README.md

metadata

license: cc-by-sa-4.0
base_model: ClassCat/roberta-small-basque
tags:
  - generated_from_trainer
datasets:
  - basque_glue
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: XLM-EusBERTa-topic-classification
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: basque_glue
          type: basque_glue
          config: bhtc
          split: validation
          args: bhtc
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6494345718901454
          - name: F1
            type: f1
            value: 0.6432667195761544
          - name: Precision
            type: precision
            value: 0.6447174737999963
          - name: Recall
            type: recall
            value: 0.6494345718901454

XLM-EusBERTa-topic-classification

This model is a fine-tuned version of ClassCat/roberta-small-basque on the basque_glue dataset. It achieves the following results on the evaluation set:

• Loss: 4.2158
• Accuracy: 0.6494
• F1: 0.6433
• Precision: 0.6447
• Recall: 0.6494

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: 0.0001
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.2439	1.0	1074	1.1310	0.6581	0.6316	0.6139	0.6581
0.9539	2.0	2148	1.3019	0.6117	0.6034	0.6465	0.6117
0.579	3.0	3222	1.5533	0.6645	0.6524	0.6661	0.6645
0.3766	4.0	4296	2.3287	0.6381	0.6283	0.6590	0.6381
0.2641	5.0	5370	2.2805	0.6597	0.6515	0.6707	0.6597
0.1707	6.0	6444	2.6621	0.6397	0.6399	0.6581	0.6397
0.1537	7.0	7518	2.9116	0.6408	0.6336	0.6452	0.6408
0.0867	8.0	8592	3.1775	0.6344	0.6337	0.6531	0.6344
0.0779	9.0	9666	3.2514	0.6543	0.6471	0.6593	0.6543
0.0587	10.0	10740	3.3244	0.6457	0.6424	0.6488	0.6457
0.0322	11.0	11814	3.8090	0.6214	0.6244	0.6488	0.6214
0.0139	12.0	12888	3.8642	0.6247	0.6176	0.6424	0.6247
0.0256	13.0	13962	3.8734	0.6419	0.6327	0.6398	0.6419
0.0046	14.0	15036	4.0934	0.6365	0.6330	0.6463	0.6365
0.0036	15.0	16110	4.0890	0.6484	0.6416	0.6469	0.6484
0.0023	16.0	17184	4.0978	0.6505	0.6440	0.6470	0.6505
0.0008	17.0	18258	4.1709	0.6478	0.6418	0.6449	0.6478
0.0014	18.0	19332	4.1715	0.6505	0.6446	0.6458	0.6505
0.0007	19.0	20406	4.2158	0.6489	0.6427	0.6443	0.6489
0.0039	20.0	21480	4.2158	0.6494	0.6433	0.6447	0.6494

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0