Model save

README.md

@@ -25,16 +25,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.8540680154972019
+      value: 0.8579982891360137
     - name: Recall
       type: recall
-      value: 0.8759381898454747
+      value: 0.8856512141280353
     - name: F1
       type: f1
-      value: 0.8648648648648649
+      value: 0.8716054746904193
     - name: Accuracy
       type: accuracy
-      value: 0.9496757457846952
+      value: 0.9511284046692607
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -44,11 +44,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [ufal/robeczech-base](https://huggingface.co/ufal/robeczech-base) on the cnec dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2682
-- Precision: 0.8541
-- Recall: 0.8759
-- F1: 0.8649
-- Accuracy: 0.9497
+- Loss: 0.3233
+- Precision: 0.8580
+- Recall: 0.8857
+- F1: 0.8716
+- Accuracy: 0.9511
 
 ## Model description
 
@@ -68,8 +68,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 32
+- 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
@@ -77,12 +77,16 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.5187        | 6.8   | 1000 | 0.3863          | 0.7882    | 0.8115 | 0.7997 | 0.9328   |
-| 0.222         | 13.61 | 2000 | 0.2829          | 0.8376    | 0.8561 | 0.8467 | 0.9463   |
-| 0.1408        | 20.41 | 3000 | 0.2662          | 0.8493    | 0.8684 | 0.8588 | 0.9493   |
-| 0.1071        | 27.21 | 4000 | 0.2682          | 0.8541    | 0.8759 | 0.8649 | 0.9497   |
+| Training Loss | Epoch | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.3724        | 3.41  | 2000  | 0.3332          | 0.7990    | 0.8230 | 0.8108 | 0.9376   |
+| 0.1863        | 6.81  | 4000  | 0.2656          | 0.8515    | 0.8636 | 0.8575 | 0.9455   |
+| 0.1109        | 10.22 | 6000  | 0.2575          | 0.8505    | 0.8737 | 0.8619 | 0.9493   |
+| 0.068         | 13.63 | 8000  | 0.2804          | 0.8567    | 0.8790 | 0.8677 | 0.9503   |
+| 0.0466        | 17.04 | 10000 | 0.2952          | 0.8573    | 0.8830 | 0.8699 | 0.9498   |
+| 0.0305        | 20.44 | 12000 | 0.2992          | 0.8618    | 0.8865 | 0.8740 | 0.9520   |
+| 0.0231        | 23.85 | 14000 | 0.3272          | 0.8567    | 0.8843 | 0.8703 | 0.9512   |
+| 0.02          | 27.26 | 16000 | 0.3233          | 0.8580    | 0.8857 | 0.8716 | 0.9511   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c851788559dbc037486d2e29f4c08690b6d531fcbb93a20539c6f3f8d5829d0
+oid sha256:0ea4e50429af3b289eb29760615425f9efbaf6f6123bc2808fa822d50cc26f62
 size 501826260