armahlovis
/

English2AkuapemTwi

@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [allenai/unifiedqa-t5-small](https://huggingface.co/allenai/unifiedqa-t5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.7965
-- Bleu: 2.9106
-- Gen Len: 18.0456
 ## Model description
@@ -37,28 +37,38 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 16
-- eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
 - mixed_precision_training: Native AMP
 ### Training results
 | Training Loss | Epoch | Step  | Validation Loss | Bleu   | Gen Len |
 |:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|
-| 2.3494        | 1.0   | 1271  | 2.1102          | 1.0276 | 18.5653 |
-| 2.2477        | 2.0   | 2542  | 2.0249          | 1.3762 | 18.45   |
-| 2.1948        | 3.0   | 3813  | 1.9608          | 1.828  | 18.2878 |
-| 2.1267        | 4.0   | 5084  | 1.9098          | 2.2275 | 18.2215 |
-| 2.0874        | 5.0   | 6355  | 1.8740          | 2.4207 | 18.1536 |
-| 2.049         | 6.0   | 7626  | 1.8446          | 2.6023 | 18.1613 |
-| 2.0445        | 7.0   | 8897  | 1.8232          | 2.7413 | 18.1157 |
-| 2.0083        | 8.0   | 10168 | 1.8085          | 2.8521 | 18.0515 |
-| 2.0093        | 9.0   | 11439 | 1.7994          | 2.8695 | 18.0852 |
-| 1.9934        | 10.0  | 12710 | 1.7965          | 2.9106 | 18.0456 |
 ### Framework versions

 This model is a fine-tuned version of [allenai/unifiedqa-t5-small](https://huggingface.co/allenai/unifiedqa-t5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 1.4994
+- Bleu: 6.0293
+- Gen Len: 17.7366
 ## Model description
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
+- 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
+- num_epochs: 20
 - mixed_precision_training: Native AMP
 ### Training results
 | Training Loss | Epoch | Step  | Validation Loss | Bleu   | Gen Len |
 |:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|
+| 2.8191        | 1.0   | 2542  | 2.5415          | 0.1546 | 18.9225 |
+| 2.5142        | 2.0   | 5084  | 2.2608          | 0.6935 | 18.7    |
+| 2.3329        | 3.0   | 7626  | 2.0914          | 1.0787 | 18.4717 |
+| 2.2235        | 4.0   | 10168 | 1.9743          | 1.581  | 18.3826 |
+| 2.1334        | 5.0   | 12710 | 1.8892          | 2.2749 | 18.1357 |
+| 2.0768        | 6.0   | 15252 | 1.8205          | 2.8561 | 17.941  |
+| 2.002         | 7.0   | 17794 | 1.7674          | 3.3599 | 17.9369 |
+| 1.9435        | 8.0   | 20336 | 1.7185          | 3.6487 | 17.9559 |
+| 1.8942        | 9.0   | 22878 | 1.6762          | 4.078  | 17.9443 |
+| 1.8609        | 10.0  | 25420 | 1.6416          | 4.4184 | 17.9453 |
+| 1.8462        | 11.0  | 27962 | 1.6123          | 4.7925 | 17.8511 |
+| 1.8036        | 12.0  | 30504 | 1.5870          | 5.0708 | 17.831  |
+| 1.7719        | 13.0  | 33046 | 1.5659          | 5.2373 | 17.7354 |
+| 1.7639        | 14.0  | 35588 | 1.5457          | 5.424  | 17.8385 |
+| 1.7486        | 15.0  | 38130 | 1.5316          | 5.5994 | 17.8004 |
+| 1.7293        | 16.0  | 40672 | 1.5204          | 5.7504 | 17.7288 |
+| 1.7257        | 17.0  | 43214 | 1.5109          | 5.8862 | 17.7461 |
+| 1.7077        | 18.0  | 45756 | 1.5045          | 5.9447 | 17.7518 |
+| 1.6986        | 19.0  | 48298 | 1.5006          | 6.0469 | 17.7217 |
+| 1.6967        | 20.0  | 50840 | 1.4994          | 6.0293 | 17.7366 |
 ### Framework versions