First training run with the ul2-small-dutch model (no prompt-tuning, full model finetuning) on the increased dataset

README.md

@@ -0,0 +1,94 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: yhavinga/ul2-small-dutch
+tags:
+- generated_from_trainer
+model-index:
+- name: ul2-small-dutch-finetuned-oba-book-search-1
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# ul2-small-dutch-finetuned-oba-book-search-1
+
+This model is a fine-tuned version of [yhavinga/ul2-small-dutch](https://huggingface.co/yhavinga/ul2-small-dutch) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.2663
+- Top-5-accuracy: 0.0597
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.3
+- 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: cosine
+- lr_scheduler_warmup_steps: 1000
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch  | Step  | Validation Loss | Top-5-accuracy |
+|:-------------:|:------:|:-----:|:---------------:|:--------------:|
+| 6.4124        | 0.0848 | 500   | 6.3536          | 0.0            |
+| 6.7952        | 0.1696 | 1000  | 6.8349          | 0.0            |
+| 6.7698        | 0.2544 | 1500  | 6.5339          | 0.0            |
+| 6.8332        | 0.3392 | 2000  | 6.5288          | 0.0            |
+| 6.7705        | 0.4239 | 2500  | 7.1165          | 0.0            |
+| 6.997         | 0.5087 | 3000  | 6.8968          | 0.0            |
+| 6.7059        | 0.5935 | 3500  | 6.8425          | 0.0            |
+| 6.8888        | 0.6783 | 4000  | 6.3301          | 0.0            |
+| 6.4895        | 0.7631 | 4500  | 6.7270          | 0.0            |
+| 6.6173        | 0.8479 | 5000  | 6.1497          | 0.0796         |
+| 6.2961        | 0.9327 | 5500  | 5.9938          | 0.0796         |
+| 6.2324        | 1.0175 | 6000  | 5.8314          | 0.0            |
+| 6.3068        | 1.1023 | 6500  | 6.1160          | 0.0            |
+| 5.9052        | 1.1870 | 7000  | 5.8330          | 0.0            |
+| 6.2636        | 1.2718 | 7500  | 5.7442          | 0.0            |
+| 5.9638        | 1.3566 | 8000  | 5.9032          | 0.0            |
+| 5.4903        | 1.4414 | 8500  | 5.2657          | 0.0            |
+| 5.3572        | 1.5262 | 9000  | 5.3021          | 0.0            |
+| 5.268         | 1.6110 | 9500  | 4.8651          | 0.0            |
+| 5.1901        | 1.6958 | 10000 | 4.8621          | 0.0            |
+| 4.9671        | 1.7806 | 10500 | 4.8482          | 0.0            |
+| 4.8376        | 1.8654 | 11000 | 4.7743          | 0.0            |
+| 4.7365        | 1.9501 | 11500 | 5.2166          | 0.0            |
+| 4.529         | 2.0349 | 12000 | 4.4299          | 0.0796         |
+| 4.3585        | 2.1197 | 12500 | 4.2365          | 0.0199         |
+| 4.2613        | 2.2045 | 13000 | 4.0390          | 0.0            |
+| 4.1379        | 2.2893 | 13500 | 3.8705          | 0.1791         |
+| 3.9627        | 2.3741 | 14000 | 3.6937          | 0.1791         |
+| 3.8334        | 2.4589 | 14500 | 3.6306          | 0.0            |
+| 3.6814        | 2.5437 | 15000 | 3.5674          | 0.2388         |
+| 3.6308        | 2.6285 | 15500 | 3.4017          | 0.1990         |
+| 3.5401        | 2.7132 | 16000 | 3.3286          | 0.0597         |
+| 3.5336        | 2.7980 | 16500 | 3.2694          | 0.0199         |
+| 3.4879        | 2.8828 | 17000 | 3.2742          | 0.0398         |
+| 3.4277        | 2.9676 | 17500 | 3.2663          | 0.0597         |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 1.13.0+cu116
+- Datasets 3.0.0
+- Tokenizers 0.19.1

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "decoder_start_token_id": 0,
+  "diversity_penalty": 0.3,
+  "early_stopping": true,
+  "eos_token_id": 1,
+  "max_new_tokens": 20,
+  "no_repeat_ngram_size": 2,
+  "num_beam_groups": 5,
+  "num_beams": 5,
+  "num_return_sequences": 5,
+  "pad_token_id": 0,
+  "transformers_version": "4.44.2"
+}