xlm-roberta-base-ary-noaug

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

• Loss: 0.3450
• F1: 0.4024
• Roc Auc: 0.6517
• Accuracy: 0.4457

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
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.6062	1.0	51	0.4275	0.0	0.5	0.2846
0.3872	2.0	102	0.3855	0.0	0.5	0.2846
0.3884	3.0	153	0.3721	0.0	0.5	0.2846
0.3547	4.0	204	0.3543	0.0065	0.5017	0.2884
0.3325	5.0	255	0.3426	0.1204	0.5385	0.3521
0.3162	6.0	306	0.3316	0.1936	0.5671	0.3596
0.2998	7.0	357	0.3277	0.3095	0.6093	0.4045
0.2596	8.0	408	0.3303	0.3446	0.6328	0.4195
0.2312	9.0	459	0.3091	0.3306	0.6201	0.4307
0.2213	10.0	510	0.3254	0.3448	0.6309	0.4382
0.1956	11.0	561	0.3310	0.3460	0.6316	0.4419
0.182	12.0	612	0.3260	0.3933	0.6458	0.4457
0.1636	13.0	663	0.3397	0.3714	0.6441	0.4757
0.1586	14.0	714	0.3461	0.3667	0.6373	0.4307
0.1552	15.0	765	0.3557	0.3881	0.6507	0.4419
0.1435	16.0	816	0.3474	0.4000	0.6510	0.4494
0.1387	17.0	867	0.3450	0.4024	0.6517	0.4457
0.1349	18.0	918	0.3461	0.3946	0.6476	0.4345
0.1349	19.0	969	0.3475	0.3945	0.6483	0.4382
0.1243	20.0	1020	0.3473	0.3945	0.6483	0.4382

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0