Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,12 @@ The model has been quantized to 8 bits to:
|
|
| 18 |
|
| 19 |
Below is an example script demonstrating how to load the 8-bit quantized model, perform translation, and decode the output:
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
```python
|
| 22 |
import torch
|
| 23 |
import transformers
|
|
|
|
| 18 |
|
| 19 |
Below is an example script demonstrating how to load the 8-bit quantized model, perform translation, and decode the output:
|
| 20 |
|
| 21 |
+
make sure to install bits and bytes
|
| 22 |
+
|
| 23 |
+
```
|
| 24 |
+
pip install -U bitsandbytes
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
```python
|
| 28 |
import torch
|
| 29 |
import transformers
|