End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [nlpie/bio-mobilebert](https://huggingface.co/nlpie/bio-mobilebert) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1613
-- Accuracy: 0.742
+- Loss: 1.5244
+- Accuracy: 0.737
 
 ## Model description
 
@@ -49,15 +49,15 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Accuracy |
 |:-------------:|:------:|:----:|:---------------:|:--------:|
-| No log        | 0.3121 | 250  | 0.8770          | 0.737    |
-| 6.4745        | 0.6242 | 500  | 0.8535          | 0.731    |
-| 6.4745        | 0.9363 | 750  | 0.8849          | 0.729    |
-| 0.7885        | 1.2484 | 1000 | 0.8966          | 0.734    |
-| 0.7885        | 1.5605 | 1250 | 0.8883          | 0.723    |
-| 0.6817        | 1.8727 | 1500 | 1.0612          | 0.729    |
-| 0.6817        | 2.1848 | 1750 | 0.9602          | 0.739    |
-| 0.5535        | 2.4969 | 2000 | 1.5924          | 0.74     |
-| 0.5535        | 2.8090 | 2250 | 1.1613          | 0.742    |
+| No log        | 0.3121 | 250  | 0.8963          | 0.718    |
+| 0.8374        | 0.6242 | 500  | 0.6912          | 0.74     |
+| 0.8374        | 0.9363 | 750  | 0.7389          | 0.747    |
+| 0.6729        | 1.2484 | 1000 | 0.8110          | 0.737    |
+| 0.6729        | 1.5605 | 1250 | 1.0646          | 0.733    |
+| 0.6059        | 1.8727 | 1500 | 0.8516          | 0.733    |
+| 0.6059        | 2.1848 | 1750 | 1.2728          | 0.734    |
+| 0.4538        | 2.4969 | 2000 | 1.5648          | 0.731    |
+| 0.4538        | 2.8090 | 2250 | 1.5244          | 0.737    |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b271d8197d4ceed9dc6e19a806e4d40683e1ba8942eccbb3ed8675245204ac80
+oid sha256:21e54704242763fb3677d262f90c491b814b0d414b34f71f73677066415f4d69
 size 98468060