RoBERTa-large-PM-M3-Voc-hf-finetuned-ner

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2320
• Precision: 0.7794
• Recall: 0.9175
• F1: 0.8429
• Accuracy: 0.9470

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: 10
• eval_batch_size: 10
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	36	1.0441	0.2844	0.0828	0.1283	0.7449
No log	2.0	72	0.7648	0.3878	0.5899	0.4680	0.7606
No log	3.0	108	0.5539	0.5096	0.6022	0.5520	0.8166
No log	4.0	144	0.6093	0.4714	0.7188	0.5694	0.7981
No log	5.0	180	0.5084	0.5185	0.7530	0.6141	0.8381
No log	6.0	216	0.3945	0.6099	0.7329	0.6658	0.8775
No log	7.0	252	0.4960	0.5273	0.7961	0.6344	0.8458
No log	8.0	288	0.3364	0.6501	0.8013	0.7178	0.9002
No log	9.0	324	0.3166	0.6601	0.8418	0.7399	0.9092
No log	10.0	360	0.2691	0.7087	0.8470	0.7717	0.9233
No log	11.0	396	0.2663	0.7215	0.8652	0.7868	0.9290
No log	12.0	432	0.2877	0.7138	0.8904	0.7924	0.9238
No log	13.0	468	0.2712	0.7353	0.8990	0.8090	0.9321
0.4329	14.0	504	0.2485	0.7511	0.8990	0.8184	0.9387
0.4329	15.0	540	0.2236	0.7859	0.9056	0.8416	0.9474
0.4329	16.0	576	0.2392	0.7696	0.9131	0.8352	0.9439
0.4329	17.0	612	0.2420	0.7684	0.9157	0.8356	0.9438
0.4329	18.0	648	0.2375	0.7708	0.9172	0.8377	0.9445
0.4329	19.0	684	0.2299	0.7832	0.9179	0.8452	0.9478
0.4329	20.0	720	0.2320	0.7794	0.9175	0.8429	0.9470

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1