metadata

license: mit
base_model: kavg/LiLT-RE-ZH
tags:
  - generated_from_trainer
datasets:
  - xfun
metrics:
  - precision
  - recall
  - f1
model-index:
  - name: checkpoints
    results: []

checkpoints

This model is a fine-tuned version of kavg/LiLT-RE-ZH on the xfun dataset. It achieves the following results on the evaluation set:

Precision: 0.4183
Recall: 0.5884
F1: 0.4890
Loss: 0.4704

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-05
train_batch_size: 8
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
training_steps: 10000

Training results

Training Loss	Epoch	Step	F1	Validation Loss	Precision	Recall
0.0761	41.67	500	0.3965	0.1487	0.3487	0.4596
0.0508	83.33	1000	0.4620	0.2456	0.4049	0.5379
0.0422	125.0	1500	0.4876	0.2545	0.4137	0.5934
0.0186	166.67	2000	0.4984	0.2960	0.4258	0.6010
0.0147	208.33	2500	0.4933	0.3388	0.4171	0.6035
0.0054	250.0	3000	0.5010	0.3819	0.4283	0.6035
0.0059	291.67	3500	0.5115	0.4177	0.4373	0.6162
0.006	333.33	4000	0.4974	0.3875	0.4281	0.5934
0.0026	375.0	4500	0.4910	0.4209	0.4226	0.5859
0.0078	416.67	5000	0.4952	0.3861	0.4275	0.5884
0.0035	458.33	5500	0.4890	0.4193	0.4183	0.5884
0.0022	500.0	6000	0.5059	0.4399	0.4395	0.5960
0.0042	541.67	6500	0.4979	0.4653	0.4288	0.5934
0.0079	583.33	7000	0.5037	0.4514	0.4309	0.6061
0.0047	625.0	7500	0.4937	0.4701	0.4227	0.5934
0.0032	666.67	8000	0.4937	0.4733	0.4227	0.5934
0.0048	708.33	8500	0.4339	0.6136	0.5084	0.5080
0.0004	750.0	9000	0.4183	0.5884	0.4890	0.4704
0.0044	791.67	9500	0.4213	0.5884	0.4910	0.4843
0.0014	833.33	10000	0.4234	0.5934	0.4942	0.5084

Framework versions

Transformers 4.38.2
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.1