Model save

README.md

@@ -24,16 +24,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.8065730914070524
+      value: 0.8169398907103825
     - name: Recall
       type: recall
-      value: 0.8426323319027181
+      value: 0.8555078683834049
     - name: F1
       type: f1
-      value: 0.8242085009620429
+      value: 0.835779175401817
     - name: Accuracy
       type: accuracy
-      value: 0.9442379182156134
+      value: 0.9460251644266514
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,11 +43,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [UWB-AIR/Czert-B-base-cased](https://huggingface.co/UWB-AIR/Czert-B-base-cased) on the cnec dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2748
-- Precision: 0.8066
-- Recall: 0.8426
-- F1: 0.8242
-- Accuracy: 0.9442
+- Loss: 0.2873
+- Precision: 0.8169
+- Recall: 0.8555
+- F1: 0.8358
+- Accuracy: 0.9460
 
 ## Model description
 
@@ -66,9 +66,9 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 4
-- eval_batch_size: 4
+- learning_rate: 2e-05
+- train_batch_size: 2
+- eval_batch_size: 2
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -76,13 +76,13 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.3028        | 1.0   | 1799 | 0.2639          | 0.7483    | 0.7965 | 0.7717 | 0.9303   |
-| 0.2143        | 2.0   | 3598 | 0.2481          | 0.7808    | 0.8140 | 0.7971 | 0.9368   |
-| 0.127         | 3.0   | 5397 | 0.2458          | 0.8030    | 0.8237 | 0.8132 | 0.9399   |
-| 0.0797        | 4.0   | 7196 | 0.2606          | 0.8088    | 0.8487 | 0.8283 | 0.9444   |
-| 0.0456        | 5.0   | 8995 | 0.2748          | 0.8066    | 0.8426 | 0.8242 | 0.9442   |
+| Training Loss | Epoch | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.2749        | 1.0   | 3597  | 0.2725          | 0.7744    | 0.7969 | 0.7855 | 0.9313   |
+| 0.2085        | 2.0   | 7194  | 0.2446          | 0.7877    | 0.8308 | 0.8087 | 0.9397   |
+| 0.1161        | 3.0   | 10791 | 0.2582          | 0.8029    | 0.8405 | 0.8212 | 0.9428   |
+| 0.0932        | 4.0   | 14388 | 0.2836          | 0.8128    | 0.8526 | 0.8323 | 0.9448   |
+| 0.0625        | 5.0   | 17985 | 0.2873          | 0.8169    | 0.8555 | 0.8358 | 0.9460   |
 
 
 ### Framework versions