metadata

license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: xlm-roberta-base-MLTC-rob
    results: []

xlm-roberta-base-MLTC-rob

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3645
F1: 0.8629
F1 Weighted: 0.8632
Roc Auc: 0.8598
Accuracy: 0.6067
Hamming Loss: 0.1401
Jaccard Score: 0.7588
Zero One Loss: 0.3933

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: 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

Training results

Training Loss	Epoch	Step	Validation Loss	F1	F1 Weighted	Roc Auc	Accuracy	Hamming Loss	Jaccard Score	Zero One Loss
0.5856	1.0	73	0.5884	0.7201	0.6619	0.6835	0.3393	0.3162	0.5627	0.6607
0.5053	2.0	146	0.4688	0.7718	0.7159	0.7712	0.4139	0.2288	0.6284	0.5861
0.3929	3.0	219	0.4002	0.8410	0.8413	0.8334	0.5347	0.1665	0.7256	0.4653
0.3293	4.0	292	0.3816	0.8471	0.8453	0.8399	0.5527	0.1600	0.7348	0.4473
0.3242	5.0	365	0.3607	0.8550	0.8538	0.8515	0.5784	0.1485	0.7467	0.4216
0.3228	6.0	438	0.3776	0.8495	0.8462	0.8437	0.5707	0.1562	0.7384	0.4293
0.2713	7.0	511	0.4086	0.8453	0.8412	0.8373	0.5630	0.1626	0.7320	0.4370
0.2519	8.0	584	0.3711	0.8534	0.8531	0.8489	0.5861	0.1510	0.7443	0.4139
0.2724	9.0	657	0.3645	0.8629	0.8632	0.8598	0.6067	0.1401	0.7588	0.3933
0.2484	10.0	730	0.3669	0.8586	0.8585	0.8553	0.5964	0.1446	0.7522	0.4036

Framework versions

Transformers 4.41.1
Pytorch 2.1.2
Datasets 2.19.1
Tokenizers 0.19.1