g-patentsbertav2-e2e

This model is a fine-tuned version of AAUBS/PatentSBERTa_V2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4385
• Accuracy: 0.8350
• Precision: 0.2942
• Recall: 0.8453
• F1: 0.4365

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: 16
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.06
• num_epochs: 5.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.4397	0.1866	2000	0.4432	0.7928	0.7523	0.8625	0.8036
0.4168	0.3731	4000	0.4196	0.8100	0.8480	0.7473	0.7945
0.4098	0.5597	6000	0.4036	0.8172	0.7941	0.8478	0.8201
0.4003	0.7463	8000	0.4040	0.8204	0.8322	0.7949	0.8131
0.3876	0.9328	10000	0.3855	0.8256	0.7981	0.8636	0.8296
0.3634	1.1194	12000	0.4008	0.8283	0.8099	0.8501	0.8295
0.3531	1.3060	14000	0.3921	0.8334	0.8050	0.8722	0.8373
0.3448	1.4925	16000	0.3845	0.8329	0.8110	0.8607	0.8351
0.3741	1.6791	18000	0.3875	0.8379	0.8525	0.8104	0.8309
0.3558	1.8657	20000	0.3771	0.8416	0.8416	0.8348	0.8382
0.3137	2.0522	22000	0.3830	0.8415	0.8304	0.8512	0.8407
0.2852	2.2388	24000	0.4188	0.8359	0.8102	0.8699	0.8390
0.2933	2.4254	26000	0.4280	0.8426	0.8333	0.8496	0.8414
0.3250	2.6119	28000	0.3846	0.8383	0.8162	0.8659	0.8403
0.3087	2.7985	30000	0.3822	0.8418	0.8313	0.8508	0.8409
0.2794	2.9851	32000	0.3997	0.8398	0.8267	0.8528	0.8395
0.2409	3.1716	34000	0.4402	0.8354	0.8198	0.8523	0.8358
0.2451	3.3582	36000	0.4359	0.8408	0.8340	0.8440	0.8390
0.2557	3.5448	38000	0.4241	0.8383	0.8382	0.8314	0.8348
0.2409	3.7313	40000	0.4388	0.8420	0.8482	0.8264	0.8372
0.2493	3.9179	42000	0.4319	0.8468	0.8357	0.8566	0.8460
0.2013	4.1045	44000	0.4871	0.8409	0.8329	0.8460	0.8394
0.2074	4.2910	46000	0.4942	0.8401	0.8373	0.8375	0.8374
0.2205	4.4776	48000	0.5179	0.8421	0.8388	0.8402	0.8395
0.1912	4.6642	50000	0.4978	0.8385	0.8274	0.8483	0.8377
0.1933	4.8507	52000	0.5141	0.8391	0.8238	0.8557	0.8395

Framework versions

• Transformers 5.0.0
• Pytorch 2.9.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2