metadata

license: mit
base_model: nielsr/lilt-xlm-roberta-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: tst_lilt_cord_xlm_ft
    results: []

tst_lilt_cord_xlm_ft

This model is a fine-tuned version of nielsr/lilt-xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2024
Precision: 0.9578
Recall: 0.9555
F1: 0.9567
Accuracy: 0.9657

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: 5e-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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.25	100	0.4066	0.8691	0.8544	0.8617	0.8774
No log	0.5	200	0.3936	0.8724	0.8633	0.8678	0.8820
No log	0.75	300	0.4283	0.8802	0.8794	0.8798	0.8902
No log	1.0	400	0.2634	0.9326	0.9183	0.9254	0.9419
0.3486	1.25	500	0.2464	0.9143	0.9150	0.9147	0.9401
0.3486	1.5	600	0.2368	0.9296	0.9296	0.9296	0.9460
0.3486	1.75	700	0.2537	0.9434	0.9442	0.9438	0.9552
0.3486	2.0	800	0.2233	0.9504	0.9466	0.9485	0.9552
0.3486	2.25	900	0.2449	0.9482	0.9482	0.9482	0.9593
0.1869	2.5	1000	0.2214	0.9540	0.9555	0.9547	0.9611
0.1869	2.75	1100	0.2304	0.9352	0.9337	0.9344	0.9446
0.1869	3.0	1200	0.2748	0.9432	0.9401	0.9417	0.9520
0.1869	3.25	1300	0.2104	0.9460	0.9490	0.9475	0.9579
0.1869	3.5	1400	0.2379	0.9496	0.9458	0.9477	0.9597
0.1043	3.75	1500	0.2067	0.9466	0.9466	0.9466	0.9579
0.1043	4.0	1600	0.2025	0.9562	0.9547	0.9555	0.9634
0.1043	4.25	1700	0.2082	0.9514	0.9506	0.9510	0.9602
0.1043	4.5	1800	0.1942	0.9547	0.9555	0.9551	0.9666
0.1043	4.75	1900	0.2007	0.9532	0.9547	0.9539	0.9652
0.0574	5.0	2000	0.2024	0.9578	0.9555	0.9567	0.9657

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1