End of training

README.md

@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0832
-- Bleu: 0.952
-- Gen Len: 17.9397
+- Loss: nan
+- Bleu: 0.6661
+- Gen Len: 17.2141
 
 ## Model description
 
@@ -45,15 +45,15 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 10
-- num_epochs: 3
+- num_epochs: 15
+- mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step  | Validation Loss | Bleu   | Gen Len |
 |:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|
-| 1.3886        | 1.0   | 12889 | 1.2435          | 0.7737 | 17.9241 |
-| 1.3043        | 2.0   | 25778 | 1.1197          | 0.9071 | 17.9235 |
-| 1.2024        | 3.0   | 38667 | 1.0832          | 0.952  | 17.9397 |
+| 0.0           | 1.0   | 7668  | nan             | 0.6661 | 17.2141 |
+| 0.0           | 2.0   | 15336 | nan             | 0.6661 | 17.2141 |
 
 
 ### Framework versions