a66967f7d810a2481010ef968b7fb56e

This model is a fine-tuned version of FacebookAI/xlm-roberta-large-finetuned-conll03-english on the ccdv/patent-classification [abstract] dataset. It achieves the following results on the evaluation set:

• Loss: 1.9987
• Data Size: 1.0
• Epoch Runtime: 129.6531
• Accuracy: 0.2218
• F1 Macro: 0.0403

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
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	2.1808	0	7.4729	0.2214	0.0413
No log	1	781	2.0553	0.0078	8.8672	0.1510	0.0292
No log	2	1562	2.0295	0.0156	10.2667	0.1510	0.0292
No log	3	2343	2.0393	0.0312	12.7833	0.2071	0.0381
0.0455	4	3124	2.0019	0.0625	16.7424	0.2218	0.0403
2.0198	5	3905	1.9943	0.125	24.6536	0.2218	0.0403
2.0246	6	4686	1.9925	0.25	40.2799	0.2071	0.0381
2.0058	7	5467	1.9822	0.5	72.0111	0.2071	0.0381
1.9838	8.0	6248	1.9903	1.0	131.3184	0.2218	0.0403
1.9809	9.0	7029	1.9875	1.0	129.4071	0.2071	0.0381
1.9866	10.0	7810	1.9927	1.0	128.4434	0.2218	0.0403
1.9769	11.0	8591	1.9987	1.0	129.6531	0.2218	0.0403

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1