layoutkv / README.md
pedro1111's picture
End of training
feeaeda verified
metadata
license: cc-by-nc-sa-4.0
base_model: microsoft/layoutlmv2-base-uncased
tags:
  - generated_from_trainer
datasets:
  - funsd
model-index:
  - name: layoutkv
    results: []

layoutkv

This model is a fine-tuned version of microsoft/layoutlmv2-base-uncased on the funsd dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6316
  • Answer: {'precision': 0.06269757639620653, 'recall': 0.14709517923362175, 'f1': 0.08792020687107498, 'number': 809}
  • Header: {'precision': 0.02142857142857143, 'recall': 0.025210084033613446, 'f1': 0.023166023166023165, 'number': 119}
  • Question: {'precision': 0.17976470588235294, 'recall': 0.3586854460093897, 'f1': 0.2394984326018809, 'number': 1065}
  • Overall Precision: 0.1211
  • Overall Recall: 0.2529
  • Overall F1: 0.1637
  • Overall Accuracy: 0.3969

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: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss Answer Header Question Overall Precision Overall Recall Overall F1 Overall Accuracy
1.9011 1.0 10 1.8281 {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 1065} 0.0 0.0 0.0 0.2901
1.7212 2.0 20 1.6755 {'precision': 0.010714285714285714, 'recall': 0.011124845488257108, 'f1': 0.01091570648878108, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.8, 'recall': 0.003755868544600939, 'f1': 0.007476635514018691, 'number': 1065} 0.0154 0.0065 0.0092 0.3484
1.6644 3.0 30 1.7453 {'precision': 0.0030959752321981426, 'recall': 0.0012360939431396785, 'f1': 0.0017667844522968195, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.4895833333333333, 'recall': 0.044131455399061034, 'f1': 0.08096468561584841, 'number': 1065} 0.1143 0.0241 0.0398 0.3130
1.5949 4.0 40 1.7670 {'precision': 0.03335250143760782, 'recall': 0.07169344870210136, 'f1': 0.04552590266875981, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.13106796116504854, 'recall': 0.2535211267605634, 'f1': 0.1728, 'number': 1065} 0.0859 0.1646 0.1129 0.3285
1.4559 5.0 50 1.5921 {'precision': 0.05108940646130729, 'recall': 0.08405438813349815, 'f1': 0.06355140186915888, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.17631224764468373, 'recall': 0.2460093896713615, 'f1': 0.20540964327714623, 'number': 1065} 0.1171 0.1656 0.1372 0.3763
1.3707 6.0 60 1.6238 {'precision': 0.049044914816726896, 'recall': 0.11742892459826947, 'f1': 0.06919155134741442, 'number': 809} {'precision': 0.006622516556291391, 'recall': 0.008403361344537815, 'f1': 0.007407407407407407, 'number': 119} {'precision': 0.16497339138848574, 'recall': 0.32018779342723, 'f1': 0.21775223499361432, 'number': 1065} 0.1052 0.2193 0.1422 0.3786
1.2836 7.0 70 1.5846 {'precision': 0.058695652173913045, 'recall': 0.1334981458590853, 'f1': 0.08154020385050964, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.15556492411467115, 'recall': 0.3464788732394366, 'f1': 0.2147221414023858, 'number': 1065} 0.1120 0.2393 0.1526 0.3851
1.2161 8.0 80 1.6814 {'precision': 0.06025974025974026, 'recall': 0.1433868974042027, 'f1': 0.08485735186539868, 'number': 809} {'precision': 0.009433962264150943, 'recall': 0.008403361344537815, 'f1': 0.008888888888888889, 'number': 119} {'precision': 0.1631912964641886, 'recall': 0.3380281690140845, 'f1': 0.22011617242433507, 'number': 1065} 0.1126 0.2393 0.1531 0.3730
1.1499 9.0 90 1.6027 {'precision': 0.06253521126760564, 'recall': 0.13720642768850433, 'f1': 0.08591331269349846, 'number': 809} {'precision': 0.011111111111111112, 'recall': 0.008403361344537815, 'f1': 0.009569377990430622, 'number': 119} {'precision': 0.1547870097005483, 'recall': 0.34460093896713617, 'f1': 0.21362048894062866, 'number': 1065} 0.1131 0.2403 0.1538 0.3710
1.1199 10.0 100 1.6616 {'precision': 0.055350553505535055, 'recall': 0.12978986402966625, 'f1': 0.07760532150776053, 'number': 809} {'precision': 0.019867549668874173, 'recall': 0.025210084033613446, 'f1': 0.022222222222222223, 'number': 119} {'precision': 0.16207042851081885, 'recall': 0.3586854460093897, 'f1': 0.22326125073056693, 'number': 1065} 0.1112 0.2459 0.1532 0.3719
1.0651 11.0 110 1.6100 {'precision': 0.06031016657093624, 'recall': 0.12978986402966625, 'f1': 0.08235294117647059, 'number': 809} {'precision': 0.015151515151515152, 'recall': 0.01680672268907563, 'f1': 0.01593625498007968, 'number': 119} {'precision': 0.163854351687389, 'recall': 0.3464788732394366, 'f1': 0.22249020198974978, 'number': 1065} 0.1154 0.2388 0.1556 0.3805
1.0454 12.0 120 1.5988 {'precision': 0.0639269406392694, 'recall': 0.138442521631644, 'f1': 0.08746583365872705, 'number': 809} {'precision': 0.022222222222222223, 'recall': 0.025210084033613446, 'f1': 0.02362204724409449, 'number': 119} {'precision': 0.17867298578199053, 'recall': 0.3539906103286385, 'f1': 0.23748031496062993, 'number': 1065} 0.1231 0.2469 0.1643 0.3977
1.0279 13.0 130 1.6209 {'precision': 0.06463104325699745, 'recall': 0.15698393077873918, 'f1': 0.09156452775775054, 'number': 809} {'precision': 0.022388059701492536, 'recall': 0.025210084033613446, 'f1': 0.02371541501976284, 'number': 119} {'precision': 0.18118811881188118, 'recall': 0.3436619718309859, 'f1': 0.23727714748784443, 'number': 1065} 0.1204 0.2489 0.1623 0.3998
1.008 14.0 140 1.6538 {'precision': 0.0633116883116883, 'recall': 0.1446229913473424, 'f1': 0.08806925103500188, 'number': 809} {'precision': 0.022556390977443608, 'recall': 0.025210084033613446, 'f1': 0.023809523809523808, 'number': 119} {'precision': 0.18536350505536833, 'recall': 0.3615023474178404, 'f1': 0.24506683640992996, 'number': 1065} 0.1244 0.2534 0.1669 0.3965
0.9812 15.0 150 1.6316 {'precision': 0.06269757639620653, 'recall': 0.14709517923362175, 'f1': 0.08792020687107498, 'number': 809} {'precision': 0.02142857142857143, 'recall': 0.025210084033613446, 'f1': 0.023166023166023165, 'number': 119} {'precision': 0.17976470588235294, 'recall': 0.3586854460093897, 'f1': 0.2394984326018809, 'number': 1065} 0.1211 0.2529 0.1637 0.3969

Framework versions

  • Transformers 4.32.0
  • Pytorch 2.0.0+cu118
  • Datasets 2.17.1
  • Tokenizers 0.13.2