huyhuyvu01/DeBERTa_large_NER_chartering_email

<!-- 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. -->

bbc-ner-deberta-largebaseline2dims

This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:

• —Loss: 0.1481
• —Cargo Dimension Precision: 0.7606
• —Cargo Dimension Recall: 0.9474
• —Cargo Dimension F1: 0.8437
• —Cargo Dimension Number: 114
• —Cargo Quantity Precision: 0.8079
• —Cargo Quantity Recall: 0.8937
• —Cargo Quantity F1: 0.8486
• —Cargo Quantity Number: 207
• —Cargo Requirements Precision: 0.4962
• —Cargo Requirements Recall: 0.6535
• —Cargo Requirements F1: 0.5641
• —Cargo Requirements Number: 202
• —Cargo Stowage Factor Precision: 0.8226
• —Cargo Stowage Factor Recall: 0.8361
• —Cargo Stowage Factor F1: 0.8293
• —Cargo Stowage Factor Number: 122
• —Cargo Type Precision: 0.7885
• —Cargo Type Recall: 0.8183
• —Cargo Type F1: 0.8031
• —Cargo Type Number: 688
• —Cargo Weigh Volume Precision: 0.8528
• —Cargo Weigh Volume Recall: 0.9026
• —Cargo Weigh Volume F1: 0.8770
• —Cargo Weigh Volume Number: 719
• —Commission Rate Precision: 0.7955
• —Commission Rate Recall: 0.8452
• —Commission Rate F1: 0.8196
• —Commission Rate Number: 336
• —Discharging Port Precision: 0.8706
• —Discharging Port Recall: 0.9015
• —Discharging Port F1: 0.8858
• —Discharging Port Number: 843
• —Laycan Date Precision: 0.8260
• —Laycan Date Recall: 0.8710
• —Laycan Date F1: 0.8479
• —Laycan Date Number: 496
• —Loading Discharging Terms Precision: 0.7211
• —Loading Discharging Terms Recall: 0.7975
• —Loading Discharging Terms F1: 0.7574
• —Loading Discharging Terms Number: 321
• —Loading Port Precision: 0.8906
• —Loading Port Recall: 0.9232
• —Loading Port F1: 0.9066
• —Loading Port Number: 899
• —Shipment Terms Precision: 0.6780
• —Shipment Terms Recall: 0.6780
• —Shipment Terms F1: 0.6780
• —Shipment Terms Number: 118
• —Vessel Requirements Precision: 0.3786
• —Vessel Requirements Recall: 0.5132
• —Vessel Requirements F1: 0.4358
• —Vessel Requirements Number: 76
• —Overall Precision: 0.8041
• —Overall Recall: 0.8598
• —Overall F1: 0.8310
• —Overall Accuracy: 0.9688

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: 5e-05
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 16
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —num_epochs: 10.0

Training results

Framework versions

• —Transformers 4.36.0.dev0
• —Pytorch 2.1.0+cu121
• —Datasets 2.14.5
• —Tokenizers 0.15.0