a93e97b0c7a71fb81e93f8c19406bbd7

This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on the ccdv/patent-classification [abstract] dataset. It achieves the following results on the evaluation set:

• Loss: 1.5320
• Data Size: 1.0
• Epoch Runtime: 26.0070
• Accuracy: 0.6192
• F1 Macro: 0.5680

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	2.2206	0	2.0347	0.0142	0.0088
No log	1	781	2.0253	0.0078	2.4416	0.2218	0.0403
No log	2	1562	1.7810	0.0156	2.5507	0.3718	0.2328
No log	3	2343	1.5324	0.0312	3.1440	0.4099	0.2481
0.0398	4	3124	1.3743	0.0625	3.8438	0.5164	0.3580
1.3624	5	3905	1.2756	0.125	5.3338	0.5513	0.3924
1.2233	6	4686	1.1850	0.25	8.4389	0.5849	0.4327
1.09	7	5467	1.0934	0.5	14.2548	0.6256	0.4970
0.9652	8.0	6248	1.0531	1.0	26.7756	0.6400	0.5729
0.7921	9.0	7029	1.0719	1.0	26.4894	0.6380	0.5736
0.594	10.0	7810	1.2279	1.0	25.9718	0.6278	0.5837
0.4339	11.0	8591	1.3685	1.0	26.1953	0.6292	0.5886
0.352	12.0	9372	1.5320	1.0	26.0070	0.6192	0.5680

Framework versions

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