--- license: cc-by-nc-sa-4.0 tags: - generated_from_trainer model-index: - name: layoutlmv3-triplet results: [] --- # layoutlmv3-triplet This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0174 - Item: {'precision': 0.946751863684771, 'recall': 0.9348054679284963, 'f1': 0.9407407407407407, 'number': 951} - Aption: {'precision': 0.9266211604095563, 'recall': 0.9225280326197758, 'f1': 0.9245700664055849, 'number': 2943} - Ootnote: {'precision': 0.841726618705036, 'recall': 0.8068965517241379, 'f1': 0.823943661971831, 'number': 145} - Ormula: {'precision': 0.9741568112133158, 'recall': 0.9754385964912281, 'f1': 0.9747972824895901, 'number': 2280} - Overall Precision: 0.9450 - Overall Recall: 0.9408 - Overall F1: 0.9429 - Overall Accuracy: 0.9980 ## 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: 3e-05 - train_batch_size: 8 - eval_batch_size: 4 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | Item | Aption | Ootnote | Ormula | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | |:-------------:|:-----:|:-----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 0.0169 | 1.0 | 8507 | 0.0149 | {'precision': 0.8774703557312253, 'recall': 0.9337539432176656, 'f1': 0.9047376464595008, 'number': 951} | {'precision': 0.8646690813302601, 'recall': 0.8922867821950391, 'f1': 0.8782608695652173, 'number': 2943} | {'precision': 0.643312101910828, 'recall': 0.696551724137931, 'f1': 0.6688741721854304, 'number': 145} | {'precision': 0.951314088754847, 'recall': 0.968421052631579, 'f1': 0.9597913497065856, 'number': 2280} | 0.8921 | 0.9215 | 0.9066 | 0.9970 | | 0.0073 | 2.0 | 17014 | 0.0120 | {'precision': 0.9236326109391125, 'recall': 0.9411146161934806, 'f1': 0.9322916666666667, 'number': 951} | {'precision': 0.9089376053962901, 'recall': 0.9157322460074754, 'f1': 0.9123222748815166, 'number': 2943} | {'precision': 0.7913669064748201, 'recall': 0.7586206896551724, 'f1': 0.7746478873239436, 'number': 145} | {'precision': 0.9669708822251195, 'recall': 0.9758771929824561, 'f1': 0.9714036236629557, 'number': 2280} | 0.9296 | 0.9376 | 0.9336 | 0.9978 | | 0.0058 | 3.0 | 25521 | 0.0126 | {'precision': 0.9508379888268157, 'recall': 0.8948475289169295, 'f1': 0.9219934994582882, 'number': 951} | {'precision': 0.9105662983425414, 'recall': 0.8960244648318043, 'f1': 0.9032368556259633, 'number': 2943} | {'precision': 0.8, 'recall': 0.7724137931034483, 'f1': 0.7859649122807018, 'number': 145} | {'precision': 0.9717314487632509, 'recall': 0.9649122807017544, 'f1': 0.9683098591549295, 'number': 2280} | 0.9362 | 0.9179 | 0.9270 | 0.9978 | | 0.0039 | 4.0 | 34028 | 0.0123 | {'precision': 0.9646522234891676, 'recall': 0.889589905362776, 'f1': 0.925601750547046, 'number': 951} | {'precision': 0.913656690746474, 'recall': 0.9024804621134896, 'f1': 0.908034188034188, 'number': 2943} | {'precision': 0.738255033557047, 'recall': 0.7586206896551724, 'f1': 0.7482993197278912, 'number': 145} | {'precision': 0.9753086419753086, 'recall': 0.9701754385964912, 'f1': 0.9727352682497802, 'number': 2280} | 0.9392 | 0.9217 | 