layoutlm-cord

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

• eval_loss: 0.1624
• eval_enu.cnt: {'precision': 0.9861111111111112, 'recall': 0.9681818181818181, 'f1': 0.9770642201834862, 'number': 220}
• eval_enu.discountprice: {'precision': 0.6666666666666666, 'recall': 0.6, 'f1': 0.631578947368421, 'number': 10}
• eval_enu.etc: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}
• eval_enu.itemsubtotal: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 6}
• eval_enu.nm: {'precision': 0.9525691699604744, 'recall': 0.9601593625498008, 'f1': 0.9563492063492064, 'number': 251}
• eval_enu.num: {'precision': 0.9090909090909091, 'recall': 0.9090909090909091, 'f1': 0.9090909090909091, 'number': 11}
• eval_enu.price: {'precision': 0.9568627450980393, 'recall': 0.991869918699187, 'f1': 0.9740518962075848, 'number': 246}
• eval_enu.sub.cnt: {'precision': 0.85, 'recall': 1.0, 'f1': 0.9189189189189189, 'number': 17}
• eval_enu.sub.nm: {'precision': 0.8285714285714286, 'recall': 0.9354838709677419, 'f1': 0.8787878787878788, 'number': 31}
• eval_enu.sub.price: {'precision': 1.0, 'recall': 0.95, 'f1': 0.9743589743589743, 'number': 20}
• eval_enu.unitprice: {'precision': 0.984375, 'recall': 0.9402985074626866, 'f1': 0.9618320610687023, 'number': 67}
• eval_otal.cashprice: {'precision': 0.9558823529411765, 'recall': 0.9558823529411765, 'f1': 0.9558823529411765, 'number': 68}
• eval_otal.changeprice: {'precision': 0.9655172413793104, 'recall': 1.0, 'f1': 0.9824561403508771, 'number': 56}
• eval_otal.creditcardprice: {'precision': 0.7647058823529411, 'recall': 0.8125, 'f1': 0.787878787878788, 'number': 16}
• eval_otal.emoneyprice: {'precision': 0.3333333333333333, 'recall': 0.5, 'f1': 0.4, 'number': 2}
• eval_otal.menuqty_cnt: {'precision': 0.9333333333333333, 'recall': 0.9655172413793104, 'f1': 0.9491525423728815, 'number': 29}
• eval_otal.menutype_cnt: {'precision': 1.0, 'recall': 0.7142857142857143, 'f1': 0.8333333333333333, 'number': 7}
• eval_otal.total_etc: {'precision': 0.5, 'recall': 0.3333333333333333, 'f1': 0.4, 'number': 3}
• eval_otal.total_price: {'precision': 0.9583333333333334, 'recall': 0.968421052631579, 'f1': 0.9633507853403142, 'number': 95}
• eval_ub_total.discount_price: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 7}
• eval_ub_total.etc: {'precision': 0.875, 'recall': 0.7777777777777778, 'f1': 0.823529411764706, 'number': 9}
• eval_ub_total.service_price: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 12}
• eval_ub_total.subtotal_price: {'precision': 0.9545454545454546, 'recall': 0.9692307692307692, 'f1': 0.9618320610687022, 'number': 65}
• eval_ub_total.tax_price: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 43}
• eval_overall_precision: 0.9522
• eval_overall_recall: 0.9544
• eval_overall_f1: 0.9533
• eval_overall_accuracy: 0.9707
• eval_runtime: 3.0438
• eval_samples_per_second: 32.853
• eval_steps_per_second: 4.271
• epoch: 1.0
• step: 50

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: 1e-06
• 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
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.36.0
• Pytorch 2.0.0
• Datasets 2.16.1
• Tokenizers 0.15.0