out_dir

This model is a fine-tuned version of d4data/biomedical-ner-all on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7439
• Accuracy: 0.8286

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0748	1.0	9	1.0905	0.7513
0.9884	2.0	18	1.0329	0.7630
0.926	3.0	27	0.9917	0.7661
0.8732	4.0	36	0.9571	0.7717
0.8217	5.0	45	0.9290	0.7779
0.7878	6.0	54	0.9038	0.7834
0.7485	7.0	63	0.8815	0.7863
0.7075	8.0	72	0.8649	0.7962
0.6833	9.0	81	0.8501	0.7981
0.659	10.0	90	0.8348	0.7997
0.6291	11.0	99	0.8247	0.8047
0.6063	12.0	108	0.8142	0.8052
0.5911	13.0	117	0.8057	0.8088
0.5676	14.0	126	0.7965	0.8097
0.5491	15.0	135	0.7948	0.8114
0.5333	16.0	144	0.7852	0.8160
0.5263	17.0	153	0.7797	0.8170
0.513	18.0	162	0.7724	0.8180
0.4978	19.0	171	0.7708	0.8202
0.4889	20.0	180	0.7695	0.8208
0.4786	21.0	189	0.7646	0.8218
0.4653	22.0	198	0.7593	0.8233
0.4507	23.0	207	0.7579	0.8242
0.4426	24.0	216	0.7564	0.8253
0.4402	25.0	225	0.7508	0.8240
0.4254	26.0	234	0.7539	0.8251
0.4237	27.0	243	0.7528	0.8266
0.4149	28.0	252	0.7494	0.8261
0.4145	29.0	261	0.7482	0.8260
0.4021	30.0	270	0.7479	0.8278
0.3996	31.0	279	0.7487	0.8271
0.3926	32.0	288	0.7478	0.8275
0.3954	33.0	297	0.7468	0.8277
0.391	34.0	306	0.7473	0.8278
0.3899	35.0	315	0.7466	0.8279
0.3852	36.0	324	0.7463	0.8282
0.3823	37.0	333	0.7451	0.8282
0.3826	38.0	342	0.7439	0.8285
0.3813	39.0	351	0.7440	0.8287
0.379	40.0	360	0.7439	0.8286

Framework versions

• Transformers 4.54.0
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.2