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: 0.2218
• Precision: 0.7339
• Recall: 0.8538
• F1: 0.7893
• Accuracy: 0.9364

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: 16
• 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	23	1.3251	0.0196	0.0004	0.0008	0.7273
No log	2.0	46	0.8784	0.3740	0.3448	0.3588	0.7640
No log	3.0	69	0.6955	0.4452	0.4643	0.4545	0.7952
No log	4.0	92	0.6009	0.4915	0.5792	0.5318	0.8149
No log	5.0	115	0.5379	0.5531	0.6262	0.5874	0.8393
No log	6.0	138	0.4651	0.5850	0.6701	0.6247	0.8613
No log	7.0	161	0.4385	0.5673	0.7214	0.6352	0.8653
No log	8.0	184	0.4528	0.5548	0.7614	0.6419	0.8587
No log	9.0	207	0.3304	0.6511	0.7723	0.7066	0.9022
No log	10.0	230	0.3257	0.6451	0.8013	0.7148	0.9019
No log	11.0	253	0.3373	0.6289	0.8068	0.7068	0.8983
No log	12.0	276	0.2834	0.6812	0.8205	0.7444	0.9180
No log	13.0	299	0.2577	0.7127	0.8284	0.7662	0.9262
No log	14.0	322	0.2315	0.7351	0.8295	0.7795	0.9349
No log	15.0	345	0.2377	0.7146	0.8409	0.7726	0.9306
No log	16.0	368	0.2445	0.7058	0.8534	0.7726	0.9286
No log	17.0	391	0.2232	0.7359	0.8538	0.7905	0.9366
No log	18.0	414	0.2239	0.7320	0.8531	0.7879	0.9359
No log	19.0	437	0.2218	0.7345	0.8542	0.7899	0.9366
No log	20.0	460	0.2218	0.7339	0.8538	0.7893	0.9364

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1