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 [Twitter/twhin-bert-base](https://huggingface.co/Twitter/twhin-bert-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9505
-- F1: 0.4
+- Loss: 0.5691
+- F1: 0.3019
 
 ## Model description
 
@@ -37,28 +37,38 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-06
+- learning_rate: 0.0002
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 20
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 0.6761        | 1.0   | 189  | 0.6155          | 0.0    |
-| 0.6234        | 2.0   | 378  | 0.6077          | 0.1961 |
-| 0.612         | 3.0   | 567  | 0.6394          | 0.2333 |
-| 0.611         | 4.0   | 756  | 0.6725          | 0.4533 |
-| 0.59          | 5.0   | 945  | 0.6929          | 0.4198 |
-| 0.5866        | 6.0   | 1134 | 0.9161          | 0.3385 |
-| 0.628         | 7.0   | 1323 | 0.8932          | 0.3824 |
-| 0.4787        | 8.0   | 1512 | 0.9278          | 0.3824 |
-| 0.5853        | 9.0   | 1701 | 0.9531          | 0.3944 |
-| 0.5249        | 10.0  | 1890 | 0.9505          | 0.4    |
+| 0.6827        | 1.0   | 189  | 0.6385          | 0.0    |
+| 0.6734        | 2.0   | 378  | 0.6212          | 0.0    |
+| 0.6462        | 3.0   | 567  | 0.6068          | 0.0    |
+| 0.6299        | 4.0   | 756  | 0.6033          | 0.0    |
+| 0.6433        | 5.0   | 945  | 0.6148          | 0.2903 |
+| 0.6516        | 6.0   | 1134 | 0.5877          | 0.0    |
+| 0.6256        | 7.0   | 1323 | 0.5976          | 0.2881 |
+| 0.6344        | 8.0   | 1512 | 0.5870          | 0.3366 |
+| 0.6373        | 9.0   | 1701 | 0.5833          | 0.3137 |
+| 0.6375        | 10.0  | 1890 | 0.5780          | 0.3368 |
+| 0.6439        | 11.0  | 2079 | 0.5771          | 0.3168 |
+| 0.6235        | 12.0  | 2268 | 0.5703          | 0.3182 |
+| 0.604         | 13.0  | 2457 | 0.5854          | 0.304  |
+| 0.6432        | 14.0  | 2646 | 0.5697          | 0.3061 |
+| 0.6009        | 15.0  | 2835 | 0.5701          | 0.3048 |
+| 0.6118        | 16.0  | 3024 | 0.5736          | 0.2957 |
+| 0.6207        | 17.0  | 3213 | 0.5733          | 0.2957 |
+| 0.6357        | 18.0  | 3402 | 0.5701          | 0.2991 |
+| 0.621         | 19.0  | 3591 | 0.5689          | 0.3019 |
+| 0.6235        | 20.0  | 3780 | 0.5691          | 0.3019 |
 
 
 ### Framework versions

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e47e77a95e463f3fecce97e0e427a0d0cc4ee16e2d56cfec0d8c7e6591232c10
+oid sha256:723633affb4d2f4f692efcf15e03085813ce0f4d5f79f2c001b1590a362161ba
 size 1115393438

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96628b7704b1f281a45df5d9feb0ef79e2c058130bf55a20bedd465146416746
-size 17082815
+oid sha256:f2c509a525eb51aebb33fb59c24ee923c1d4c1db23c3ae81fe05ccf354084f7b
+size 17082758

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cef82a51722ba12abdcad65c57e267b1439c8577b38065cfb732fb608c0493f9
-size 4536
+oid sha256:3de5528471e131f424915bd06548ac5d69114bc8df74648e5d6dbadaf9f41727
+size 4600