mtl-dev/allergy_check_bert_turk_v1

README.md

@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [dbmdz/bert-base-turkish-cased](https://huggingface.co/dbmdz/bert-base-turkish-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3577
+- Loss: 0.3560
 - Accuracy: 0.8955
 
 ## Model description
@@ -44,18 +44,16 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 5
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.7828        | 1.0   | 135  | 0.3802          | 0.8843   |
-| 0.3478        | 2.0   | 270  | 0.3407          | 0.8955   |
-| 0.2755        | 3.0   | 405  | 0.3109          | 0.8993   |
-| 0.191         | 4.0   | 540  | 0.3558          | 0.8955   |
-| 0.1216        | 5.0   | 675  | 0.3577          | 0.8955   |
+| 0.9326        | 1.0   | 135  | 0.4260          | 0.8694   |
+| 0.3599        | 2.0   | 270  | 0.3860          | 0.8769   |
+| 0.2744        | 3.0   | 405  | 0.3560          | 0.8955   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71cedaf674644e1581bfe0df271d9e4c636754bc2eb4afb07bbca0cc9a67a9b9
+oid sha256:c7b361de2ac4993b923bcf16856e0185bb01f752735cf570e3b8e1a83e6307ba
 size 442511368

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20def1d805cb67bf04e0270b90e45594349dd3f3096649d460bc6ce42aab0d2a
+oid sha256:0a8c14647db7e89ab5eef0deb832fa7eaae28ad6ad943890b56d33453a913595
 size 5368