xlm-roberta-large_product_classifier

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

• Loss: 1.3981
• Accuracy: 0.8169

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 OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	490	0.8869	0.7423
1.3297	2.0	980	0.7796	0.7798
0.7265	3.0	1470	0.7592	0.7872
0.5509	4.0	1960	0.8112	0.7949
0.4258	5.0	2450	0.8498	0.7875
0.3307	6.0	2940	0.8326	0.8036
0.2702	7.0	3430	0.8833	0.8066
0.2078	8.0	3920	0.9260	0.8066
0.1571	9.0	4410	0.9800	0.8087
0.1242	10.0	4900	1.0725	0.8043
0.0962	11.0	5390	1.2147	0.7946
0.0857	12.0	5880	1.1705	0.8123
0.0667	13.0	6370	1.2551	0.8041
0.052	14.0	6860	1.2762	0.8184
0.0414	15.0	7350	1.3442	0.8115
0.0313	16.0	7840	1.3510	0.8130
0.0247	17.0	8330	1.3754	0.8133
0.0158	18.0	8820	1.3915	0.8135
0.0162	19.0	9310	1.3975	0.8186
0.0109	20.0	9800	1.3981	0.8169

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu124
• Datasets 2.21.0
• Tokenizers 0.21.0