update readme
Browse files
README.md
CHANGED
@@ -19,6 +19,7 @@ This model is a fine-tuned version of [google/mobilebert-uncased](https://huggin
|
|
19 |
Notice that this model only has the first 14 transformer blocks. It is quantized and structually pruned by NNCF. The sparsity in remaining linear layers is 39.8%.
|
20 |
|
21 |
- Torch f1: 90.15
|
|
|
22 |
|
23 |
### Framework versions
|
24 |
|
|
|
19 |
Notice that this model only has the first 14 transformer blocks. It is quantized and structually pruned by NNCF. The sparsity in remaining linear layers is 39.8%.
|
20 |
|
21 |
- Torch f1: 90.15
|
22 |
+
- IR f1: 89.8414
|
23 |
|
24 |
### Framework versions
|
25 |
|