esahit
/

ul2-large-dutch-finetuned-oba-book-search

PEFT

TensorBoard

Safetensors

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

esahit commited on Sep 21

Commit

b1d2b94

•

1 Parent(s): ad91fd7

Second full training run on complete dataset

Browse files

Files changed (1) hide show

README.md +32 -32

README.md CHANGED Viewed

@@ -14,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
 # ul2-large-dutch-finetuned-oba-book-search
-This model is a fine-tuned version of [yhavinga/ul2-large-dutch](https://huggingface.co/yhavinga/ul2-large-dutch) on a sample dataset from the public library of Amsterdam (OBA).
 It achieves the following results on the evaluation set:
-- Loss: 5.4042
-- Top-5-accuracy: 0.0
 ## Model description
@@ -36,7 +36,7 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 0.001
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
@@ -48,34 +48,34 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch  | Step | Validation Loss | Top-5-accuracy |
 |:-------------:|:------:|:----:|:---------------:|:--------------:|
-| 8.2793        | 0.1729 | 200  | 5.9519          | 0.0            |
-| 8.1572        | 0.3457 | 400  | 5.9147          | 0.0            |
-| 8.1924        | 0.5186 | 600  | 5.8604          | 0.0            |
-| 8.021         | 0.6914 | 800  | 5.8270          | 0.0            |
-| 7.9558        | 0.8643 | 1000 | 5.7974          | 0.0            |
-| 7.9786        | 1.0372 | 1200 | 5.7586          | 0.0            |
-| 7.9113        | 1.2100 | 1400 | 5.7306          | 0.0            |
-| 8.0483        | 1.3829 | 1600 | 5.7506          | 0.0            |
-| 7.8481        | 1.5557 | 1800 | 5.7116          | 0.0            |
-| 7.9376        | 1.7286 | 2000 | 5.6599          | 0.0            |
-| 7.7537        | 1.9015 | 2200 | 5.6289          | 0.0            |
-| 7.7101        | 2.0743 | 2400 | 5.5863          | 0.0            |
-| 7.653         | 2.2472 | 2600 | 5.5719          | 0.0            |
-| 7.7515        | 2.4201 | 2800 | 5.5510          | 0.0            |
-| 7.6844        | 2.5929 | 3000 | 5.5245          | 0.0            |
-| 7.7322        | 2.7658 | 3200 | 5.5087          | 0.0            |
-| 7.7169        | 2.9386 | 3400 | 5.5065          | 0.0            |
-| 7.6177        | 3.1115 | 3600 | 5.4846          | 0.0            |
-| 7.6558        | 3.2844 | 3800 | 5.4712          | 0.0            |
-| 7.6453        | 3.4572 | 4000 | 5.4564          | 0.0            |
-| 7.5664        | 3.6301 | 4200 | 5.4431          | 0.0            |
-| 7.5475        | 3.8029 | 4400 | 5.4432          | 0.0            |
-| 7.5741        | 3.9758 | 4600 | 5.4393          | 0.0            |
-| 7.5523        | 4.1487 | 4800 | 5.4268          | 0.0            |
-| 7.6833        | 4.3215 | 5000 | 5.4243          | 0.0            |
-| 7.6817        | 4.4944 | 5200 | 5.4098          | 0.0            |
-| 7.544         | 4.6672 | 5400 | 5.4070          | 0.0            |
-| 7.6062        | 4.8401 | 5600 | 5.4033          | 0.0            |
 ### Framework versions

 # ul2-large-dutch-finetuned-oba-book-search
+This model is a fine-tuned version of [yhavinga/ul2-large-dutch](https://huggingface.co/yhavinga/ul2-large-dutch) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 4.4355
+- Top-5-accuracy: 0.3172
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 0.03
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 | Training Loss | Epoch  | Step | Validation Loss | Top-5-accuracy |
 |:-------------:|:------:|:----:|:---------------:|:--------------:|
+| 7.5872        | 0.1729 | 200  | 5.1429          | 0.0            |
+| 6.5938        | 0.3457 | 400  | 4.5886          | 0.0680         |
+| 6.2525        | 0.5186 | 600  | 4.7437          | 0.0453         |
+| 6.1056        | 0.6914 | 800  | 4.6423          | 0.2493         |
+| 6.0637        | 0.8643 | 1000 | 4.6631          | 0.1586         |
+| 5.9667        | 1.0372 | 1200 | 4.5781          | 0.0227         |
+| 5.8045        | 1.2100 | 1400 | 4.4807          | 0.1360         |
+| 5.7737        | 1.3829 | 1600 | 4.4840          | 0.0453         |
+| 5.7215        | 1.5557 | 1800 | 4.5203          | 0.0680         |
+| 5.6041        | 1.7286 | 2000 | 4.4304          | 0.0680         |
+| 5.5924        | 1.9015 | 2200 | 4.5078          | 0.0453         |
+| 5.5273        | 2.0743 | 2400 | 4.5331          | 0.1133         |
+| 5.5034        | 2.2472 | 2600 | 4.4927          | 0.0453         |
+| 5.5556        | 2.4201 | 2800 | 4.4918          | 0.0227         |
+| 5.5071        | 2.5929 | 3000 | 4.4780          | 0.1133         |
+| 5.4722        | 2.7658 | 3200 | 4.4836          | 0.0453         |
+| 5.4154        | 2.9386 | 3400 | 4.4614          | 0.0680         |
+| 5.4479        | 3.1115 | 3600 | 4.4920          | 0.0680         |
+| 5.3862        | 3.2844 | 3800 | 4.5000          | 0.0680         |
+| 5.3983        | 3.4572 | 4000 | 4.4335          | 0.1360         |
+| 5.3797        | 3.6301 | 4200 | 4.5029          | 0.1586         |
+| 5.3729        | 3.8029 | 4400 | 4.4466          | 0.1586         |
+| 5.4011        | 3.9758 | 4600 | 4.4359          | 0.1813         |
+| 5.3342        | 4.1487 | 4800 | 4.4264          | 0.2039         |
+| 5.3995        | 4.3215 | 5000 | 4.4568          | 0.2039         |
+| 5.2805        | 4.4944 | 5200 | 4.4460          | 0.2719         |
+| 5.3532        | 4.6672 | 5400 | 4.4434          | 0.3172         |
+| 5.3444        | 4.8401 | 5600 | 4.4355          | 0.3172         |
 ### Framework versions