0.9304 | 0.9978 | | 0.0029 | 5.0 | 42535 | 0.0133 | {'precision': 0.9404761904761905, 'recall': 0.9137749737118822, 'f1': 0.9269333333333334, 'number': 951} | {'precision': 0.9182130584192439, 'recall': 0.90791709140333, 'f1': 0.9130360498889458, 'number': 2943} | {'precision': 0.8273381294964028, 'recall': 0.7931034482758621, 'f1': 0.8098591549295774, 'number': 145} | {'precision': 0.9727472527472527, 'recall': 0.9706140350877193, 'f1': 0.9716794731064764, 'number': 2280} | 0.9393 | 0.9288 | 0.9340 | 0.9979 | | 0.0022 | 6.0 | 51042 | 0.0148 | {'precision': 0.9139896373056995, 'recall': 0.9274447949526814, 'f1': 0.9206680584551148, 'number': 951} | {'precision': 0.9104477611940298, 'recall': 0.9119945633707102, 'f1': 0.911220505856391, 'number': 2943} | {'precision': 0.8226950354609929, 'recall': 0.8, 'f1': 0.8111888111888113, 'number': 145} | {'precision': 0.9750765194578049, 'recall': 0.9780701754385965, 'f1': 0.976571053207795, 'number': 2280} | 0.9323 | 0.9356 | 0.9340 | 0.9976 | | 0.0016 | 7.0 | 59549 | 0.0170 | {'precision': 0.9418729817007535, 'recall': 0.9200841219768665, 'f1': 0.9308510638297872, 'number': 951} | {'precision': 0.9165808444902163, 'recall': 0.9072375127420998, 'f1': 0.9118852459016393, 'number': 2943} | {'precision': 0.8405797101449275, 'recall': 0.8, 'f1': 0.8197879858657243, 'number': 145} | {'precision': 0.9752102700309871, 'recall': 0.9662280701754385, 'f1': 0.9706983917162371, 'number': 2280} | 0.9399 | 0.9280 | 0.9339 | 0.9977 | | 0.0013 | 8.0 | 68056 | 0.0187 | {'precision': 0.9455709711846318, 'recall': 0.9316508937960042, 'f1': 0.9385593220338984, 'number': 951} | {'precision': 0.9238387978142076, 'recall': 0.9191301393136255, 'f1': 0.9214784534150912, 'number': 2943} | {'precision': 0.9047619047619048, 'recall': 0.7862068965517242, 'f1': 0.8413284132841328, 'number': 145} | {'precision': 0.9723562966213252, 'recall': 0.9719298245614035, 'f1': 0.9721430138188198, 'number': 2280} | 0.9443 | 0.9370 | 0.9407 | 0.9979 | | 0.0009 | 9.0 | 76563 | 0.0169 | {'precision': 0.9375, 'recall': 0.9305993690851735, 'f1': 0.9340369393139841, 'number': 951} | {'precision': 0.9234449760765551, 'recall': 0.9181107713217805, 'f1': 0.9207701482364968, 'number': 2943} | {'precision': 0.8656716417910447, 'recall': 0.8, 'f1': 0.8315412186379928, 'number': 145} | {'precision': 0.9750328515111695, 'recall': 0.9763157894736842, 'f1': 0.9756738987508219, 'number': 2280} | 0.9431 | 0.9383 | 0.9407 | 0.9979 | | 0.0008 | 10.0 | 85070 | 0.0174 | {'precision': 0.946751863684771, 'recall': 0.9348054679284963, 'f1': 0.9407407407407407, 'number': 951} | {'precision': 0.9266211604095563, 'recall': 0.9225280326197758, 'f1': 0.9245700664055849, 'number': 2943} | {'precision': 0.841726618705036, 'recall': 0.8068965517241379, 'f1': 0.823943661971831, 'number': 145} | {'precision': 0.9741568112133158, 'recall': 0.9754385964912281, 'f1': 0.9747972824895901, 'number': 2280} | 0.9450 | 0.9408 | 0.9429 | 0.9980 | ### Framework versions - Transformers 4.26.0 - Pytorch 1.12.1 - Datasets 2.9.0 - Tokenizers 0.13.2