metadata

library_name: transformers
license: mit
base_model: FacebookAI/xlm-roberta-large
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: xlmr_synset_classifier_marked
    results: []

xlmr_synset_classifier_marked

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

Loss: 0.5065
Accuracy: 0.8542
F1: 0.8462
Precision: 0.8503
Recall: 0.8542
F1 Macro: 0.6998
Precision Macro: 0.6940
Recall Macro: 0.7209
F1 Micro: 0.8542
Precision Micro: 0.8542
Recall Micro: 0.8542

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: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 50
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	F1 Macro	Precision Macro	Recall Macro	F1 Micro	Precision Micro	Recall Micro
3.4502	0.6221	100	1.8070	0.6245	0.5355	0.5022	0.6245	0.2357	0.2456	0.2565	0.6245	0.6245	0.6245
1.1977	1.2442	200	0.7296	0.8116	0.7934	0.8034	0.8116	0.5365	0.5496	0.5643	0.8116	0.8116	0.8116
0.724	1.8663	300	0.6379	0.8282	0.8150	0.8301	0.8282	0.5981	0.5903	0.6309	0.8282	0.8282	0.8282
0.5655	2.4883	400	0.5609	0.8398	0.8267	0.8326	0.8398	0.6235	0.6094	0.6519	0.8398	0.8398	0.8398
0.5095	3.1104	500	0.5166	0.8488	0.8389	0.8470	0.8488	0.6594	0.6492	0.6862	0.8488	0.8488	0.8488
0.4206	3.7325	600	0.4964	0.8479	0.8396	0.8412	0.8479	0.6778	0.6770	0.6923	0.8479	0.8479	0.8479
0.386	4.3546	700	0.5091	0.8502	0.8418	0.8468	0.8502	0.6949	0.6930	0.7146	0.8502	0.8502	0.8502
0.3463	4.9767	800	0.5065	0.8542	0.8462	0.8503	0.8542	0.6998	0.6940	0.7209	0.8542	0.8542	0.8542

Framework versions

Transformers 4.45.2
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.20.3