update model card README.md

README.md

@@ -17,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [moussaKam/barthez-orangesum-abstract](https://huggingface.co/moussaKam/barthez-orangesum-abstract) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 7.5955
-- Rouge1: 0.7166
-- Rouge2: 0.7001
-- Rougel: 0.7163
-- Rougelsum: 0.7161
+- Loss: 0.1792
+- Rouge1: 0.6715
+- Rouge2: 0.6537
+- Rougel: 0.6717
+- Rougelsum: 0.6717
 - Gen Len: 20.0
 
 ## Model description
@@ -42,22 +42,24 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 4
-- eval_batch_size: 4
+- train_batch_size: 16
+- eval_batch_size: 32
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 8
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
-| No log        | 0.96  | 12   | 9.5195          | 0.7166 | 0.7001 | 0.7163 | 0.7161    | 20.0    |
-| No log        | 2.0   | 25   | 8.0721          | 0.7166 | 0.7001 | 0.7163 | 0.7161    | 20.0    |
-| No log        | 2.88  | 36   | 7.5955          | 0.7166 | 0.7001 | 0.7163 | 0.7161    | 20.0    |
+| No log        | 1.0   | 31   | 0.2054          | 0.6716 | 0.6537 | 0.6717 | 0.6717    | 20.0    |
+| No log        | 1.99  | 62   | 0.1884          | 0.6715 | 0.6537 | 0.6717 | 0.6717    | 20.0    |
+| No log        | 2.99  | 93   | 0.1824          | 0.6715 | 0.6537 | 0.6717 | 0.6717    | 20.0    |
+| No log        | 3.98  | 124  | 0.1798          | 0.6715 | 0.6537 | 0.6717 | 0.6717    | 20.0    |
+| No log        | 4.98  | 155  | 0.1792          | 0.6715 | 0.6537 | 0.6717 | 0.6717    | 20.0    |
 
 
 ### Framework versions