layoutlm-funsd

This model is a fine-tuned version of microsoft/layoutlm-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8927
• Column: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}
• Ignore: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}
• Key: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 17}
• Value: {'precision': 0.6666666666666666, 'recall': 0.48484848484848486, 'f1': 0.5614035087719298, 'number': 33}
• Overall Precision: 0.6875
• Overall Recall: 0.4231
• Overall F1: 0.5238
• Overall Accuracy: 0.7947

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	Column	Ignore	Key	Value	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
2.4627	1.0	2	2.1288	{'precision': 0.23529411764705882, 'recall': 0.16, 'f1': 0.19047619047619052, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.06060606060606061, 'recall': 0.06060606060606061, 'f1': 0.06060606060606061, 'number': 33}	0.0870	0.0769	0.0816	0.6887
2.1025	2.0	4	1.7650	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 33}	0.0	0.0	0.0	0.6921
1.7503	3.0	6	1.4611	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 33}	0.0	0.0	0.0	0.6904
1.4557	4.0	8	1.2624	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 33}	0.0	0.0	0.0	0.6904
1.3067	5.0	10	1.1889	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 33}	0.0	0.0	0.0	0.6904
1.1884	6.0	12	1.1436	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 33}	0.0	0.0	0.0	0.6904
1.1456	7.0	14	1.0901	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 17}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 33}	0.0	0.0	0.0	0.6904
1.0915	8.0	16	1.0410	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 0.11764705882352941, 'f1': 0.21052631578947367, 'number': 17}	{'precision': 0.3333333333333333, 'recall': 0.030303030303030304, 'f1': 0.05555555555555555, 'number': 33}	0.6	0.0385	0.0723	0.6937
1.0428	9.0	18	0.9990	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 0.29411764705882354, 'f1': 0.45454545454545453, 'number': 17}	{'precision': 0.23529411764705882, 'recall': 0.12121212121212122, 'f1': 0.16, 'number': 33}	0.2727	0.1154	0.1622	0.7252
0.9819	10.0	20	0.9639	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 0.4117647058823529, 'f1': 0.5833333333333334, 'number': 17}	{'precision': 0.2631578947368421, 'recall': 0.15151515151515152, 'f1': 0.19230769230769232, 'number': 33}	0.3243	0.1538	0.2087	0.7517
0.9592	11.0	22	0.9344	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 0.6470588235294118, 'f1': 0.7857142857142858, 'number': 17}	{'precision': 0.3684210526315789, 'recall': 0.21212121212121213, 'f1': 0.2692307692307693, 'number': 33}	0.4737	0.2308	0.3103	0.7781
0.9011	12.0	24	0.9105	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 17}	{'precision': 0.64, 'recall': 0.48484848484848486, 'f1': 0.5517241379310344, 'number': 33}	0.66	0.4231	0.5156	0.7930
0.9426	13.0	26	0.8927	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 17}	{'precision': 0.6666666666666666, 'recall': 0.48484848484848486, 'f1': 0.5614035087719298, 'number': 33}	0.6875	0.4231	0.5238	0.7947
0.8809	14.0	28	0.8821	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 17}	{'precision': 0.6666666666666666, 'recall': 0.48484848484848486, 'f1': 0.5614035087719298, 'number': 33}	0.6875	0.4231	0.5238	0.7947
0.9188	15.0	30	0.8774	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 25}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 17}	{'precision': 0.6666666666666666, 'recall': 0.48484848484848486, 'f1': 0.5614035087719298, 'number': 33}	0.6875	0.4231	0.5238	0.7947

Framework versions

• Transformers 4.32.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3