Add link to Neuron-optimized version
#13
by
badaoui
HF Staff
- opened
README.md
CHANGED
|
@@ -232,3 +232,15 @@ Glue test results:
|
|
| 232 |
<a href="https://huggingface.co/exbert/?model=roberta-base">
|
| 233 |
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
|
| 234 |
</a>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 232 |
<a href="https://huggingface.co/exbert/?model=roberta-base">
|
| 233 |
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
|
| 234 |
</a>
|
| 235 |
+
|
| 236 |
+
---
|
| 237 |
+
## ๐ AWS Neuron Optimized Version Available
|
| 238 |
+
|
| 239 |
+
A Neuron-optimized version of this model is available for improved performance on AWS Inferentia/Trainium instances:
|
| 240 |
+
|
| 241 |
+
**[badaoui/FacebookAI-roberta-base-neuron](https://huggingface.co/badaoui/FacebookAI-roberta-base-neuron)**
|
| 242 |
+
|
| 243 |
+
The Neuron-optimized version provides:
|
| 244 |
+
- Pre-compiled artifacts for faster loading
|
| 245 |
+
- Optimized performance on AWS Neuron devices
|
| 246 |
+
- Same model capabilities with improved inference speed
|