lilt-invoices2

This model is a fine-tuned version of SCUT-DLVCLab/lilt-roberta-en-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0032
• Amount: {'precision': 0.9982517482517482, 'recall': 1.0, 'f1': 0.9991251093613298, 'number': 571}
• Billingaddress: {'precision': 1.0, 'recall': 0.9937888198757764, 'f1': 0.9968847352024921, 'number': 161}
• Description: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 612}
• Invoicedate: {'precision': 0.9942196531791907, 'recall': 1.0, 'f1': 0.9971014492753623, 'number': 172}
• Invoicetotal: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 207}
• Quantity: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 545}
• Subtotal: {'precision': 1.0, 'recall': 0.9933774834437086, 'f1': 0.9966777408637874, 'number': 151}
• Totaltax: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 139}
• Unitprice: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 492}
• Vendorname: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 208}
• Overall Precision: 0.9994
• Overall Recall: 0.9994
• Overall F1: 0.9994
• Overall Accuracy: 0.9994

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
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Amount	Billingaddress	Description	Invoicedate	Invoicetotal	Quantity	Subtotal	Totaltax	Unitprice	Vendorname	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.6178	4.35	100	0.1659	{'precision': 0.8553654743390358, 'recall': 0.9632224168126094, 'f1': 0.9060955518945634, 'number': 571}	{'precision': 0.9815950920245399, 'recall': 0.9937888198757764, 'f1': 0.9876543209876544, 'number': 161}	{'precision': 0.9775641025641025, 'recall': 0.9967320261437909, 'f1': 0.9870550161812297, 'number': 612}	{'precision': 0.9940476190476191, 'recall': 0.9709302325581395, 'f1': 0.9823529411764705, 'number': 172}	{'precision': 0.8571428571428571, 'recall': 0.8985507246376812, 'f1': 0.8773584905660375, 'number': 207}	{'precision': 0.9890909090909091, 'recall': 0.998165137614679, 'f1': 0.993607305936073, 'number': 545}	{'precision': 0.7664233576642335, 'recall': 0.695364238410596, 'f1': 0.7291666666666665, 'number': 151}	{'precision': 0.8818897637795275, 'recall': 0.8057553956834532, 'f1': 0.8421052631578947, 'number': 139}	{'precision': 0.9809523809523809, 'recall': 0.8373983739837398, 'f1': 0.9035087719298245, 'number': 492}	{'precision': 0.9856459330143541, 'recall': 0.9903846153846154, 'f1': 0.988009592326139, 'number': 208}	0.9368	0.9368	0.9368	0.9368
0.1653	8.7	200	0.0668	{'precision': 0.9420529801324503, 'recall': 0.9964973730297724, 'f1': 0.9685106382978723, 'number': 571}	{'precision': 0.9876543209876543, 'recall': 0.9937888198757764, 'f1': 0.9907120743034055, 'number': 161}	{'precision': 1.0, 'recall': 0.9901960784313726, 'f1': 0.9950738916256158, 'number': 612}	{'precision': 0.9941520467836257, 'recall': 0.9883720930232558, 'f1': 0.9912536443148688, 'number': 172}	{'precision': 0.9140271493212669, 'recall': 0.9758454106280193, 'f1': 0.9439252336448598, 'number': 207}	{'precision': 0.9945255474452555, 'recall': 1.0, 'f1': 0.9972552607502287, 'number': 545}	{'precision': 0.9328358208955224, 'recall': 0.8278145695364238, 'f1': 0.8771929824561403, 'number': 151}	{'precision': 0.9615384615384616, 'recall': 0.8992805755395683, 'f1': 0.929368029739777, 'number': 139}	{'precision': 0.9978947368421053, 'recall': 0.9634146341463414, 'f1': 0.9803516028955533, 'number': 492}	{'precision': 1.0, 'recall': 0.9951923076923077, 'f1': 0.9975903614457832, 'number': 208}	0.9770	0.9770	0.9770	0.9770
0.0676	13.04	300	0.0208	{'precision': 0.9861111111111112, 'recall': 0.9947460595446584, 'f1': 0.990409764603313, 'number': 571}	{'precision': 1.0, 'recall': 0.9937888198757764, 'f1': 0.9968847352024921, 'number': 161}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 612}	{'precision': 0.9941860465116279, 'recall': 0.9941860465116279, 'f1': 0.9941860465116279, 'number': 172}	{'precision': 0.9951219512195122, 'recall': 0.9855072463768116, 'f1': 0.9902912621359223, 'number': 207}	{'precision': 0.9963369963369964, 'recall': 0.998165137614679, 'f1': 0.9972502291475711, 'number': 545}	{'precision': 1.0, 'recall': 0.9602649006622517, 'f1': 0.9797297297297297, 'number': 151}	{'precision': 0.9787234042553191, 'recall': 0.9928057553956835, 'f1': 0.9857142857142858, 'number': 139}	{'precision': 0.9918864097363083, 'recall': 0.9939024390243902, 'f1': 0.9928934010152284, 'number': 492}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 208}	0.9942	0.9942	0.9942	0.9942
0.0296	17.39	400	0.0067	{'precision': 0.9982456140350877, 'recall': 0.9964973730297724, 'f1': 0.9973707274320772, 'number': 571}	{'precision': 1.0, 'recall': 0.9937888198757764, 'f1': 0.9968847352024921, 'number': 161}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 612}	{'precision': 0.9942196531791907, 'recall': 1.0, 'f1': 0.9971014492753623, 'number': 172}	{'precision': 0.9951923076923077, 'recall': 1.0, 'f1': 0.9975903614457832, 'number': 207}	{'precision': 0.9981684981684982, 'recall': 1.0, 'f1': 0.999083409715857, 'number': 545}	{'precision': 0.9933333333333333, 'recall': 0.9867549668874173, 'f1': 0.9900332225913622, 'number': 151}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 139}	{'precision': 0.9979674796747967, 'recall': 0.9979674796747967, 'f1': 0.9979674796747967, 'number': 492}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 208}	0.9982	0.9982	0.9982	0.9982
0.0143	21.74	500	0.0032	{'precision': 0.9982517482517482, 'recall': 1.0, 'f1': 0.9991251093613298, 'number': 571}	{'precision': 1.0, 'recall': 0.9937888198757764, 'f1': 0.9968847352024921, 'number': 161}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 612}	{'precision': 0.9942196531791907, 'recall': 1.0, 'f1': 0.9971014492753623, 'number': 172}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 207}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 545}	{'precision': 1.0, 'recall': 0.9933774834437086, 'f1': 0.9966777408637874, 'number': 151}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 139}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 492}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 208}	0.9994	0.9994	0.9994	0.9994

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3