yacine-djm/fg-bert-sustainability-15-1.5e-05-0.02-64

<!-- 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. -->

fg-bert-sustainability-15-1.5e-05-0.02-64

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

• —Loss: 0.0711
• —F1: 0.9215
• —Roc Auc: 0.9565
• —Accuracy: 0.8846

On the validation dataset :

• —The accuracy with hamming loss is 0.7800788954635107
• —The acccuracy as a metric is 0.8326530612244898
• —The following is the global precision score: 0.8695652173913043
• —The following is the global recall score: 0.8536585365853658
• —The following is the global f1-score: 0.8615384615384616

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: 1.5e-05
• —trainbatchsize: 64
• —evalbatchsize: 64
• —seed: 42
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —num_epochs: 15

Training results

Framework versions

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