Norm | 5.0118 |
| Learning Rate | 1.88e-8 |
| Evaluation Samples/Sec | 23.205 |
| Evaluation F1 Score | 0.8132 |
| Wandb Runtime | 3822 seconds |
| Runtime | 3821.17 seconds |
| Evaluation Loss | 0.5319 |
| Evaluation Runtime | 46.067 seconds |
| Evaluation Precision | 0.8206 |
| Global Training Steps | 270 |
| Training Steps/Sec | 0.072 |
| Training Samples/Sec | 5.681 |