End of training

README.md

@@ -19,10 +19,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [dumitrescustefan/bert-base-romanian-cased-v1](https://huggingface.co/dumitrescustefan/bert-base-romanian-cased-v1) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2801
-- F1: 0.5082
-- Roc Auc: 0.6987
-- Accuracy: 0.4107
+- Loss: 0.2460
+- F1: 0.6924
+- Roc Auc: 0.8005
+- Accuracy: 0.5967
 
 ## Model description
 
@@ -41,25 +41,28 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
+- learning_rate: 2e-05
 - train_batch_size: 128
-- eval_batch_size: 16
+- eval_batch_size: 128
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 7
+- num_epochs: 10
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | F1     | Roc Auc | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
-| No log        | 1.0   | 32   | 0.4216          | 0.0    | 0.5     | 0.0      |
-| No log        | 2.0   | 64   | 0.3749          | 0.0770 | 0.5273  | 0.0460   |
-| No log        | 3.0   | 96   | 0.3358          | 0.2623 | 0.5907  | 0.1768   |
-| No log        | 4.0   | 128  | 0.3070          | 0.4103 | 0.6528  | 0.3131   |
-| No log        | 5.0   | 160  | 0.2903          | 0.4629 | 0.6811  | 0.3738   |
-| No log        | 6.0   | 192  | 0.2831          | 0.4937 | 0.6923  | 0.3959   |
-| No log        | 7.0   | 224  | 0.2801          | 0.5082 | 0.6987  | 0.4107   |
+| No log        | 1.0   | 32   | 0.3874          | 0.0317 | 0.5090  | 0.0147   |
+| No log        | 2.0   | 64   | 0.3179          | 0.3614 | 0.6333  | 0.2597   |
+| No log        | 3.0   | 96   | 0.2706          | 0.5534 | 0.7215  | 0.4512   |
+| No log        | 4.0   | 128  | 0.2541          | 0.6257 | 0.7549  | 0.5267   |
+| No log        | 5.0   | 160  | 0.2448          | 0.6575 | 0.7748  | 0.5562   |
+| No log        | 6.0   | 192  | 0.2433          | 0.6745 | 0.7835  | 0.5746   |
+| No log        | 7.0   | 224  | 0.2451          | 0.6837 | 0.7963  | 0.5875   |
+| No log        | 8.0   | 256  | 0.2451          | 0.6752 | 0.7885  | 0.5875   |
+| No log        | 9.0   | 288  | 0.2452          | 0.6845 | 0.7939  | 0.5930   |
+| No log        | 10.0  | 320  | 0.2460          | 0.6924 | 0.8005  | 0.5967   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb51da53119c46a9049ac0f915a2d984d2cd7c2981926d3da4feb7c1982eb631
+oid sha256:3dccce92fd17b40e49f5a362e20e5513e0608b1daa66d08a7926987706a46518
 size 497810452

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:751c6fb1a59fbdd459f4122bed43cd95147b96aa50414c337a33edbcb35814d3
+oid sha256:e2849ca0c8080b495a0bda4f7456cbb0829ef61be2a1402831d02ac059d9420b
 size 5176