Alfahluzi/bert2bert-dropout-0.3-lr-5e-05-ds-canonical add 3 more epoch (total 6)

README.md

@@ -15,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on the id_liputan6 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.2037
-- Rouge2 Precision: 0.1783
-- Rouge2 Recall: 0.1901
-- Rouge2 Fmeasure: 0.1818
+- Loss: 2.2773
+- Rouge2 Precision: 0.1744
+- Rouge2 Recall: 0.187
+- Rouge2 Fmeasure: 0.1784
 
 ## Model description
 
@@ -43,16 +43,14 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 1
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
-|:-------------:|:-----:|:-----:|:---------------:|:----------------:|:-------------:|:---------------:|
-| 1.7398        | 1.0   | 4040  | 2.2892          | 0.1789           | 0.1933        | 0.1835          |
-| 1.5953        | 2.0   | 8080  | 2.2380          | 0.1789           | 0.1917        | 0.1829          |
-| 1.5771        | 3.0   | 12120 | 2.2037          | 0.1783           | 0.1901        | 0.1818          |
+| Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
+|:-------------:|:-----:|:----:|:---------------:|:----------------:|:-------------:|:---------------:|
+| 1.3309        | 1.0   | 4040 | 2.2773          | 0.1744           | 0.187         | 0.1784          |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba1131f5f2de4b02cfd9a13bb532ff08829fa2418e79d91f8f9a2f0039c2f869
+oid sha256:5a01d0fafcaf021b08871c150ab052b9bd905f1fcb7c5547fd9633f060daebe8
 size 998132132