angrotanak-khmer-intent-classifier

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: 0.3511
• Accuracy: 0.9333
• F1: 0.9339
• Precision: 0.9417
• Recall: 0.9333

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: 8
• eval_batch_size: 8
• 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: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
2.0705	1.0	15	2.0505	0.2	0.1179	0.0984	0.2
1.9508	2.0	30	2.0051	0.1667	0.0476	0.0278	0.1667
1.8727	3.0	45	1.8381	0.2	0.0990	0.1287	0.2
1.7077	4.0	60	1.5842	0.3667	0.2852	0.3446	0.3667
1.5137	5.0	75	1.3019	0.5667	0.5114	0.5722	0.5667
1.33	6.0	90	1.1958	0.6333	0.5721	0.6734	0.6333
1.1648	7.0	105	0.9677	0.8667	0.8563	0.8718	0.8667
0.9707	8.0	120	0.8538	0.8	0.7853	0.8389	0.8
0.841	9.0	135	0.7069	0.8	0.7888	0.8290	0.8
0.736	10.0	150	0.5623	0.9	0.8872	0.8861	0.9
0.6398	11.0	165	0.4984	0.9	0.8967	0.9194	0.9
0.5907	12.0	180	0.3862	0.9333	0.9341	0.9472	0.9333
0.4658	13.0	195	0.3766	0.9333	0.9339	0.9417	0.9333
0.4504	14.0	210	0.3618	0.9333	0.9339	0.9417	0.9333
0.5012	15.0	225	0.3511	0.9333	0.9339	0.9417	0.9333

Framework versions

• Transformers 4.52.3
• Pytorch 2.7.0+cpu
• Datasets 3.6.0
• Tokenizers 0.21.1