Training complete

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7135
-- Sacrebleu: 0.0000
+- Loss: 0.2966
+- Sacrebleu: 0.0
 
 ## Model description
 
@@ -39,21 +39,23 @@ 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: 8
+- eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 16
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 2
+- num_epochs: 4
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
-|:-------------:|:-----:|:----:|:---------------:|:---------:|
-| No log        | 0.992 | 62   | 0.9029          | 0.0000    |
-| No log        | 1.984 | 124  | 0.7135          | 0.0000    |
+| Training Loss | Epoch  | Step | Validation Loss | Sacrebleu |
+|:-------------:|:------:|:----:|:---------------:|:---------:|
+| No log        | 0.992  | 93   | 0.4620          | 0.0       |
+| No log        | 1.9947 | 187  | 0.3466          | 0.0       |
+| No log        | 2.9973 | 281  | 0.3070          | 0.0       |
+| No log        | 3.968  | 372  | 0.2966          | 0.0       |
 
 
 ### Framework versions