bastianchinchon/nominal-groups-recognition-roberta-clinical-wl-es

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

nominal-groups-recognition-roberta-clinical-wl-es

This model is a fine-tuned version of plncmm/roberta-clinical-wl-es on the bastianchinchon/spanishnominalgroups_conll2003 dataset. It achieves the following results on the evaluation set:

• —Loss: 0.2226
• —Body Part Precision: 0.7427
• —Body Part Recall: 0.7966
• —Body Part F1: 0.7687
• —Body Part Number: 413
• —Disease Precision: 0.7915
• —Disease Recall: 0.8174
• —Disease F1: 0.8042
• —Disease Number: 975
• —Family Member Precision: 0.8286
• —Family Member Recall: 0.9667
• —Family Member F1: 0.8923
• —Family Member Number: 30
• —Medication Precision: 0.7905
• —Medication Recall: 0.8925
• —Medication F1: 0.8384
• —Medication Number: 93
• —Procedure Precision: 0.7105
• —Procedure Recall: 0.7814
• —Procedure F1: 0.7443
• —Procedure Number: 311
• —Overall Precision: 0.7666
• —Overall Recall: 0.8128
• —Overall F1: 0.7890
• —Overall Accuracy: 0.9374

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
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 13
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —num_epochs: 2

Training results

Framework versions

• —Transformers 4.30.2
• —Pytorch 2.0.1+cu118
• —Datasets 2.13.1
• —Tokenizers 0.13.3